HP Switch Software IPv6 Configuration Guide - HPE Support Center

Loading...

HP Switch Software IPv6 Configuration Guide K/KA.15.14 nl

Abstract This switch software guide is intended for network administrators and support personnel, and applies to the switch models listed on this page unless otherwise noted. This guide does not provide information about upgrading or replacing switch hardware. The information in this guide is subject to change without notice. Applicable Products HP HP HP HP HP

Switch Switch Switch Switch Switch

3500 Series 3500yl Series 3800 Series 5406zl Series 5412zl Series

HP Part Number: 5998-4901 Published: October 2013 Edition: 3

HP HP HP HP

Switch Switch Switch Switch

6200yl-24G (J8992A) 8206zl (J9475A) 8212zl (J8715A/B) 6600 Series

© Copyright 2008, 2013 Hewlett-Packard Development Company, L.P. Confidential computer software. Valid license from HP required for possession, use or copying. Consistent with FAR 12.211 and 12.212, Commercial Computer Software, Computer Software Documentation, and Technical Data for Commercial Items are licensed to the U.S. Government under vendor's standard commercial license. The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein. UNIX is a registered trademark of The Open Group. Acknowledgments Microsoft, Windows, Windows XP, and Windows NT are U.S. registered trademarks of Microsoft Corporation. Java and Oracle are registered trademarks of Oracle and/or its affiliates. Warranty For the software end user license agreement and the hardware limited warranty information for HP Networking products, visit www.hp.com/ networking/support.

Contents 1 IPv6 Addressing Configuration...................................................................11 Introduction............................................................................................................................13 General configuration steps.....................................................................................................13 Configuring IPv6 addressing....................................................................................................14 Configuring IP source-interface addressing.................................................................................14 Enabling IPv6 with an automatically configured link-local address.................................................15 Viewing currently configured IPv6 unicast addresses...............................................................16 Enabling autoconfiguration of a global unicast address and a default router identity on a VLAN.......16 Viewing current IPv6 autoconfiguration settings......................................................................16 Default IPv6 Gateway....................................................................................................17 Enabling DHCPv6..................................................................................................................17 Authenticating the DHCPv6 client.............................................................................................18 Viewing DHCP client information..............................................................................................18 DHCPv6 client authentication operating notes........................................................................19 Viewing configured DHCPv6 addresses................................................................................20 General operating notes for DHCPv6...................................................................................20 Statically configuring a link-local unicast address........................................................................20 Statically configuring a global unicast address...........................................................................21 Viewing the currently configured static IPv6 addresses per-VLAN..............................................22 Operating notes................................................................................................................22 Assigning an IPv6 address to a loopback interface.....................................................................22 Displaying loopback interface configurations.........................................................................23 Viewing the current IPv6 addressing configuration.......................................................................23 View IPv6 gateway, route, and router neighbors.........................................................................28 Viewing gateway and IPv6 route information.........................................................................28 Viewing IPv6 router information...........................................................................................29 Address lifetimes....................................................................................................................30 Preferred lifetime................................................................................................................30 Valid lifetime.....................................................................................................................30 DHCPv6 client........................................................................................................................31 Duplicate address detection (DAD) for statically configured addresses............................................31 IPv6 loopback interfaces..........................................................................................................31 About disabling IPv6 on a VLAN..............................................................................................32 Neighbor discovery................................................................................................................32 Duplicate address detection (DAD).......................................................................................33 DAD operation.............................................................................................................33 Configuring DAD....................................................................................................................33 Operating notes for ND......................................................................................................34 Router access and default router selection..................................................................................35 Router advertisements.........................................................................................................35 IPv6 router advertisement options for DNS configuration.....................................................35 Global configuration.....................................................................................................35 IP interface configuration...........................................................................................36 Displaying the configuration.......................................................................................36 Router solicitations.............................................................................................................36 Default IPv6 router.............................................................................................................37 Router redirection...............................................................................................................37

2 IPv6 Management Features........................................................................38 Viewing the neighbor cache.....................................................................................................38 Clearing the Neighbor Cache.............................................................................................40 IPv6 Telnet operation...............................................................................................................41 Contents

3

Using outbound Telnet to another device...............................................................................41 Viewing the current telnet activity on a switch.........................................................................42 Enabling or disabling inbound Telnet access.........................................................................43 Viewing the current inbound Telnet configuration....................................................................43 Configuring an IPv6 address for an SNTP server.........................................................................44 Configuring (enabling or disabling) the Timep mode..............................................................46 Transferring TFTP files over IPv6................................................................................................48 Enabling TFTP for IPv6........................................................................................................48 Copying files over IPv6 using TFTP.......................................................................................49 Using auto-TFTP for IPv6.....................................................................................................51 SNMP Configuration Commands Supported...............................................................................51 SNMPv1 and v2c..............................................................................................................52 SNMPv3...........................................................................................................................52 IP preserve for IPv6.................................................................................................................54 Configuring IP preserve......................................................................................................54 Downloading an IP preserve configuration file to an IPv6-based switch.....................................54 Verifying how IP preserve was implemented in a switch...........................................................55 View the neighbor cache.........................................................................................................55 Clear the neighbor cache...................................................................................................56 SNTP and Timep....................................................................................................................56 About configuring (enabling or disabling) the SNTP mode......................................................56 About configuring an IPv6 address for an SNTP server............................................................56 TFTP file transfers over IPv6......................................................................................................57 SNMP management for IPv6....................................................................................................57 SNMP features supported...................................................................................................57

3 IPv6 Management Security Features............................................................58 Configuring authorized IP managers for switch access.................................................................58 Configuring single station access.........................................................................................58 Configuring multiple station access.......................................................................................59 Viewing an authorized IP managers configuration..................................................................59 Authorizing manager access...............................................................................................60 Editing an existing authorized IP manager entry.....................................................................61 Deleting an authorized IP manager entry..............................................................................61 Configuring SSH for IPv6.........................................................................................................62 Displaying an SSH configuration..........................................................................................64 Performing secure file transfers to and from IPv4 and IPv6 client devices.........................................64 Authorized IP managers for IPv6...............................................................................................65 About using a mask to configure authorized management stations................................................66 About configuring multiple station access..............................................................................66 Secure shell (SSH) for IPv6.......................................................................................................69 SCP and SFTP for IPv6............................................................................................................70

4 Multicast Listener Discovery (MLD) Snooping................................................71 Overview..............................................................................................................................72 Enabling or Disabling MLD Snooping on a VLAN.......................................................................72 Setting the MLD Version...........................................................................................................73 Configuring per-port MLD traffic filters.......................................................................................73 Configuring the querier...........................................................................................................74 Configuring the Query Interval.................................................................................................75 Configuring the Query Maximum Response Time........................................................................75 Configuring the Number of Times to Retry a Query.....................................................................75 Configuring the Last Member Query Interval..............................................................................76 Configuring Fast Learn.............................................................................................................76 Configuring fastleave..............................................................................................................76 Configuring forced fastleave.....................................................................................................77 4

Contents

Viewing the current MLD status.................................................................................................77 Configuring the current MLD.....................................................................................................78 Listing ports currently joined.....................................................................................................79 Viewing MLD statistics.............................................................................................................80 Viewing MLD counters.............................................................................................................82 MLD snooping........................................................................................................................83 General operation.............................................................................................................84 Forwarding in MLD snooping..............................................................................................85 Listeners and joins..............................................................................................................86 Queries............................................................................................................................86 Leaves..............................................................................................................................86 Fast leaves and forced fast leaves.........................................................................................87 Current MLD status..................................................................................................................87 Current MLD configuration.......................................................................................................88 Counters................................................................................................................................90

5 IPv6 Access Control Lists (ACLs)..................................................................91 Introduction............................................................................................................................93 Command Summary for Configuring ACLs............................................................................94 Command Summary for Enabling, Disabling, and Displaying ACLs..........................................95 Overview..............................................................................................................................99 Types of IPv6 ACLs.............................................................................................................99 Concurrent IPv4 and IPv6 ACLs............................................................................................99 IPv6 ACL applications........................................................................................................99 RACL applications.........................................................................................................99 VACL applications.......................................................................................................100 IPv6 static port ACL applications...................................................................................101 RADIUS-assigned (dynamic) port ACL applications..........................................................101 Effect of RADIUS-assigned ACLs when multiple clients are using the same port................101 802.1X user-based and port-based applications..........................................................102 Operating notes for IPv6 applications............................................................................102 Multiple ACL assignments on an interface...........................................................................103 About filtering inbound traffic with multiple ACLS.............................................................103 Filtering outbound traffic...............................................................................................104 Permitting traffic filtered through multiple ACLs................................................................104 Features common to all ACL applications............................................................................105 General steps for planning and configuring ACLs.....................................................................106 IPv6 ACL operation...............................................................................................................107 Packet-filtering process......................................................................................................108 Planning an ACL application..................................................................................................109 IPv6 traffic management and improved network performance................................................110 Security..........................................................................................................................111 Guidelines for planning the structure of an ACL....................................................................111 ACL configuration and operating rules................................................................................112 How an ACE uses a prefix to screen packets for SA and DA matches......................................113 Prefix usage differences between ACLs and other IPv6 addressing.....................................115 Configuring and assigning an IPv6 ACL..................................................................................115 Implementing IPv6 ACLs....................................................................................................115 Permit/deny options.........................................................................................................116 Overriding an implicit deny...............................................................................................116 ACL configuration............................................................................................................117 ACL Configuration Structure..........................................................................................117 ACL configuration factors..................................................................................................119 The sequence of entries in an ACL is significant...............................................................119 Implied deny function..................................................................................................120 Contents

5

Assignment of an ACL to an interface............................................................................120 Assignment of an ACL name to an interface....................................................................120 Creating an ACL using the CLI................................................................................................120 General ACE rules................................................................................................................120 Adding or inserting an ACE in an ACL...............................................................................120 Deleting an ACE..............................................................................................................121 Duplicate ACE sequence numbers.................................................................................121 Configuration Commands......................................................................................................121 Creating an ACL and/or entering the IPv6 ACL (ipv6-acl) context...........................................121 Configuring ACEs in an ACL.............................................................................................122 Configuring TCP and UDP traffic in IPv6 ACLs......................................................................125 Filtering ICMP traffic.........................................................................................................127 Filtering routed IPv6 traffic.................................................................................................128 Filtering routed or switched IPv6 traffic inbound on a VLAN...................................................130 Filtering inbound IPv6 traffic per port and trunk...............................................................131 Deleting an ACL..............................................................................................................132 Inserting an ACE in an existing ACL...................................................................................132 Deleting an ACE from an existing ACL.....................................................................................134 Resequencing the ACEs in an IPv6 ACL...................................................................................135 Attaching a remark to an ACE................................................................................................136 Appending remarks and related ACEs to the end of an ACL.......................................................137 Inserting remarks and related ACEs within an existing list...........................................................137 Inserting a remark for an ACE that already exists in an ACL.......................................................138 Replacing an existing remark.................................................................................................138 Removing a remark from an existing ACE................................................................................138 Operating notes for remarks..............................................................................................138 Viewing ACL configuration data.............................................................................................139 Viewing an ACL summary......................................................................................................140 Viewing the content of all ACLs on the switch...........................................................................141 Viewing the IPv4 and IPv6 VACL assignments for a VLAN..........................................................142 Viewing the IPv4 and IPv6 RACL and VACL assignments for a VLAN...........................................143 Viewing static port (and trunk) ACL assignments........................................................................144 Viewing the content of a specific ACL......................................................................................145 Creating or editing an ACL offline..........................................................................................150 The offline process...........................................................................................................150 Example of using the offline process...................................................................................151 Enabling ACL logging on the switch........................................................................................152 Monitoring static ACL performance.........................................................................................156 ACE counter operation.....................................................................................................157 Resetting ACE hit counters to zero......................................................................................157 Example of ACL performance monitoring............................................................................157 Example of resetting ACE hit counters to zero......................................................................158 Options for applying IPv6 ACLs on the switch...........................................................................159 Static ACLS.....................................................................................................................159 RADIUS-assigned ACLs.....................................................................................................160 Using CIDR notation to enter the IPv6 ACL prefix length.............................................................160 Overview of IPv6 ACLs..........................................................................................................160 Commands to create, enter, and configure an ACL...............................................................161 Example: IPv6 ACL configuration in a routed environment..........................................................161 Editing an existing ACL.........................................................................................................162 General editing rules........................................................................................................162 Sequence numbering in ACLs............................................................................................163 About displaying All ACLs and their assignments in the switch startup-config file and running-config file.................................................................................................................................164 Testing and troubleshooting ACLs...........................................................................................164 6

Contents

Enable IPv6 ACL "Deny" or “Permit” logging.......................................................................164 Requirements for using IPv6 ACL logging........................................................................164 ACL logging operation.................................................................................................165 IPv6 counter operation with multiple interface assignments.........................................................165 IPv4 counter operation with multiple interface assignments....................................................167 General ACL operating notes.................................................................................................170 Unable to Delete an ACL in the Running Configuration..........................................................172

6 IPv6 Routing Basics.................................................................................173 IPv6 routing overview ...........................................................................................................173 Dual stack IPv4/IPv6 operation..........................................................................................174 IPv6 Routing Features ......................................................................................................174 Viewing the router ID............................................................................................................175 Manually configuring a router ID............................................................................................175 Changing an existing router ID...............................................................................................175 Viewing a manually configured router ID.................................................................................176 Configuring the IPv6 hop limit................................................................................................176 Configuring the IPv6 default route...........................................................................................177 Viewing the IPv6 routing table................................................................................................177 Enabling IPv6 routing............................................................................................................178 Configuring a router ID..........................................................................................................179 IPv6 networks and subnets.....................................................................................................179 VLANs and routing..........................................................................................................180 Link-local.........................................................................................................................180 Global unicast.................................................................................................................180 IPv6 management interface...............................................................................................180 IPv6 routing operation...........................................................................................................181 Concurrent static and dynamic routing operation..................................................................181 Router Advertisements (RAs)..............................................................................................181 DHCPv6-relay.................................................................................................................182 General Steps for Enabling IPv6 Routing.............................................................................182 Configuring global IPv6 routing parameters.............................................................................183 System router ID...............................................................................................................183 Automatic router ID selection........................................................................................183 Different route types in the IPv6 routing table............................................................................183 Routing table content........................................................................................................184 Destination network..........................................................................................................184 Gateway for forwarding routed traffic.................................................................................185 Metric and administrative distance.....................................................................................185

7 IPv6 Static Routing..................................................................................187 Adding static and null routes to IPv6 table...............................................................................187 Viewing static route information..............................................................................................188 About static routing ..............................................................................................................189 Advantages.....................................................................................................................190 Disadvantages.................................................................................................................190 Static route types.............................................................................................................190 Static routing default settings.............................................................................................191 Static route states follow VLAN states..................................................................................191 Static routes for ECMP applications....................................................................................191

8 IPv6 Router Advertisements......................................................................193 Global configuration context commands..................................................................................194 Enabling or disabling IPv6 Router Advertisement generation..................................................194 Enabling or disabling IPv6 routing.....................................................................................194 VLAN or tunnel context ND configuration................................................................................194 Contents

7

Configuring DHCPv6 service requirements..........................................................................194 Configuring the range for intervals between RA transmissions on a VLAN...............................195 Setting or changing the hop-limit for host-generated packets.............................................196 Setting or changing the default router lifetime.................................................................196 Changing the reachable time duration for neighbors............................................................196 Setting or changing the Neighbor Discovery retransmit timer.................................................197 Configuring the global unicast prefix and lifetime for hosts on a VLAN....................................197 Suppressing Router Advertisements on a VLAN....................................................................202 Restricting IPv6 Router Advertisements.................................................................................203 Configuring RA Guard.................................................................................................203 Operating Notes.........................................................................................................203 Displaying the Router Advertisement configuration.....................................................................204 General RA operation...........................................................................................................207 RA basics.......................................................................................................................208 About setting up your IPv6 RA policy..................................................................................208 Steps for configuring IPv6 RAs................................................................................................208 Guidelines for configuring RAs on multiple routing switches for the same VLAN............................209

9 DHCPv6-Relay.......................................................................................211 Configuring DHCPv6-relay.....................................................................................................211 Viewing the DHCPv6-relay configuration.............................................................................212 DHCPv6-relay operating notes...........................................................................................214 About configuring DHCPv6 relay............................................................................................214 DHCPv6 request forwarding..............................................................................................215 DHCPv6-relay helper addresses.........................................................................................215 General steps for enabling DHCP relay operation................................................................215 Multiple-hop forwarding of DHCPv6 service requests.................................................................216

10 OSPFv3 Routing...................................................................................218 OSPFv3 RFC compliance.......................................................................................................220 Activating OSPFv3................................................................................................................220 Configuring OSPFv3 on the routing switch...............................................................................220 Router ID or IPv4 loopback address requirement..................................................................220 Enabling IPv6 Routing......................................................................................................220 Enabling global OSPFv3 routing........................................................................................221 Assigning the routing switch to OSPFv3 areas......................................................................221 Configuring an OSPFv3 backbone or normal area..........................................................222 Configuring a stub or NSSA area..................................................................................222 Enabling OSPFv3 on an interface and assigning one or more VLANs to each area..................224 Assigning IPv6 loopback addresses to an area (optional)......................................................225 OSPFv3 redistribution of loopback addresses..................................................................225 Configuring route-maps.........................................................................................................226 Enabling route redistribution...................................................................................................226 Modifying the default metric for redistribution...........................................................................227 Modifying the redistribution metric type...................................................................................227 Enabling redistribution of loopback IPv6 addresses in OSPFv3 when the addresses are not assigned to an OSPFv3 area..........................................................................................................228 Verifying the OSPFv3 redistribution of loopback interfaces....................................................228 Configuring ranges on an ABR to reduce advertising to the backbone.........................................229 Influencing route choices by changing the administrative distance default.....................................231 Enforcing strict LSA operation for graceful restart helper mode....................................................232 Adjusting performance by changing the VLAN interface settings.................................................232 Cost per interface............................................................................................................232 Dead interval per interface................................................................................................232 Hello interval per interface................................................................................................233 Priority per interface.........................................................................................................233 8

Contents

Retransmit interval per interface.........................................................................................233 Transit delay per interface.................................................................................................233 Configuring a virtual link.......................................................................................................233 Adjusting a dead interval on a virtual link...........................................................................234 Adjusting a hello interval on a virtual link............................................................................235 Adjusting the retransmit interval on a virtual link...................................................................236 Adjusting transit-delay on a virtual link................................................................................236 Configuring OSPFv3 passive.............................................................................................237 Troubleshooting: Logging neighbor adjacency change events................................................238 Displaying OSPv3F Information..............................................................................................239 Viewing a summary of OSPFv3 configuration information..........................................................240 Viewing general OSPFv3 configuration information..............................................................241 Viewing OSPFv3 route information.....................................................................................242 Viewing OSPFv3 area information......................................................................................243 Viewing OSPFv3 interface information................................................................................243 Viewing OSPFv3 interface information for neighbor routers...................................................246 Viewing or clearing OSPFv3 packet statistics counters...........................................................247 Viewing OSPFv3 link-state AS-scope information..................................................................248 Viewing OSPFv3 link-state Area-Scope information...............................................................250 Viewing OSPFv3 link-state link-scope information..................................................................252 Viewing OSPFv3 redistribution information..........................................................................253 Viewing OSPFv3 virtual link information..............................................................................254 Viewing OSPFv3 virtual neighbor information......................................................................254 Viewing OSPFv3 SPF statistics...........................................................................................255 Debugging OSFP routing messages........................................................................................256 Graceful Shutdown of OSPF Routing.......................................................................................256 Modules Operating in NonStop Mode...............................................................................256 Enabling load-sharing among next-hop routes..........................................................................256 Displaying the current IP load-sharing configuration...................................................................258 License requirements.............................................................................................................258 Overview of OSPFv3.............................................................................................................259 Hello packets..................................................................................................................259 Link-state advertisements (LSAs)..........................................................................................259 OSPFv3 router types.............................................................................................................260 Interior routers.................................................................................................................260 Area border routers (ABRs)................................................................................................260 Autonomous system boundary router (ASBR)........................................................................261 Designated routers......................................................................................................261 OSPFv3 area types...............................................................................................................263 Normal area...................................................................................................................263 Backbone area................................................................................................................264 Stub area........................................................................................................................264 Not-so-stubby-area (NSSA)................................................................................................264 Reducing AS-external-LSAs and inter-area-prefix-LSAs.................................................................265 Algorithm for AS-external-LSA reduction..............................................................................265 About replacing inter-area-prefix-LSAs and type-7-external-LSA default routes with an AS-external-LSA default route....................................................................................................................265 About equal-cost multi-path routing.........................................................................................266 OSPFv3 Activation and Dynamic Configuration........................................................................268 General configuration steps for OSPFv3..............................................................................268 Configuration rules...........................................................................................................268 OSPFv3 global and interface settings......................................................................................268 OSPFv3 redistribution of loopback addresses...........................................................................269 About configuring for external route redistribution in an OSPFv3 domain (optional).......................269 About configuring ranges on an ABR to reduce advertising to the backbone................................270 Contents

9

About influencing route choices by changing the administrative distance default (optional).............270 About enforcing strict LSA operation for graceful restart helper mode (optional).............................270 About adjusting performance by changing the VLAN interface settings (optional)..........................271 About configuring an ABR to use a virtual link to the backbone..................................................271 About adjusting virtual link performance by changing the interface settings.............................272 OSPFv3 passive...................................................................................................................272

11 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels.....................273 Overview............................................................................................................................273 Configuring a Tunnel Interface................................................................................................274 Configuring the Tunnel Mode............................................................................................275 Configuring the Tunnel Source...........................................................................................275 Configuring the Tunnel Destination.....................................................................................276 Configuring the Static MTU...............................................................................................276 Configuring a Value for TOS.............................................................................................276 Configuring a Value for Time-to-Live (TTL)............................................................................277 Example: Manual 6in4 Tunneling...........................................................................................277 Configure Switch B...........................................................................................................278 Configure Switch C..........................................................................................................279 Example: Tunneling Using Policy-Based Routing (PBR)................................................................280 Configure Switch B...........................................................................................................280 Configure Switch C..........................................................................................................281 Displaying Tunnel Configuration and Status Information.............................................................282 Show Interface Tunnel Brief...............................................................................................283 Show IPv6 ND RA Prefix for a Specific Tunnel......................................................................283 Show IP Counters for a Tunnel...........................................................................................285

12 IPv6 Diagnostic and Troubleshooting.......................................................288 Introduction..........................................................................................................................288 ICMP Rate-Limiting................................................................................................................289 Ping for IPv6 (Ping6).............................................................................................................290 Traceroute for IPv6................................................................................................................291 DNS Resolver for IPv6...........................................................................................................293 DNS Configuration..........................................................................................................293 Viewing the Current Configuration......................................................................................295 Operating Notes.............................................................................................................295 Debug/Syslog for IPv6..........................................................................................................295 Configuring Debug and Event Log Messaging.....................................................................295 Debug Command............................................................................................................296 Configuring Debug Destinations.........................................................................................297 Configuring an IPv6 Syslog Server.....................................................................................298 Logging Command..........................................................................................................298 Displaying a Debug/Syslog for Configuration......................................................................299

IPv6 Glossary...........................................................................................300 Index.......................................................................................................302

10

Contents

1 IPv6 Addressing Configuration Table 1 Summary of commands Command syntax

Description

Default

CLI page reference

[no] ipv6 enable

Enables IPv6 on VLAN and automatically configures the VLAN's link-local unicast address with a 64-bit EUI-64 interface identifier generated from the VLAN MAC address.

Disabled

15

[no] ipv6 address autoconfig

Implements unicast address autoconfiguration.

Disabled

16

ipv6 source-interface [protocol-ID | all][loopback | vlan | ] [no]ipv6 source-interface[protocol-ID | all]

Defines the source IP address selection policy for the application protocols.

[no] ipv6 address dhcp full rapid-commit

Configures DHCPv6 on a VLAN, which initiates transmission of DHCPv6 requests for service.

Disabled

17

[no] ipv6 dhcp-client authentication [ mode [ md5 ] | key-chain chain-name-str ]

Allows the configuration of authentication options.

-

18

show ipv6 dhcp-client [ authentication [ vlan vid ] | statistics [ vlan vid ] ]

Displays information about IPv6 DHCP client authentication parameters or DHCP client statistics.

-

18

show ipv6

Lists IPv6 addresses for all VLANs configured on the switch.

-

20

show ipv6 vlan vid

Lists IPv6 addresses configured on the VLAN.

-

20

[no] ipv6 address fe80::interface-id link-local

Configures a link-local unicast address.

-

20

[no] ipv6 address [ network-prefix interface-id/prefix-length ] [no] ipv6 address [ network-prefix ::/prefix-length ] eui-64

Statically configures a global unicast address.

-

21

show run

Displays the currently configured static IPv6 addresses per-VLAN.

-

22

11

Table 1 Summary of commands (continued) Command syntax

Description

Default

CLI page reference

[no] interface loopback 0 - 7 ipv6 address ipv6-addr

Configures an IPv6 address on a loopback interface.

-

22

show ipv6 vlan [ vlan vid | tunnel tunnel-id ]

Displays: IP and IPv6 global configuration settings.

-

23

IPv6 status for the specified VLAN. IPv6 addresses (with prefix lengths) configured on the specified VLAN. Expiration data (Expiry) for each address.

12

show ipv6 nd

Displays the current IPv6 ND settings on the configured VLAN interfaces.

-

25

show ipv6 route ipv6-addr connected

Displays the routes in the switch's IPv6 routing table.

-

28

show ipv6 routers vlan vid

Lists the switch's IPv6 router table entries for all VLANs configured on the switch or for a single VLAN.

-

29

ipv6 nd dad-attempts 0 - 255

Executed at the global 3 (enabled) 33 config level, configures the number of neighbor solicitations to send when performing DAD for a unicast address configured on a VLAN interface.

ipv6 nd ns-interval milliseconds

Used on VLAN interfaces to reconfigure the neighbor discovery time in milliseconds between DAD neighbor solicitations sent for an unresolved destination, or between duplicate address detection neighbor solicitation requests.

1000 ms

33

ipv6 nd reachable-time milliseconds

Used on VLAN interfaces to configure the length of time in milliseconds a neighbor will be considered reachable

30000 ms

34

IPv6 Addressing Configuration

Table 1 Summary of commands (continued) Command syntax

Description

Default

CLI page reference

after the Neighbor Unreachability Detection algorithm has confirmed it to be reachable.

More information, see “Viewing the current IPv6 addressing configuration” (page 23), (page 25).

Introduction In the default configuration, IPv6 operation is disabled on the switch. This section describes the general steps and individual commands for enabling IPv6 operation. This chapter provides the following: •

general steps for IPv6 configuration



IPv6 command syntax descriptions, including show commands

Most IPv6 configuration commands are applied per-VLAN. The exceptions are ICMP, ND, and the (optional) authorized-managers feature, which are configured at the global configuration level. (ICMP and ND for IPv6 are enabled with default values when IPv6 is first enabled and can either be left in their default settings or reconfigured, as needed.) NOTE: Beginning with software release K.13.01, the switch is capable of operating in dual-stack mode, where IPv4 and IPv6 run concurrently on a given VLAN.

General configuration steps The IPv6 configuration on switches running software release K.13.01 or greater includes global and per-VLAN settings. This section provides an overview of the general configuration steps for enabling IPv6 on a given VLAN and can be enabled by any one of several commands. The following steps provide a suggested progression for getting started. NOTE: The ICMP and ND parameters are set to default values at the global configuration level, are satisfactory for many applications, and generally do not need adjustment when you are first configuring IPv6 on the switch. In the default configuration, IPv6 is disabled on all VLANs or tunnels. 1.

If IPv6 DHCP service is available, enable IPv6 DHCP on the VLAN. If IPv6 is not already enabled on the VLAN, enabling DHCPv6 also enables IPv6 and automatically configures a link-local address using the EUI-64 format. NOTE: If IPv6 is not already enabled on the VLAN, enabling DHCPv6 causes the switch to automatically generate a link-local address. DHCPv6 does not assign a link-local address. A DHCPv6 server can provide other services, such as the addresses of time servers. For this reason, you may want to enable DHCP even if you are using another method to configure IPv6 addressing on the VLAN.

Introduction

13

2.

If IPv6 DHCP service is not enabled on the VLAN, do either of the following: •

Enable IPv6 on the VLAN. This automatically configures a link-local address with an EUI-64 interface identifier.



Statically configure a unicast IPv6 address on the VLAN. This enables IPv6 on the VLAN and, if you configure anything other than a link-local address, the link-local address is automatically configured with an EUI-64 interface identifier.

3.

4.

If an IPv6 router is connected on the VLAN, enable IPv6 address autoconfiguration to automatically configure global unicast addresses with prefixes included in advertisements received from the router. The interface identifier used in addresses configured by this method is the same as the interface identifier in the current link-local address. If needed, statically configure IPv6 unicast addressing on the VLAN interface as needed. This can include statically replacing the automatically generated link-local address.

Configuring IPv6 addressing In the default configuration on a VLAN, any one of the following commands enables IPv6 and creates a link-local address. Thus, while any one of these methods is configured on a VLAN, IPv6 remains enabled and a link-local address is present. ipv6 enable See “Enabling IPv6 with an automatically configured link-local address” (page 15). ipv6 address autoconfig See “Enabling autoconfiguration of a global unicast address and a default router identity on a VLAN” (page 16). ipv6 address dhcp full [ rapid-commit ] See “Enabling DHCPv6” (page 17). ipv6 address fe80:0:0:0: See interface-identifier link-local “Statically configuring a link-local unicast address” (page 20). ipv6 address See prefix:interface-identifier “Statically configuring a global unicast address” (page 21). NOTE: Addresses created by any of these methods remain tentative until verified as unique by DAD. See “Duplicate address detection (DAD)” (page 33).

Configuring IP source-interface addressing Syntax ipv6 source-interface [protocol-ID | all][loopback | vlan | ] [no]ipv6 source-interface[protocol-ID | all] Defines the source IP address selection policy for the application protocols.

14

IPv6 Addressing Configuration

The use of the [no] form of the command reverts the application protocols to a default behavior when the IP address of the outgoing interface is used as the source. protocol-ID

The name of application protocol to apply the source IP address selection policy to.

loopback

The loopback interface to use as the source IP address provider for the application protocol. If multiple IP addresses are configured on the interface the lexicographically lowest IP address will be used.

vlan

The VLAN interface to use as the source IP address provider for the application protocol. If multiple IP addresses are configured on the interface the lexicographically lowest IP address will be used.



The specific IPv6 address to use as the source in the protocol generated outgoing IP packets.

Example HP-Switch# show ipv6 source-interface detail Protocol : Radius Admin Policy : Outgoing Interface Oper Policy : Outgoing Interface Source IPv6 Interface : vlan- 1 Source IPv6 Address : 192.168.1.2 Source Interface State : N/A

Shows the source IPv6 configuration, status or detailed information. In this example the command is invoked without parameters therefore it shows the configuration information for all protocols. •

If status is specified the operational status information will be shown.



If detail is specified the detailed operational status information will be shown.



If protocol-ID is specified the information only relevant to the specified protocol will be shown.

Enabling IPv6 with an automatically configured link-local address Syntax: [no] ipv6 enable If IPv6 has not already been enabled on a VLAN by another IPv6 command option described in this chapter, this command enables IPv6 on the VLAN and automatically configures the VLAN's link-local unicast address with a 64-bit EUI-64 interface identifier generated from the VLAN MAC address. NOTE: Only one link-local IPv6 address is allowed on the VLAN interface. Subsequent static or DHCP configuration of another link-local address overwrites the existing link-local address. A link-local address always uses the prefix fe80:0:0:0. With IPv6 enabled, the VLAN uses received RAs to designate the default IPv6 router. See “Default IPv6 router” (page 37). After verification of uniqueness by DAD, a link-local IPv6 address assigned automatically is set to the preferred status, with a "permanent" lifetime. Default: Disabled The no form of the command disables IPv6 on the VLAN if no other IPv6-enabling command is configured on the VLAN. See “About disabling IPv6 on a VLAN” (page 32). Enabling IPv6 with an automatically configured link-local address

15

Viewing currently configured IPv6 unicast addresses To view the current IPv6 enable setting and any statically configured IPv6 addresses per-VLAN, use show run. To view all currently configured IPv6 unicast addresses, use the following: •

show ipv6 (Lists IPv6 addresses for all VLANs configured on the switch.)



show ipv6 vlan vid (Lists IPv6 addresses configured on the VLAN.)



show ipv6 tunnel tunnel-id (Lists IPv6 addresses configured on the tunnel.)

For more information, see “Viewing the current IPv6 addressing configuration” (page 23).

Enabling autoconfiguration of a global unicast address and a default router identity on a VLAN Enabling autoconfig or rebooting the switch with autoconfig enabled on a VLAN causes the switch to configure IPv6 addressing on the VLAN using RAs and an EUI-64 interface identifier.

Syntax: [no] ipv6 address autoconfig Implements unicast address autoconfiguration as follows: •

If IPv6 is not already enabled on the VLAN, enables IPv6 and generates a link-local ( EUI-64) address.



Generates router solicitations (RSs) on the VLAN.



If an RA is received on the VLAN, the switch uses the route prefix in the RA to configure a global unicast address. The interface identifier for this address is the same as the interface identifier used in the current link-local address at the time the RA is received. This can be either a statically configured or the (automatic) EUI-64 interface identifier, depending on how the link-local address was configured. If an RA is not received on the VLAN after autoconfig is enabled, a link-local address is present, but no global unicast addresses are autoconfigured. NOTE: If a link-local address is already configured on the VLAN, a later, autoconfigured global unicast address, uses the same interface identifier as the link-local address. Autoconfigured and DHCPv6-assigned global unicast addresses with the same prefix are mutually exclusive on a VLAN. On a given switch, if both options are configured on the same VLAN, only the first to acquire a global unicast address is used.

After verification of uniqueness by DAD, an IPv6 address assigned to a VLAN by autoconfiguration is set to the preferred and valid lifetimes specified by the RA used to generate the address and is configured as a preferred address. Default: Disabled. The no form of the command produces different results, depending on how IPv6 is configured on the VLAN: If IPv6 was enabled only by the autoconfig command, deleting this command disables IPv6 on the VLAN. See “About disabling IPv6 on a VLAN” (page 32).

Viewing current IPv6 autoconfiguration settings To view the current IPv6 autoconfiguration settings per-VLAN or per-tunnel, use show run.

16

IPv6 Addressing Configuration

To view all currently configured IPv6 unicast addresses, use the following: •

show ipv6 Lists IPv6 addresses for all VLANs configured on the switch.



show ipv6 vlan vid Lists IPv6 addresses configured on the VLAN.



show ipv6 tunnel tunnel Lists IPv6 addresses configured on the tunnel.

For more information, see “Viewing the current IPv6 addressing configuration” (page 23).

Default IPv6 Gateway Instead of using static or DHCPv6 configuration, a default IPv6 gateway for an interface (VLAN or tunnel) is determined from the default router list of reachable or probably reachable routers the switch detects from periodic multicast router advertisements (RAs) received on the interface. For a given interface, there can be multiple default gateways, with different nodes on the link using different gateways. If the switch does not detect any IPv6 routers that are reachable from a given interface, it assumes (for that interface) that it can reach only the other devices connected to the interface. See “Router access and default router selection” (page 35). NOTE: In IPv6 for the switches covered in this guide, the default route cannot be statically configured. Also, DHCPv6 does not include default route configuration.)

Enabling DHCPv6 Enabling the DHCPv6 option on a VLAN allows the switch to obtain a global unicast address. It also allows the switch to obtain additional information such as an NTP server address or a DNS server address that can be used by the switch.

Syntax: [no] ipv6 address dhcp full [ rapid-commit ] Configures DHCPv6 on a VLAN, which initiates transmission of DHCPv6 requests for service. If IPv6 is not already enabled on the VLAN by the ipv6 enable command, this option enables IPv6 and causes the switch to autoconfigure a link-local unicast address with an EUI-64 interface identifier. [ rapid-commit ] Expedites DHCP configuration by using a two-message exchange with the server (solicit-reply) instead of the default four-message exchange (solicit-advertise-request-reply). NOTE: A DHCPv6 server does not assign link-local addresses, and enabling DHCPv6 on a VLAN does not affect a pre-existing link-local address. A DHCPv6-assigned address can be configured on a VLAN when the following is true: •

The assigned address is not on the same subnet as a previously configured autoconfig address.



The maximum IPv6 address limit on the VLAN or the switch has not been reached.

If the switch is an IPv6 host, ipv6 address dhcp full must be configured on the DHCPv6 client to obtain relevant information from the DHCPv6 server. M-bit and O-bit settings in RAs from a router are not used by the switch in host mode. If Enabling DHCPv6

17

the switch is operating as an IPv6 router, it includes M-bit and O-bit values in the RAs it transmits. See chapter “IPv6 Router Advertisements” (page 193) for routing switch operation. After verification of uniqueness by DAD, an IPv6 address assigned to the VLAN by a DHCPv6 server is set to the preferred and valid lifetimes specified in an RA received on the VLAN for the prefix used in the assigned address and is configured as a preferred address. See section “Address lifetimes” (page 30). Default: Disabled The no form of the command removes the DHCPv6 option from the configuration and, if no other IPv6-enabling command is configured on the VLAN, disables IPv6 on the VLAN. See “About disabling IPv6 on a VLAN” (page 32).

Authenticating the DHCPv6 client DHCPv6 client authentication allows the configuration of authentication options such as mode and key-chain. For more information, see “DHCPv6 client” (page 31).

Syntax: [no] ipv6 dhcp-client authentication [ mode [ md5 ] | key-chain chain-name-str ] Allows the configuration of authentication options. The authentication information carried in the option can be used to identify the source of a DHCPv6 message and to confirm the message has not been tampered with. The no form of the ipv6 dhcp-client authentication command disables authentication. mode [ md5 ] The only supported mode is MD5-based HMAC. key-chain chain-name-str The name of the key chain. The key chain must have been created with an ID and a duration before it can be entered as the chain-name-str. See chapter “Key Management System" in the Access Security Guide for your switch for more information about creating key chains.

Viewing DHCP client information To view DHCP client authentication or statistical information, enter the show dhcpv6 client command with the appropriate option. To view information about a specific VLAN, enter the vlan command with the VLAN name or identifier.

Syntax: show ipv6 dhcp-client [ authentication [ vlan vid ] | statistics [ vlan vid ] ] Displays information about IPv6 DHCP client authentication parameters or DHCP client statistics. authentication Displays authentication information such as enabled or disabled, mode, key-chain, and key-id statistics Displays information such as transmitted packets, received packets, and authentication failures.

18

IPv6 Addressing Configuration

Example 1 DHCP client authentication show command for all VLANs HP Switch(config)# show ipv6 dhcp-client authentication DHCPv6 Authentication Information Vlan Name Authentication Authentication mode Key-Chain Key-Id

: : : : :

DEFAULT_VLAN Enabled HMAC-MD5 DHCP10 1

Vlan Name Authentication Authentication mode Key-Chain Key-Id

: : : : :

VLAN2 Enabled HMAC-MD5 DHCP10 1

Example 2 DHCP client authentication show command for a specific VLAN HP Switch(config)# show ipv6 dhcp-client authentication vlan 1 DHCPv6 Authentication Information Vlan Name Authentication Authentication Mode Key-Chain

: : : :

DEFAULT_VLAN Enabled HMAC-MD5 DHCP10

Example 3 DHCP client statistics for all VLANs HP Switch(config)# show ipv6 dhcp-client statistics DHCPv6 Statistics Information Vlan Name Transmitted Packets Received Packets Authentication Failures

: : : :

DEFAULT_VLAN 10000 9998 2

Vlan Name Transmitted Packets Received Packets Authentication Failures

: : : :

VLAN2 5000 3500 0

Example 4 DHCP client statistics for a single VLAN HP Switch(config)# show ipv6 dhcp-client statistics vlan 1 DHCPv6 Statistics Information Vlan Name Transmitted Packets Received Packets Authentication Failures

: : : :

DEFAULT_VLAN 10000 9998 2

DHCPv6 client authentication operating notes •

The DHCPv6 client must be stopped before setting the authentication mode or entering a key-chain name. Use the no ipv6 address dhcp full command to disable DHCPv6. Viewing DHCP client information

19

Commands are executed in VLAN context. •

You must disable the DHCPv6 client before disabling DHCPv6 authentication.



You must create a key chain with an ID and a duration before configuring DHCPv6 client authentication with a key-chain name.



The DHCPv6 client and DHCPv4 Relay can be enabled on the same interface and operate at the same time.

Viewing configured DHCPv6 addresses To view the current IPv6 DHCPv6 settings per-VLAN, use show run. To view all currently configured IPv6 unicast addresses, use the following: •

show ipv6 (Lists IPv6 addresses for all VLANs configured on the switch.)



show ipv6 vlan vid (Lists IPv6 addresses configured on the VLAN.)

For more information, see “Viewing the current IPv6 addressing configuration” (page 23).

General operating notes for DHCPv6 •

If multiple DHCPv6 servers are available, the switch selects a server based on the preference value sent in DHCPv6 messages from the servers.



The switch supports both DHCPv4 and DHCPv6 client operation on the same VLAN.



With IPv6 enabled, the switch determines the default IPv6 router for the VLAN from the RAs it receives. See “Default IPv6 router” (page 37).



DHCPv6 and statically configured global unicast addresses are mutually exclusive on a given VLAN. That is, configuring DHCPv6 on a VLAN erases any static global unicast addresses previously configured on that VLAN, and the reverse. (A statically configured link-local address is not affected by configuring DHCPv6 on the VLAN.)



For the same subnet on the switch, a DHCPv6 global unicast address assignment takes precedence over an autoconfigured address assignment, regardless of which address type was the first to be configured. If DHCPv6 is subsequently removed from the configuration, an autoconfigured address assignment replaces it after the next RA is received on the VLAN. DHCPv6 and autoconfigured addresses co-exist on the same VLAN if they belong to different subnets.

For related information see: •

RFC 3315: "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)"



RFC 3633: "IPv6 Prefix Options for Dynamic Host Configuration Protocol (DHCP) version 6"



RFC 3736: "Stateless Dynamic Host Configuration Protocol (DHCP) Service for IPv6"

Statically configuring a link-local unicast address Syntax: [no] ipv6 address fe80::interface-id link-local •

If IPv6 is not already enabled on the VLAN, this command enables IPv6 and configures a static link-local address.



If IPv6 is already enabled on the VLAN, this command overwrites the current, link-local address with the specified static address. (One link-local address is allowed per VLAN interface.)

interface-id The low-order 64 bits, in 16-bit blocks, comprise this value in a link-local address: 20

IPv6 Addressing Configuration

xxxx xxxx : xxxx xxxx : xxxx xxxx : xxxx xxxx Where a static link-local address is already configured, a new, autoconfigured global unicast addresses assignment uses the same interface identifier as the link-local address. NOTE: An existing link-local address is replaced, and is not deprecated, when a static replacement is configured. The prefix for a statically configured link-local address is always 64 bits, with all blocks after fe80 set to zero. That is: fe80:0:0:0. After verification of uniqueness by DAD, a statically configured link-local address status is set to preferred, with a permanent lifetime. See “Address lifetimes” (page 30). For link-local addressing, the no form of the static IPv6 address command produces different results, depending on how IPv6 is configured on the VLAN: •

If IPv6 was enabled only by a statically configured link-local address, deleting the link-local address disables IPv6 on the VLAN.



If other IPv6-enabling commands have been configured on the VLAN, deleting the statically configured link-local address causes the switch to replace it with the default (EUI-64) link-local address for the VLAN, and IPv6 remains enabled.

See also “About disabling IPv6 on a VLAN” (page 32).

Statically configuring a global unicast address Syntax: [no] ipv6 address [ network-prefix interface-id/prefix-length ] [no] ipv6 address [ network-prefix::/prefix-length ] eui-64 If IPv6 is not already enabled on a VLAN, either of these command options do the following: •

Enable IPv6 on the VLAN



Configure a link-local address using the EUI-64 format



Statically configure a global unicast address

If IPv6 is already enabled on the VLAN, the above commands statically configure a global unicast address, but have no effect on the current link-local address. network-prefix This includes the global routing prefix and the subnet ID for the address. interface-id Enters a user-defined interface identity. prefix-length Specifies the number of bits in the network prefix. If you are using the eui-64 option, this value must be 64. eui-64 Specifies using the EUI format to create an interface identifier based on the VLAN MAC address. After verification of uniqueness by DAD, the lifetime of a statically configured IPv6 address assigned to a VLAN is set to permanent and is configured as a preferred address. Statically configuring a global unicast address

21

The no form of the command erases the specified address and, if no other IPv6-enabling command is configured on the VLAN, disables IPv6 on the VLAN. See “About disabling IPv6 on a VLAN” (page 32).

Viewing the currently configured static IPv6 addresses per-VLAN To view the currently configured static IPv6 addresses per-VLAN, use show run. To view all currently configured IPv6 unicast addresses, use the following: •

show ipv6 (Lists IPv6 addresses for all VLANs configured on the switch.)



show ipv6 vlan vid (Lists IPv6 addresses configured on VLAN vid.)



show ipv6 tunnel tunnel-id (Lists IPv6 addresses configured on tunnel tunnel-id.)

For more information, see “Viewing the current IPv6 addressing configuration” (page 23).

Operating notes •

With IPv6 enabled, the switch determines the default IPv6 router for the VLAN from the RAs it receives. See “View IPv6 gateway, route, and router neighbors” (page 28).



If DHCPv6 is configured on a VLAN, configuring a static global unicast address on the VLAN removes DHCPv6 from the VLAN's configuration and deletes the DHCPv6-assigned global unicast address.



For a statically configured global unicast address to be routable, a gateway router must be transmitting RAs on the VLAN.



If an autoconfigured global unicast address already exists for the same subnet as a new, statically configured global unicast address, the statically configured address is denied. In the reverse case, you can add an autoconfig command to the VLAN configuration, but it is not implemented unless the static address is removed from the configuration.

Assigning an IPv6 address to a loopback interface The following command enables nondefault IPv6 address configuration on loopback interfaces. For more information on loopback interfaces, see “IPv6 loopback interfaces” (page 31).

Syntax: [no] interface loopback 0 - 7 ipv6 address ipv6-addr Configures an IPv6 address on a loopback interface identified by 0 - 7. Use the no form of the command to remove nondefault IPv6 addresses from the loopback interface. NOTE: You cannot remove the (reserved) default loopback interface address ::1/128 from lo-0. You can configure up to 32 IPv6 addresses (and up to 32 IPv4 addresses) on a loopback interface. To configure an IPv6 address for the loopback interface, enter the ipv6 addressip-address command at the loopback interface configuration level, as shown in the following example. When you configure an IPv6 address for a loopback interface, you do not specify a prefix. The default prefix/128 applies automatically.

22

IPv6 Addressing Configuration

Example 5 Configuring an IPv6 address on a loopback interface HP Switch(config)# interface loopback 1 HP Switch(lo-1)# IPv6 address 2001:db8::1

NOTE: •

You can configure a loopback interface only from the CLI; you cannot configure a loopback interface from the Menu interface.



IPv6 loopback interfaces share the same IPv6 address space with VLAN configurations. The maximum number of IPv6 addresses supported on a switch is 2048, which includes all IPv6 addresses configured for both VLANs and loopback interfaces (except for the default loopback IPv6 address, ::1 /128).



Each IPv6 address that you configure on a loopback interface must be unique in the AS; see “OSPFv3 Routing” (page 218). This means that the address cannot be used by a VLAN interface or another loopback interface.



You can configure up to 32 IPv6 and 32 IPv4 addresses on a loopback interface (lo0 to lo7).

Displaying loopback interface configurations Use show ipv6 to display the list of loopback interfaces configured with nondefault IPv6 addresses. (Loopback interface 0, if configured only with the default ::1/128 IPv6 address, does not appear in this listing.) NOTE: A loopback interface does not appear in the show ipv6 command output unless it is configured with a non-default IPv6 address. For information about displaying IPv6 on a VLAN, see “About disabling IPv6 on a VLAN” (page 32).

Viewing the current IPv6 addressing configuration Use these commands to view the current status of the IPv6 configuration on the switch.

Syntax: show ipv6 Lists the current, global IPv6 settings and per-VLAN IPv6 addressing on the switch. IPv6 Routing For software releases K.13.01 through K.14.01, this setting is always Disabled. This is a global setting and is not configured per-VLAN. See “View IPv6 gateway, route, and router neighbors” (page 28). Default Gateway Lists the IPv4 default gateway, if any, configured on the switch. This is a globally configured router gateway address and is not configured per-VLAN. ND DAD Indicates whether DAD is enabled (the default) or disabled. Using ipv6 nd dad-attempts 0 disables ND. See “Duplicate address detection (DAD)” (page 33). DAD Attempts Indicates the number of neighbor solicitations the switch transmits per-address for duplicate (IPv6) address detection. Implemented when a new address is configured or when an interface with configured addresses comes up (such as after a reboot). Viewing the current IPv6 addressing configuration

23

Default: 3; Range: 0 – 255 ms. 0 disables duplicate address detection. See “Duplicate address detection (DAD)” (page 33). VLAN Name Lists the name of a VLAN statically configured on the switch. VLAN Name Lists the name of a VLAN statically configured on the switch. IPv6 Status For the indicated VLAN, shows whether IPv6 is disabled (the default) or enabled. See “Configuring IPv6 addressing” (page 14). Address Origin Autoconfig The address was configured using SLAAC. In this case, the interface identifier for global unicast addresses copied from the current link-local unicast address. DHCP The address was assigned by a DHCPv6 server. Addresses having a DHCP origin are listed with a 128-bit prefix length. Manual The address was statically configured on the VLAN. IPv6 Address/Prefix Length Lists each IPv6 address and prefix length configured on the indicated VLAN. Address Status Tentative DAD has not yet confirmed the address as unique, and it is not usable for sending and receiving traffic. Preferred The address has been confirmed as unique by DAD and usable for sending and receiving traffic. The Expiry time shown for this address by the show ipv6 vlan vid command output is the preferred lifetime assigned to the address. See “Address lifetimes” (page 30). Deprecated The preferred lifetime for the address has been exceeded, but there is time remaining in the valid lifetime. Duplicate Indicates a statically configured IPv6 address that is a duplicate of another IPv6 address that already exists on another device belonging to the same VLAN interface. A duplicate address is not used. Example 6 (page 25) shows the output on a switch having IPv6 enabled on one VLAN.

24

IPv6 Addressing Configuration

Example 6 show IPv6 command output HP Switch(tunnel-3)# show ipv6 Internet (IPv6) Service IPv6 Routing ND DAD DAD Attempts

: Enabled : Enabled : 3

Interface Name IPv6 Status Layer 3 Status

VLAN Interfaces : DEFAULT_VLAN : Disabled : Enabled

Interface Name IPv6 Status Layer 3 Status

: VLAN22 : Enabled : Enabled

Address Origin ---------autoconfig

| | IPv6 Address/Prefix Length + ---------------------------------| fe80::218:71ff:feb9:8500/64

Address Status --------tentative

Tunnel Interfaces Interface Name IPv6 Status Layer 3 Status Address Origin ---------autoconfig

: TUNNEL3 : Enabled : NA

| | IPv6 Address/Prefix Length + --------------------------------—| fe80::218:71ff:feb9:8500/64

Address Status --------tentative

Syntax: show ipv6 nd Displays the current IPv6 ND settings on the configured VLAN interfaces. start here “show IPv6 nd output with default settings” (page 25) shows the output on a switch having IPv6 enabled on VLANs 1 and 22. Example 7 show IPv6 nd output with default settings HP Switch# show ipv6 nd IPV6 Neighbor Discovery Configuration Interface Name --------1 22

DAD Attempts --------3 3

RCH Time (msecs) ---------30000 30000

NS Interval (msecs) ----------1000 1000

Syntax: show ipv6 vlan [ vlan vid | tunnel tunnel-id ] Viewing the current IPv6 addressing configuration

25

Displays IP and IPv6 global configuration settings, the IPv6 status for the specified VLAN, the IPv6 addresses (with prefix lengths) configured on the specified VLAN, and the expiration data (Expiry) for each address. IPv6 Routing For software releases K.13.01 through K.14.01, this setting is always Disabled. See “View IPv6 gateway, route, and router neighbors” (page 28). Default Gateway Lists the IPv4 default gateway, if any, configured on the switch. This is a globally configured router gateway address and is not configured per-VLAN ND DAD Shows whether ND is enabled. The default setting is Enabled. Using ipv6 nd dad- attempts 0 disables ND. DAD Attempts Indicates the number of neighbor solicitations the switch transmits per-address for duplicate (IPv6) address detection. Implemented when a new address is configured or when an interface with configured addresses comes up (such as after a reboot). The default setting is 3, and the range is 0 to 255. A setting of "0" disables DAD. See “Duplicate address detection (DAD)” (page 33). VLAN Name Lists the name of a VLAN statically configured on the switch. IPv6 Status For the indicated VLAN, shows whether IPv6 is disabled (the default) or enabled. See “Configuring IPv6 addressing” (page 14). IPv6 Address/Prefix Length Lists each IPv6 address and prefix length configured on the indicated VLAN. Expiry Lists the lifetime status of each IPv6 address listed for a VLAN: Permanent The address does not time out and need renewal or replacement. date/time The date and time that the address expires. Expiration date and time is specified in the router advertisement used to create the prefix for automatically configured, global unicast addresses. The Address Status field in the show ipv6 command output indicates whether this date/time is for the preferred or valid lifetime assigned to the corresponding address. See “Address lifetimes” (page 30).

26

IPv6 Addressing Configuration

Example 8 show IPv6 VLAN vid output HP Switch# show ipv6 vlan 10 Internet (IPv6) Service IPv6 Routing Default Gateway ND DAD DAD Attempts

: : : :

Disabled fe80::213:c4ff:fedd:14b0%vlan10 Enabled 3

Vlan Name IPv6 Status

: VLAN10 : Enabled

IPv6 Address/Prefixlength Expiry --------------------------------- -----------------------2001:db8:a03:e102::1:101/64 Fri May 19 11:51:15 2009 fe80::1:101/64

permanent

Syntax: show run In addition to the other elements of the current configuration, this command lists the statically configured, global unicast IPv6 addressing and the current IPv6 configuration per-VLAN. The listing may include one or more of the following, depending on what other IPv6 options are configured on the VLAN. Any SLAAC commands in the configuration are also listed in the output, but the actual addresses resulting from these commands are not included in the output. ipv6 ipv6 ipv6 ipv6 ipv6 ipv6

enable address fe80::interface-id link-local address prefix:interface-id/prefix-length address autoconfig address dhcp full [ rapid-commit ] global-unicast-address/prefix

Viewing the current IPv6 addressing configuration

27

Example 9 show run output listing the current IPv6 addressing commands HP Switch(config)# show run Running configuration: . . . vlan 10 name "VLAN10" untagged 1-12 ipv6 address fe80::1:101 link-local 1 ipv6 address dhcp full rapid-commit 2 . . . 1 2

Statically configured IPv6 addresses appear in the show run output. Commands for automatic IPv6 address configuration appear in the show run output, but the addresses resulting from these commands do not appear in the output.

View IPv6 gateway, route, and router neighbors Use these commands to view the switch's current routing table content and connectivity to routers per VLAN. This includes information received in RAs from IPv6 routers on VLANs enabled with IPv6 on the switch.

Viewing gateway and IPv6 route information Syntax: show ipv6 route ipv6-addr connected Displays the routes in the switch's IPv6 routing table. ipv6-addr Optional. Limits the output to show the gateway to the specified IPv6 address. connected Optional. Limits the output to show only the gateways to IPv6 addresses connected to VLAN interfaces configured on the switch, including the loopback (::1/128) address. Dest The destination address for a detected route. Gateway The IPv6 address or VLAN interface used to reach the destination. (Includes the loopback address.) Type Indicates route type (static, connected, or OSPF). Distance The route's administrative distance, used to determine the best path to the destination. Metric Indicates the route cost for the selected destination.

28

IPv6 Addressing Configuration

Example 10 show IPv6 route output Examples of addresses in ouput: “Unknown” Address Dest : ::/0 Gateway : fe80::213:c4ff:fedd:14b0%vlan10

Type : static Dist. : 40 Metric : 0

Loopback Address Dest : ::1/128 Gateway : lo0

Type : connected Dist. : 0 Metric : 1

Global Unicast Address Configured on the Switch Dest : 2001:db8:a03:e102::/64 Gateway : VLAN10

Type : connected Dist. : 0 Metric : 1

Link-Local Address Configured on the Switch Dest : fe80::%vlan10 Gateway : VLAN10

Type : connected Dist. : 0 Metric : 1

Link-Local Address Assigned to the Loopback Address Dest : fe80::1%lo0 Gateway : lo0

Type : connected Dist. : 0 Metric : 1

Viewing IPv6 router information Syntax: show ipv6 routers [ vlan vid ] Lists the switch's IPv6 router table entries for all VLANs configured on the switch or for a single VLAN. This output provides information about the IPv6 routers from which RAs have been received on the switch. NOTE: This command reports on IPv6 routers the switch has learned of through operation as a host or client of a router. If the switch itself is configured as an IPv6 router (routing switch), the output of this command is empty. vlan vid Optional. Specifies only the information on IPv6 routers on the indicated VLAN. Router Address The IPv6 address of the router interface. Preference The relative priority of prefix assignments received from the router when prefix assignments are also received on the same switch VLAN interface from other IPv6 routers. Interface The VLAN interface on which the router exists. MTU The maximum transmission unit (in bytes) allowed for frames on the path to the indicated router. Hop Limit The maximum number of router hops allowed. View IPv6 gateway, route, and router neighbors

29

Prefix Advertised Lists the prefix and prefix size (number of leftmost bits in an address) originating with the indicated router. Valid Lifetime The total time the address is available, including the preferred lifetime and the additional time (if any) allowed for the address to exist in the deprecated state. See “Valid lifetime” (page 30). Preferred Lifetime The length of time during which the address can be used freely as both a source and a destination address for traffic exchanges with other devices. See “Preferred lifetime” (page 30). On/Off Link Indicates whether the entry source is on the same VLAN as is indicated in the Interface field. Example 11 (page 30) indicates that the switch is receiving router advertisements from a single router that exists on VLAN 10. Example 11 show IPv6 routers output HP Switch(config)# show ipv6 routers IPv6 Router Table Entries Router Address Preference Interface MTU Hop Limit

: : : : :

fe80::213:c4ff:fedd:14b0 Medium VLAN10 1500 64

Valid Prefix Advertised Lifetime(s) -------------------------- -----------2001:db8:a03:e102::/64 864000

Preferred Lifetime(s) -----------604800

On/Off Link ------Onlink

Address lifetimes Every configured IPv6 unicast address has a lifetime setting that determines how long the address can be used before it must be refreshed or replaced. Some addresses are set as "permanent" and do not expire. Others have both a "preferred" and a "valid" lifetime that specify the duration of their use and availability.

Preferred lifetime This is the length of time during which the address can be used freely as both a source and a destination address for traffic exchanges with other devices. This time span is equal to or less than the valid lifetime also assigned to the address. If this time expires without the address being refreshed, the address becomes deprecated and should be replaced with a new, preferred address. In the deprecated state, an address can continue to be used as a destination for existing communication exchanges but is not used for new exchanges or as a source for traffic sent from the interface. A new, preferred address and its deprecated counterpart both appear in the show ipv6 vlan vid output as long as the deprecated address is within its valid lifetime.

Valid lifetime The valid lifetime, which is the total time the address is available, is equal to or greater than the preferred lifetime. The valid lifetime enables communication to continue for transactions that began 30

IPv6 Addressing Configuration

before the address became deprecated. However, in this time frame, the address should no longer be used for new communications. If this time expires without the deprecated address being refreshed, the address becomes invalid and may be assigned to another interface. Table 2 IPv6 unicast addresses lifetimes Address source

Lifetime criteria

Link-local

Permanent

Statically configured unicast

Permanent

Autoconfigured global

Finite preferred and valid lifetimes

DHCPv6-configured

Finite preferred and valid lifetimes

A new, preferred address used as a replacement for a deprecated address can be acquired from a manual, DHCPv6, or autoconfiguration source.

DHCPv6 client The DHCPv6 client allows a host to request global unicast IPv6 address assignments from a DHCPv6 server. If there are multiple DHCPv6 servers, the client can select a server based on the preference value sent in DHCPv6 messages. The DHCPv6 client can request that the server send only the configuration information. In this case, a router on the same interface (VLAN) as the host provides the global IPv6 address to the host through router advertisements. NOTE: If the switch is rebooted with a default configuration, only the default DHCPv4 client is enabled on the default VLAN. The DHCPv6 client has to be explicitly enabled on a VLAN using the command ipv6 address dhcp or ipv6 address autoconfig.

Duplicate address detection (DAD) for statically configured addresses Statically configured IPv6 addresses are designated as permanent. If DAD determines that a statically configured address duplicates a previously configured and reachable address on another device belonging to the VLAN, the more recent, duplicate address is designated as duplicate. For more information, see: •

“Duplicate address detection (DAD)” (page 33)



“Viewing the current IPv6 addressing configuration” (page 23)

IPv6 loopback interfaces By default, each switch has eight internal IPv6 loopback interfaces (lo-0 as through lo-7) with IPv6 address ::1/128 configured by default on lo-0. This address (::1/128) is used only for internal traffic transmitted within the switch and is not used in packet headers in egress traffic sent to network devices. Each loopback interface can have multiple IPv6 addresses, all of which must be unique. Routing protocols such as OSPFv3, advertise the configured loopback addresses throughout a network or autonomous system. User-defined IPv6 loopback addresses provide these benefits when a routing protocol is enabled: •

A loopback interface is a virtual interface that is always up and reachable as long as at least one of the IPv6 interfaces on the switch is operational. As a result, a loopback interface is useful for debugging tasks because its address can always be pinged if any other switch interface is up.



You can use a loopback interface to establish a Telnet session, ping the switch, and access the switch through SNMP. DHCPv6 client

31

For information about how to configure an IPv6 loopback address to participate in an OSPF broadcast area, see “Assigning IPv6 loopback addresses to an area (optional)” (page 225).

About disabling IPv6 on a VLAN While one IPv6-enabling command is configured on a VLAN, IPv6 remains enabled on that VLAN. In this case, removing the only IPv6-enabling command from the configuration disables IPv6 operation on the VLAN. That is, to disable IPv6 on a VLAN, the following commands must be removed from the VLAN's configuration: ipv6 ipv6 ipv6 ipv6 ipv6

enable address address address address

dhcp full rapid-commit autoconfig fe80::interface-id link-local prefix:interface-id

If any of the above remain enabled, IPv6 remains enabled on the VLAN and, at a minimum, a link-local unicast address is present.

Neighbor discovery Neighbor Discovery (ND) is the IPv6 equivalent of the IPv4 ARP for layer 2 address resolution. ND uses IPv6 ICMP messages to provide for discovery of IPv6 devices such as other switches, routers, management stations, and servers on the same interface. ND runs automatically in the default configuration and provides services in addition to those provided in IPv4 by ARP. For example: •

Determine the link-layer address of neighbors on the same VLAN or tunnel interface.



Verify that a neighbor is reachable. Track neighbor (local) routers.

• Track neighbor (local) routers. Neighbor Discovery enables functions such as the following: •

router and neighbor solicitation and discovery



detecting address changes for devices on a VLAN or tunnel



identifying a replacement for a router or router path that has become unavailable



duplicate address detection (DAD)



router advertisement processing



neighbor reachability



autoconfiguration of unicast addresses



resolution of destination addresses



changes to link-layer addresses

An instance of ND is triggered on a device when a new (tentative) or changed IPv6 address is detected. (This includes stateless, stateful, and static address configuration.) ND operates in a per-VLAN scope, that is, within the VLAN on which the device running the ND instance is a member. ND actually occurs when there is communication between devices on a VLAN. That is, a device needing to determine the link-layer address of another device on the VLAN initiates a (multicast) neighbor solicitation message (containing a solicited-node multicast address that corresponds to the IPv6 address of the destination device) on the VLAN. When the destination device receives the neighbor solicitation, it responds with a neighbor advertisement message identifying its link-layer address. When the initiating device receives this advertisement, the two devices are ready to exchange traffic on the VLAN interface. Also, when an IPv6 interface becomes operational, it transmits a router solicitation on the interface and listens for an RA.

32

IPv6 Addressing Configuration

NOTE: Neighbor and router solicitations must originate on the same VLAN as the receiving device. To support this operation, IPv6 is designed to discard any incoming neighbor or router solicitation that does not have a value of 255 in the IP Hop Limit field. For a complete list of requirements, see RFC 246. When a pair of IPv6 devices in a VLAN exchange communication, they enter each other's IPv6 and corresponding MAC addresses in their respective neighbor caches. These entries are maintained for a time after communication ceases and then dropped. To view or clear the content of the neighbor cache, see “Viewing the neighbor cache” (page 38). For related information, see RFC 2461: "Neighbor Discovery for IP Version 6 (IPv6)."

Duplicate address detection (DAD) DAD verifies that a configured unicast IPv6 address is unique before it is assigned to a VLAN interface on the switch. DAD is enabled in the default IPv6 configuration and can be reconfigured, disabled, or re-enabled at the global config or per-interface command level. DAD can be useful in helping to troubleshoot erroneous replies to DAD requests, or where the neighbor cache contains a large number of invalid entries caused by an unauthorized station sending false replies to the switch's ND queries. If DAD verifies that a unicast IPv6 address is a duplicate, the address is not used. If the link-local address of the VLAN interface is found to be a duplicate of an address for another device on the interface, the interface stops processing IPv6 traffic.

DAD operation On a given VLAN interface, when a new unicast address is configured, the switch runs DAD for this address by sending a neighbor solicitation to the All-Nodes multicast address (ff02::1). This operation discovers other devices on the VLAN and verifies whether the proposed unicast address assignment is unique on the VLAN. (During this time, the address being checked for uniqueness is held in a tentative state and cannot be used to receive traffic other than neighbor solicitations and neighbor advertisements.) A device that receives the neighbor solicitation responds with a neighbor advertisement that includes its link-local address. If the newly configured address is from a static or DHCPv6 source and is found to be a duplicate, it is labeled as duplicate in the "Address Status" field of the show ipv6 command and is not used. If an autoconfigured address is found to be a duplicate, it is dropped and the following message appears in the Event Log: W date time 00019 ip: ip address IPv6-address removed from vlan id vid DAD does not perform periodic checks of existing addresses. However, when a VLAN comes up with IPv6 unicast addresses configured (as can occur during a reboot), the switch runs DAD for each address on the interface by sending neighbor solicitations to the All-Nodes multicast address, as described above. If an address is configured while DAD is disabled, the address is assumed to be unique and is assigned to the interface. If you want to verify the uniqueness of an address configured while DAD was disabled, re-enable DAD and then either delete and reconfigure the address, or reboot the switch.

Configuring DAD Syntax: ipv6 nd dad-attempts 0 - 255 This command is executed at the global or per-interface config level, and configures the number of neighbor solicitations to send when performing DAD for a unicast address configured on a VLAN or tunnel interface. A per-interface configuration overrides a globally set configuration. Configuring DAD

33

0 - 255 The number of consecutive neighbor solicitation messages sent for DAD inquiries on an interface. Setting this value to 0 disables DAD on the interface, which bypasses checks for uniqueness on newly configured addresses. If a reboot is performed while DAD is disabled, the duplicate address check is not performed on any IPv6 addresses configured on the switch. Default: 3 (enabled); 0 (disabled); Range: 0 - 255 (0 = disabled) The no form of the command restores the default setting (3). NOTE: Software version K.14.xx supports a dad-attempts range of 0 to 600. However, software version K.15.xx or greater supports a range of 0 to 255. If dad-attempts is set higher than 255, updating from K.14.xx to K.15.xx or greater and rebooting truncates this value to 255. Similarly, if the switch is running version K.15.xx or greater: (1) downloading a configuration file created using version K.14.xx with dad-attempts set higher than 255 and then (2) rebooting the switch using this configuration file truncates the setting to 255.

Syntax: ipv6 nd ns-interval milliseconds Used on VLAN interfaces to reconfigure the ND time in milliseconds between DAD neighbor solicitations sent for an unresolved destination, or between duplicate address detection neighbor solicitation requests. Increasing this setting is indicated where neighbor solicitation retries or failures are occurring, or in a "slow" (WAN) network. This value can be configured in a router advertisement to help ensure that all hosts on a VLAN are using the same retransmit interval for ND. See “Setting or changing the Neighbor Discovery retransmit timer” (page 197). To view the current setting, use show ipv6 nd. Default: 1000 ms; Range: 1000 to 3600000 ms

Syntax: ipv6 nd reachable-time milliseconds Used on VLAN interfaces to configure the length of time in milliseconds a neighbor is considered reachable after the Neighbor Unreachability Detection algorithm has confirmed it to be reachable. When the switch operates in host mode, this setting can be overridden by a reachable time received in a router advertisement. This value can be configured in an RA a router advertisement to help ensure that all hosts on a VLAN are using the same reachable time in their neighbor cache. To view the current setting, use show ipv6 nd. Default: 30000 ms; Range: 1000 to 2147483647 ms

Operating notes for ND

34



A verified link-local unicast address must exist on a VLAN interface before the switch can run DAD on other addresses associated with the interface.



If a previously configured unicast address is changed, a neighbor advertisement (an all-nodes multicast message--ff02::1) is sent to notify other devices on the VLAN and to perform DAD.



IPv6 addresses on a VLAN interface are assigned to multicast address groups identified with well-known prefixes.



DAD is performed on all stateful, stateless, and statically configured unicast addresses.

IPv6 Addressing Configuration



Neighbor solicitations for DAD do not cause the neighbor cache of neighboring switches to be updated.



If a previously configured unicast address is changed, a neighbor advertisement is sent on the VLAN to notify other devices and for duplicate address detection.



If DAD is disabled when an address is configured, the address is assumed to be unique and is assigned to the interface.

Router access and default router selection Traffic can be routed between destinations on different VLANs configured on the switch or to a destination on an off-switch VLAN. This is done by placing the switch on the same VLAN interface or subnet as an IPv6-capable router configured to route traffic to other IPv6 interfaces or to tunnel IPv6 traffic across an IPv4 network.

Router advertisements An IPv6 router periodically transmits RAs on the VLANs to which it belongs to notify other devices of its presence. The switch uses these advertisements for purposes such as: •

Learning the MAC and link-local addresses of IPv6 routers on the VLAN. (For devices other than routers, the switch must use ND to learn these addresses.)



Building a list of default (reachable) routers, along with router lifetime and prefix lifetime data.



Learning the prefixes and the valid and preferred lifetimes to use for stateless (autoconfigured) global unicast addresses. (This is required for autoconfiguration of global unicast IPv6 addresses.)



Learning the hop limit for traffic leaving the VLAN interface.



Learning the MTU to apply to frames intended to be routed.

IPv6 router advertisement options for DNS configuration Two new options in IPv6 Router Advertisements allow IPv6 routers to advertise a list of recursive DNS Server (RDNSS) addresses and a DNS Search List (DNSSL) to IPv6 hosts. RA-based DNS configuration enables the full configuration of basic networking information for hosts without requiring DHCPv6. An IPv6 host can acquire the DNS configuration (that is, the DNS recursive server addresses and DNS Search List) for the links to which the host is connected. The host learns this DNS configuration from the same RA message that provides configuration information for the link. DNS options are included by default in every emitted RA unless the inclusion is suppressed via CLI commands or the SNMP MIB. The suppress option can be configured either globally or for an IP interface. NOTE:

This is supported in RFC 6106

Global configuration This command suppresses the inclusion of RDNSS and SNSSL in outgoing Router Advertisements across all interfaces on a switch. Syntax [no]nd suppress-ra-dns Suppresses DNS options in router advertisements.

Router access and default router selection

35

Example 12 Suppressing the inclusion of RDNSS and SNSSL in outgoing RAs globally HP Switch(config)# ipv6 nd suppress-ra-dns

IP interface configuration This example shows the command that suppresses the inclusion of RDNSS and SNSSL in outgoing Router Advertisements for an IP interface. The command is executed in VLAN context. Example 13 Suppressing the inclusion of RDNSS and SNSSL in outgoing RAs for an IP interface HP Switch(config)# vlan 2 HP Switch(vlan-2)# ipv6 nd ra suppress-dns

Displaying the configuration The following example is the output showing which interfaces has RA-DNS suppressed. Example 14 Displaying the configuration HP Switch(config)# show ipv6 nd ra IPv6 Router Advertisement Configuration Global RA Suppress: : No Global RA-DNS Suppress : Yes Global Hop Limit : 64 IPv6 Unicast Routing : Disabled

Interface --------vlan-1 vlan-2 vlan-3

Suppress RA -------No No No

Suppress RA-DNS -------No Yes No

Interval Min/Max -------200/600 200/600 200/600

Lifetime (sec) -------1800 1800 1800

Mngd Flag ---No No No

Other Flag ----No No No

RCH Time (ms) -------0 0 0

Interval (ms) -------0 0 0

Hop Lim --64 64 64

Example 15 Example shows the output from running-config command with router advertisement DNS suppressed HP Switch(config)# show running-config Running configuration: ; J8698A Configuration Editor; Created on release #K.15.10.XXXX ; Ver #03:01.1f.ef:f2 Hostname “HP Switch” module 1 type J9550A ipv6 nd suppress-ra-dns vlan 1 name “DEFAULT_VLAN” untagged A1-A24 ip address dhcp-bootp exit vlan 2 name “vlan2” no ip address ipv6 nd ra suppress-dns exit spanning-tree

Router solicitations When an IPv6 interface becomes operational on the switch, a router solicitation is automatically sent to trigger an RA from any IPv6 routers reachable on the VLAN. (Router solicitations are sent to the All-Routers multicast address; ff02::2. If an RA is not received within one second of sending 36

IPv6 Addressing Configuration

the initial router solicitation, the switch sends up to three additional solicitations at intervals of four seconds. If an RA is received, the sending router is added to the switch's default router list and the switch stops sending router solicitations. If an RA is not received, IPv6 traffic on that VLAN cannot be routed, and the only usable unicast IPv6 address on the VLAN is the link-local address. NOTE: If the switch does not receive an RA after sending the router solicitations, as described above, no further router solicitations are sent on that VLAN unless a new IPv6 setting is configured, IPv6 on the VLAN is disabled and then re-enabled, or the VLAN itself is disconnected and then reconnected.

Default IPv6 router If IPv6 is enabled on a VLAN where there is at least one accessible IPv6 router, the switch selects a default IPv6 router. See “Enabling autoconfiguration of a global unicast address and a default router identity on a VLAN” (page 16). •

If the switch receives RAs from a single IPv6 router on the same VLAN or subnet, the switch configures a global unicast address and selects the advertising router as the default IPv6 router.



If multiple IPv6 routers on a VLAN send RAs advertising the same network, the switch configures one global unicast address and selects one router as the default router, based on the router's relative reachability, using factors such as router priority and route cost.



If multiple IPv6 routers on a VLAN send RAs advertising different subnets, the switch configures a corresponding global unicast address for each RA and selects one of the routers as the default IPv6 router, based on route cost. When multiple RAs are received on a VLAN, the switch uses the router priority and route cost information included in the RAs to identify the default router for the VLAN.

Router redirection With multiple routers on a VLAN, if the default (first-hop) router for an IPv6-enabled VLAN on the switch determines that there is a better first-hop router for reaching a given remote destination, the default router can redirect the switch to use that other router as the default router. For further information on routing IPv6 traffic, see the documentation provided for the IPv6 router. For related information, see RFC 2461: "Neighbor Discovery for IP Version 6."

Router access and default router selection

37

2 IPv6 Management Features Table 3 Summary of Commands Command syntax

Description

show ipv6 neighbors [ vlan vid ] clear ipv6 neighbors

An IPv6 neighbor cache is populated as a result of communication with other devices on the same VLAN. n/a

38, 40

telnet link-local-addr %vlan vid telnet global-unicast-addr show telnet [no] telnet-server [ listen [ oobm | data | both ]]

Telnet operation for IPv6 on the switch.

Enabled

41, 42, 43

[no] sntp server priority 1 - 3 link-local-addr %vlan 1 - 3 [ oobm ] [ 1 - 7 ] [no] sntp server priority 1 - 3 globalunicast-addr [ oobm ] [ 1 - 7 ]

Configures an IPv6 address for an SNTP server.

None

44

show timep timesync [ sntp | timep ] ip timep dhcp [ interval ] 1 - 9999 ip timep manual ipv6-addr interval 1 - 9999 [no] ip timep

Timep and related commands, including an example of using an IPv6 address. None

46

[no] tftp client | server [ listen oobm | | both ] copy tftp target ipv6-addr filename [ oobm ] auto-tftp ipv6-addr filename

Default

Upload or download data files to and from a physically connected device or a remote TFTP server. -

CLI page reference

48

snmp-server host [ ipv6-addr | ipv4-addr ] snmpv3 targetaddress name params params_name

Configure SNMP Trap Receivers. None

51

snmp-server host [ ipv6-addr | ipv4-addr ] [ community-name ] [ none | all | non-info | critical | debug ] [ inform [ retries count ] [ timeout interval ]]

SNMP Configuration Commands Supported SNMPv1 and v2c

52

snmpv3 targetaddress name params params_name [ ipv4-addr | ipv6-addr ] [ addr-mask ip4-addr ] [ filter none | debug | all | not-info | critical [ max-msg-size 484-65535 ] [ port-mask tcp-udp port ] [ retries 0 - 255 ] [ taglist tag_name ] [ timeout 0 - 2147483647 ] [ udp-port port-number ]

SNMP Configuration Commands Supported SNMPv3

52

]

This chapter focuses on the IPv6 application of management features that support both IPv6 and IPv4 operation. For additional information on these features, see the current Management and Configuration Guide for your switch.

Viewing the neighbor cache Neighbor discovery occurs when there is communication between the switch and another, reachable IPv6 device on the same VLAN. A neighbor destination is reachable from a given source address 38

IPv6 Management Features

if a confirmation (neighbor solicitation) has been received at the source verifying that traffic has been received at the destination. The switch maintains an IPv6 neighbor cache that is populated as a result of communication with other devices on the same VLAN. You can view the contents of the neighbor cache using the commands described in this section. For more information, see “View the neighbor cache” (page 55).

Syntax: show ipv6 neighbors [ vlan vid ] Displays IPv6 neighbor information currently held in the neighbor cache. After a period without communication with a given neighbor, the switch drops that neighbor's data from the cache. The command lists neighbors for all VLAN interfaces on the switch or for only the specified VLAN. The following fields are included for each entry in the cache: IPv6 Address Lists the 128-bit addresses for the local host and any neighbors (on the same VLAN) with whom there has been recent communication. MAC Address The MAC Address corresponding to each of the listed IPv6 addresses. VLAN vid Optional. Causes the switch to list only the IPv6 neighbors on a specific VLAN configured on the switch. Type Appears only when VLAN is not specified and indicates whether the corresponding address is local (configured on the switch) or dynamic (configured on a neighbor device). Age Appears only when the VLAN is specified and indicates the length of time the entry has remained unused. Port Identifies the switch port on which the entry was learned. If this field is empty for a given address, the address is configured on the switch itself. State A neighbor destination is reachable from a given source address if confirmation has been received at the source verifying that traffic has been received at the destination. This field shows the reachability status of each listed address: INCMP (Incomplete): Neighbor address resolution is in progress, but has not yet been determined. REACH (Reachable): The neighbor is known to have been reachable recently. STALE A timeout has occurred for reachability of the neighbor, and an unsolicited discovery packet has been received from the neighbor address. If the path to the neighbor is then used successfully, this state is restored to REACH.

Viewing the neighbor cache

39

DELAY Indicates waiting for a response to traffic sent recently to the neighbor address. The time for determining the neighbor's reachability has been extended. PROBE The neighbor may not be reachable. Periodic, unicast neighbor solicitations are being sent to verify reachability. Example 16 Neighbor cache without specifying a VLAN HP Switch(config)# show ipv6 IPv6 ND Cache Entries IPv6 Address --------------------------2001:db8:260:212::101 2001:db8:260:214::1:15 fe80::1:1 fe80::10:27 fe80::213:c4ff:fedd:14b0

neighbor MAC Address ------------0013c4-dd14b0 001279-88a100 001279-88a100 001560-7aadc0 0013c4-dd14b0

State ----STALE REACH REACH REACH REACH

Type ------dynamic local local dynamic dynamic

Port ---1

3 1

Example 17 Neighbor cache content for a specific VLAN HP Switch(config)# show ipv6 neighbor vlan 10 IPv6 ND Cache Entries IPv6 Address -----------------------2001:db8:260:212::101 2001:db8:260:214::1:15 fe80:1a3::1:1 fe80:::10:27 fe80::213:c4ff:fedd:14b0

MAC Address ------------0013c4-dd14b0 001279-88a100 001279-88a100 001560-7aadc0 0013c4-dd14b0

State ----STALE REACH REACH REACH REACH

Age ------------5h:13m:44s 11h:15m:23s 9h:35m:11s 22h:26m:12s 23 0h:32m:36s

Port ---1 17 12 3 1

Clearing the Neighbor Cache You can clear the contents of the neighbor cache using the commands described in this section. For more information, see “Clear the neighbor cache” (page 56).

Syntax: clear ipv6 neighbors Executed at the global config level, this command removes all non-local IPv6 neighbor addresses and corresponding MAC addresses from the neighbor cache, except neighbor entries specified as next-hops for active routes. Note that the Layer-2 address information for any next-hop route is cleared until the route is refreshed in the neighbor cache.

40

IPv6 Management Features

Example 18 Clearing the IPv6 neighbors cache HP Switch(config)# clear ipv6 neighbors HP Switch(config)# show ipv6 neighbors HP Switch# show ipv6 neighbors IPv6 ND Cache Entries IPv6 Address MAC Address State Type Port --------------------------- ------------- ----- ------- ---fe80::213:c4ff:fedd:14b0 000000-000000 INCMP dynamic 1 1

For an active-route next-hop, the MAC address and source port data is removed, and the State is set to “Incomplete” (INCMP) until the route is refreshed in the neighbor cache.

IPv6 Telnet operation This section describes Telnet operation for IPv6 on the switch. For IPv4 Telnet operation, see the Management and Configuration Guide for your switch.

Using outbound Telnet to another device Syntax: telnet link-local-addr %vlan vid [ oobm ] telnet global-unicast-addr [ oobm ] Outbound Telnet establishes a Telnet session from the switch CLI to another IPv6 device and includes these options. •

Telnet for link-local addresses on the same VLAN requires the link-local address and interface scope: link-local-addr : Specifies the link-local IPv6 address of the destination device. %vlan vid : Suffix specifying the interface on which the destination device is located. No spaces are allowed in the suffix.



Telnet for global unicast addresses requires a global unicast address for the destination. Also, the switch must be receiving RAs from an IPv6 gateway router. global-unicast-addr : Specifies the global IPv6 address of the destination device.

For switches that have a separate OOBM port, the oobm parameter specifies that the Telnet traffic goes out from the OOBM interface. If this parameter is not specified, the Telnet traffic goes out from the data interface. The oobm parameter is not available on switches that do not have a separate OOBM port. For more information on OOBM, see "Network Out-of-Band Management" Appendix in the Management and Configuration Guide.

IPv6 Telnet operation

41

Example 19 Telnet to another device To Telnet to another IPv6 device having a link-local address of fe80::215:60ff:fe79:8980 and on the same VLAN interface (VLAN 10), use the following command: HP Switch(config)# telnet fe80::215:60ff:fe79:980%vlan10 If the switch is receiving RAs from an IPv6 default gateway router, you can Telnet to a device on the same VLAN or another VLAN or subnet by using its global unicast address. To Telnet to a device having an IPv6 global unicast address of 2001:db8::215:60ff:fe79:980, enter the following command: HP Switch(config)# telnet 2001:db8::215:60ff:fe79:980

Viewing the current telnet activity on a switch Syntax: show telnet This command shows the active incoming and outgoing Telnet sessions on the switch (for both IPv4 and IPv6). Command output includes the following: Session The session number. The switch allows one outbound session and up to five inbound sessions. Privilege Manager or Operator. From Console (for outbound sessions) or the source IP address of the inbound session. To The destination of the outbound session, if in use. “show telnet output with three sessions active” (page 43) shows that the switch is running one outbound IPv4 session and is being accessed by two inbound sessions.

42

IPv6 Management Features

Example 20 show telnet output with three sessions active HP Switch# show telnet Telnet Activity ---------------------------------------------------Session : 1 Privilege: Manager From : Console To : 10.0.10.140 ---------------------------------------------------Session : 2 Privilege: Manager From : 2620:0:260:212::2:219 To : ---------------------------------------------------Session : ** 3 1 Privilege: Manager From : fe80::2:101 To : 1

The ** in “Session:” indicates the session through which show telnet was run.

Enabling or disabling inbound Telnet access Syntax: [no] telnet-server listen [ oobm | data | both ] This command is used at the global config level to enable (the default) or disable all (IPv4 and IPv6) inbound Telnet access to the switch. The no form of the command disables inbound telnet. The listen parameter is available only on switches that have a separate OOBM port. Values for this parameter are: oobm Inbound Telnet access is enabled only on the OOBM port. data Inbound Telnet access is enabled only on the data ports. both Inbound Telnet access is enabled on both the OOBM port and on the data ports. This is the default value. For more information on OOBM, see the "Network Out-of-Band Management" Appendix in the Management and Configuration Guide. The listen parameter is not available on switches that do not have a separate OOBM port. Example 21 To disable IPv4 and IPv6 Telnet access to the switch, use this command: HP Switch(config)# no telnet-server

Viewing the current inbound Telnet configuration Syntax: show console IPv6 Telnet operation

43

This command shows the current configuration of IPv4 and IPv6 inbound Telnet permissions, as well as other information. For both protocols, the default setting allows inbound sessions. Example 22 show console output Showing default console configuration HP Switch(config)# show console Console/Serial Link Inbound Telnet Enabled [Yes] : Yes 1 Web Agent Enabled [Yes] : Yes Terminal Type [VT100] : VT100 Screen Refresh Interval (sec) [3] : 3 Displayed Events [All] : All Baud Rate [Speed Sense] : speed-sense Flow Control [XON/XOFF] : XON/XOFF Session Inactivity Time (min) [0] : 0 1

Inbound Telnet Setting for IPv4 and IPv6 Telnet is Yes

Configuring an IPv6 address for an SNTP server For more information, see “SNTP and Timep” (page 56).

Syntax: [no] sntp server priority 1 - 3 link-local-addr %vlan 1 - 3 [ oobm ] [ 1 - 7 ] [no] sntp server priority 1 - 3 globalunicast-addr [ oobm ] [ 1 - 7 ] Configures an IPv6 address for an SNTP server. server priority 1 - 3 Specifies the priority of the server addressing being configured. When the SNTP mode is set to unicast and more than one server is configured, this value determines the order in which the configured servers will be accessed for a time value. The switch polls multiple servers in order until a response is received or until all servers on the list have been tried without success. Up to three server addresses (IPv6 and/or IPv4) can be configured. link-local-addr Specifies the link-local IPv6 address of the destination device. %vlan vid Suffix specifying the interface on which the destination device is located. No spaces are allowed in the suffix. global-unicast-addr Specifies the global IPv6 address of the destination device. oobm For switches that have a separate OOBM port, oobm specifies that SNTP traffic goes through that port. (By default, SNTP traffic goes through the data ports.) [ 1 - 7 ] This optional setting specifies the SNTP server version expected for the specified server. Default: 3

44

IPv6 Management Features

Example 23 Configuring link-local and global unicast SNTP server addresses To configure link-local and global unicast SNTP server addresses of: •

fe80::215:60ff:fe7a:adc0 (on VLAN 10, configured on the switch)



2001:db8::215:60ff:fe79:8980

as the priority "1" and "2" SNTP servers, respectively, using version 7, you would enter these commands at the global config level, as shown below. HP Switch(config)# sntp server priority 1 fe80::215:60ff:fe7a:adc0%vlan10 7 HP Switch(config)# sntp server priority 2 2001:db8::215:60ff:fe79:8980 7

NOTE: In the preceding example, using a link-local address requires that you specify the local scope for the address; VLAN 10 in this case. This is always indicated by %vlan followed immediately (without spaces) by the VLAN identifier.

Syntax: show sntp Displays the current SNTP configuration, including the following: Time Sync Mode Indicates whether timesync is disabled or set to either SNTP or Timep. Default: timep SNTP Mode Indicates whether SNTP uses the broadcast or unicast method of contacting a time server. The broadcast option does not require you to configure a time server address. The unicast option does require configuration of a time server address. Poll Interval Indicates the interval between consecutive time requests to an SNTP server. Priority Indicates the configured priority for the corresponding SNTP server address. SNTP Server Address Lists the currently configured SNTP server addresses. Protocol Version Lists the SNTP server protocol version to expect from the server at the corresponding address.

Configuring an IPv6 address for an SNTP server

45

Example 24 show sntp output with both an IPv6 and an IPv4 server address configured The show sntp output for the proceeding sntp server command example would appear as follows: HP Switch(config)# show sntp SNTP Configuration Time Sync Mode: Sntp SNTP Mode : Broadcast Poll Interval (sec) [720] : 719 Priority -------1 2

NOTE:

SNTP Server Address ----------------------------------2001:db8::215:60ff:fe79:8980 10.255.5.24

Protocol Version ---------------7 3

The show management command can also be used to display SNTP server information.

Configuring (enabling or disabling) the Timep mode Software release K.13.01 and greater enables configuration of a global unicast address for IPv6 Timep time server. For the details of configuring Timep on the switch, see the "Time Protocols" chapter in the Management and Configuration Guide for your switch. This section lists the Timep and related commands, including an example of using an IPv6 address. The following commands are available at the global config level for Timep operation. Commands Affecting Timep

Function

show timep

Display the current timep configuration.

timesync [

sntp

ip timep dhcp [ 1 - 9999

|

]

timep

interval

]

Enable either SNTP or Timep as the time synchronization method on the switch without affecting the configuration of either. Enable Timep operation with a Timep server assignment configured from an IPv4 or IPv6 DHCP server. Optionally change the interval between time requests.

ip timep manual ipv6-addr interval 1 - 9999

Enable Timep operation with a statically configured IPv6 address for a Timep server. Optionally change the interval between time requests.

[no] ip timep

Disables Timep operation. To re-enable Timep, it is necessary to reconfigure either the DHCP or the static option.

NOTE: To use a global unicast IPv6 address to configure an IPv6 Timep server on the switch, the switch must be receiving advertisements from an IPv6 router on a VLAN configured on the switch. To use a link-local IPv6 address to configure an IPv6 Timep server on the switch, it is necessary to append %vlan followed (without spaces) by the VLAN ID of the VLAN on which the server address is available. The VLAN must be configured on the switch. For example: fe80::11:215%vlan10

46

IPv6 Management Features

Syntax: ip timep dhcp interval 1 - 9999 ip timep manual { ipv6-addr | ipv4-addr } [ interval 1 - 9999 ] [ oobm ] Used at the global config level to configure a Timep server address. NOTE:

The switch allows one Timep server configuration.

timep dhcp Configures the switch to obtain the address of a Timep server from an IPv4 or IPv6 DHCP server. timep manual Specifies static configuration of a Timep server address. ipv6-addr Specifies the IPv6 address of an SNTP server. See the preceding Note. [ interval 1 - 9999 ] This optional setting specifies the interval in minutes between Timep requests. Default: 720 [ oobm ] For switches that have a separate OOBM port, oobm specifies that Timep traffic goes through that port. (By default, Timep traffic goes through the data ports.) Example 25 Configuring a link-local Timep server address To configure a link-local Timep server address of: fe80::215:60ff:fe7a:adc0 where the address is on VLAN 10, configured on the switch, enter this command at the global config level: HP Switch(config)# ip timep manual fe80::215:60ff:fe7a:adc0%vlan10

NOTE: In the preceding example, using a link-local address requires that you specify the local scope for the address; VLAN 10 in this case. This is always indicated by %vlan followed immediately (without spaces) by the VLAN identifier. For a global unicast address, you would enter the address without the %vlan suffix.

Syntax: show timep Displays the current Timep configuration, including the following: Time Sync Mode Indicates whether timesync is disabled or set to either SNTP or Timep. Default: Disabled Timep Mode Indicates whether Timep is configured to use a DHCP server to acquire a Timep server address or to use a statically configured Timep server address. Server Address Lists the currently configured Timep server address.

Configuring an IPv6 address for an SNTP server

47

Poll Interval (min) [ 720 ] Indicates the interval between consecutive time requests to the configured Timep server. Example 26 show timep The show timep output for the preceding ip timep manual command example would appear as follows: HP Switch(config)# show timep Timep Configuration Time Sync Mode: Timep TimeP Mode [Disabled] : Manual Server Address : fe80::215:60ff:fe7a:adc0%vlan10 Poll Interval (min) [720] : 720

Transferring TFTP files over IPv6 You can use TFTP copy commands over IPv6 to upload, or download files to and from a physically connected device or a remote TFTP server, including: •

Switch software



Software images



Switch configurations



ACL command files



Diagnostic data (crash data, crash log, and event log)

For complete information on how to configure TFTP file transfers between the switch and a TFTP server or other host device on the network, see the File Transfers appendix in the Management and Configuration Guide for your switch. To upload and/or download files to the switch using TFTP in an IPv6 network, you must: 1. Enable TFTP for IPv6 on the switch; see “Enabling TFTP for IPv6” (page 48). 2. Enter a TFTP copy command with the IPv6 address of a TFTP server in the command syntax; see “Copying files over IPv6 using TFTP” (page 49). 3. (Optional) To enable auto-TFTP operation, enter the auto-tftp command; see “Using auto-TFTP for IPv6” (page 51).

Enabling TFTP for IPv6 Client and server TFTP for IPv6 is enabled by default on the switch. However, if it is disabled, you can re-enable it by specifying TFTP client or server functionality with the tftp client|server command. Enter this command at the global configuration level.

Syntax: [no] tftp [ client ] | server [ listen oobm | data | both ] Enables TFTP for IPv4 and IPv6 client or server functionality so that the switch can:

48



Use TFTP client functionality to access IPv4- or IPv6-based TFTP servers in the network to receive downloaded files.



Use TFTP server functionality on the switch to be accessed by other IPv4 or IPv6 hosts requesting to upload files.



For switches that have a separate OOBM port, the listen parameter in a server configuration allows you to specify whether transfers take place through the OOBM interface, the data interface, or both. For more information

IPv6 Management Features

on OOBM, see Appendix I, "Networked Out-of-Band Management (OOBM)" in this guide. The no form of the command disables the client or server functionality. Default: TFTP client and server functionality enabled NOTE: To disable all TFTP client or server operation on the switch except for the auto-TFTP feature, enter the no tftp [ client | server ] command. To re-enable TFTP client or server operation, re-enter the tftp [ client | server ] command. (Entering no tftp without specifying client or server affects only the client functionality. To disable or re-enable the TFTP server functionality, you must specify server in the command.) When TFTP is disabled, instances of TFTP in the CLI copy command and the Menu interface "Download OS" screen become unavailable. The no tftp [ client|server ] command does not affect auto-TFTP operation. For more information, see “Using auto-TFTP for IPv6” (page 51).

Copying files over IPv6 using TFTP Syntax: copy tftp target ipv6-addr filename [ oobm ] Copies (downloads) a data file from a TFTP server at the specified IPv6 address to a target file on a switch that is enabled with TFTP server functionality. ipv6-addr If this is a link-local address, use this IPv6 address format: fe80::device-id %vlan vid For example: fe80::123%vlan10 If this is a global unicast address, use this IPv6 format: ipv6-addr For example: 2001:db8::123 target One of the following values: autorun-cert-file Copies an autorun trusted certificate to the switch. autorun-key-file Copies an autorun key file to the switch. command-file Copies a file stored on a remote host and executes the ACL command script on the switch. Depending on the ACL commands stored in the file, one of the following actions is performed in the running-config file on the switch: •

A new ACL is created.



An existing ACL is replaced.



match, permit, or deny statements are added to an existing ACL. For more information on ACLs, see "Creating an ACL Offline" in the “Access Control Lists (ACLs)” chapter in the Access Security Guide. Transferring TFTP files over IPv6

49

config filename Copies the contents of a file on a remote host to a configuration file on the switch. flash [ primary | secondary ] Copies a software file stored on a remote host to primary or secondary flash memory on the switch. To run a newly downloaded software image, enter the reload or boot system flash command. pub-key-file Copies a public-key file to the switch. startup-config Copies a configuration file on a remote host to the startup configuration file on the switch. oobm For switches that have a separate OOBM port, specifies that the transfer will be through the OOBM interface. (Default is transfer through the data interface.)

Syntax: copy source tftp ipv6-addr filename [ pc | unix ] [ oobm ] Copies (uploads) a source data file on a switch that is enabled with TFTP server functionality to a file on the TFTP server at the specified IPv6 address, where source is one of the following values: command-output cli-command Copies the output of a CLI command to the specified file on a remote host. config filename Copies the specified configuration file to a remote file on a TFTP server. crash-data [ slot-id | master ] Copies the contents of the crash data file to the specified file path on a remote host. The crash data is software-specific and used to determine the cause of a system crash. You can copy crash information from an individual slot or from the master crash file on the switch. crash-log [ slot-id | master ] Copies the contents of the crash log to the specified file path on a remote host. The crash log contains processor-specific operational data that is used to determine the cause of a system crash. You can copy the contents of the crash log from an individual slot or from the master crash log on the switch. event-log Copies the contents of the Event Log on the switch to the specified file path on a remote host. flash [ primary | secondary ] Copies the software file used as the primary or secondary flash image on the switch to a file on a remote host. startup-config Copies the startup configuration file in flash memory to a remote file on a TFTP server. running-config Copies the running configuration file to a remote file on a TFTP server.

50

IPv6 Management Features

ipv6-addr If this is a link-local address, use this IPv6 address format: fe80::device-id%vlan vid For example: fe80::123%vlan10 If this is a global unicast address, use this IPv6 format: ipv6-addr For example: 2001:db8::123 oobm For switches that have a separate OOBM port, specifies that the transfer will be through the OOBM interface. (Default is transfer through the data interface.)

Using auto-TFTP for IPv6 At switch startup, the auto-TFTP for IPv6 feature automatically downloads a software image to the switch from a specified TFTP server and then reboots the switch. To implement the process, the switch must first reboot using one of the following methods: •

Enter the boot system flash primary command in the CLI.



With the default flash boot image set to primary flash (the default), enter the boot or the reload command, or cycle the power to the switch. (To reset the boot image to primary flash, use boot set-default flash primary.)

Syntax: auto-tftp ipv6-addr filename Configures the switch to automatically download the specified software file from the TFTP server at the specified IPv6 address. The file is downloaded into primary flash memory at switch startup. The switch then automatically reboots from primary flash. NOTE: To enable auto-TFTP to copy a software image to primary flash memory, the version number of the downloaded software file (for example, K_14_01.swi) must be different from the version number currently in the primary flash image. The current TFTP client status (enabled or disabled) does not affect auto-TFTP operation. See “Enabling TFTP for IPv6” (page 48). Completion of the auto-TFTP process may require several minutes while the switch executes the TFTP transfer to primary flash and then reboots again. The no form of the command disables auto-TFTP operation by deleting the auto-tftp entry from the startup configuration. The no auto-tftp command does not affect the current TFTP-enabled configuration on the switch. However, entering the ip ssh filetransfer command automatically disables both auto-tftp and tftp operation.

SNMP Configuration Commands Supported For more information on each SNMP configuration procedure, see the "Configuring for Network Management Applications" chapter in the current Management and Configuration Guide for your switch.

SNMP Configuration Commands Supported

51

SNMPv1 and v2c Syntax: snmp-server host [ ipv6-addr | ipv4-addr ] [ community-name ] [ none | all | non-info | critical | debug ] [ inform [ retries count ] [ timeout interval ]] Executed at the global config level to configure an SNMP trap receiver to receive SNMPv1 and SNMPv2c traps, SNMPv2c informs, and (optionally) Event Log messages. snmp-server listen [ oobm | data | both ] For switches with a separate OOBM port, specifies whether the switch listens for SNMP traps on the OOBM interface, the data interface, or both.

SNMPv3 Syntax: snmpv3 targetaddress name params params_name [ ipv4-addr | ipv6-addr ] [ addr-mask ip4-addr ] [ filter none | debug | all | not-info | critical ] [ max-msg-size 484-65535 ] [ port-mask tcp-udp port ] [ retries 0 - 255 ] [ taglist tag_name ] [ timeout 0 - 2147483647 ] [ udp-port port-number ] Executed at the global config level to configure an SNMPv3 management station to which notifications (traps and informs) are sent. IPv6 addresses are supported in SNMP show command output, as shown in Example 27 (page 53) and Example 28 (page 54). The show snmp-server command displays the current SNMP policy configuration, including SNMP communities, network security notifications, link-change traps, trap receivers (including the IPv4 or IPv6 address) that can receive SNMPv1 and SNMPv2c traps, and the source IP (interface) address used in IP headers when sending SNMP notifications (traps and informs) or responses to SNMP requests.

52

IPv6 Management Features

Example 27 show snmp-server command output with IPv6 address HP Switch(config)# show snmp-server SNMP Communities Community Name -------------------public marker

MIB View -------Manager Manager

Write Access -----------Unrestricted Unrestricted

Trap Receivers Link-Change Traps Enabled on Ports [All] : All Traps Category ---------------------------SNMP Authentication Password change Login failures Port-Security Authorization Server Contact DHCP-Snooping Dynamic ARP Protection Address --------------------15.29.17.218 15.29.17.219 2620:0000:0260:0211 :0217:a4ff:feff:1f70 1

: : : : : : :

Community --------public public marker

Current Status --------------Extended Enabled Enabled Enabled Enabled Enabled Enabled Events -------All Critical

Type ---trap trap

Retry ----3 3

Critical trap 3

Timeout ------15 15 15

Excluded MIBs Snmp Response Pdu Source-IP Information Selection Policy : rfc1517

1

An IPv6 address is displayed on two lines.

The show snmpv3 targetaddress command displays the configuration (including the IPv4 or IPv6 address) of the SNMPv3 management stations to which notification messages are sent.

SNMP Configuration Commands Supported

53

Example 28 snmpv3 targetaddress command output with IPv6 address HP Switch(config)# show snmpv3 targetaddress snmpTargetAddrTable [rfc2573] Target Name ---------------1 2 PP.217 PP.218

1

IP Address ----------------------15.29.17.218 15.29.17.219 15.29.17.217 2620:0:260:211 :217:a4ff:feff:1f70 1

Parameter ---------1 2 marker_p marker_p

An IPv6 Address is displayed on two lines

IP preserve for IPv6 IPv6 supports the IP preserve feature, which allows you to copy a configuration file from a TFTP server to multiple switches without overwriting the IPv6 address and subnet mask on VLAN 1 (the default VLAN) in each switch, and the Gateway IPv6 address assigned to the switch.

Configuring IP preserve Enter the ip preserve statement at the end of the configuration file to be downloaded from a TFTP server. (You do not invoke IP preserve by entering a command from the CLI.) Example 29 How to enter IP preserve in a configuration file ; J8697A Configuration Editor; Created on release #K.15.xx hostname "HP Switch" time daylight-time-rule None * * * * * * password manager password operator ip preserve

Entering an ip preserve statement as the last line in a configuration file stored on a TFTP server allows you to download and execute the file as the startup-config file on an IPv6 switch. When the switch reboots, the configuration settings in the downloaded file are implemented without changing the IPv6 address and gateway assigned to the switch as shown in Example 30 (page 55).

Downloading an IP preserve configuration file to an IPv6-based switch Enter the TFTP copy command, as described in “SNMP management for IPv6” (page 57), to copy the file as the new startup-config file on a switch. When you download an IP Preserve configuration file, the following rules apply: •

54

If the switch's current IPv6 address for VLAN 1 was statically configured and not dynamically assigned by a DHCP/Bootp server, the switch reboots and retains its current IPv6 address,

IPv6 Management Features

subnet mask, and gateway address. All other configuration settings in the downloaded configuration file are applied. •

If the switch's current IPv6 address for VLAN 1 was assigned from a DHCP server and not statically configured, IP preserve is suspended. The IPv6 addressing specified in the downloaded configuration file is implemented when the switch copies the file and reboots. •

If the downloaded file specifies DHCP/Bootp as the source for the IPv6 address of VLAN 1, the switch uses the IPv6 address assigned by the DHCP/Bootp server.



If the file specifies a dedicated IPv6 address and subnet mask for VLAN 1 and a Gateway IPv6 address, the switch implements these settings in the startup-config file.

Verifying how IP preserve was implemented in a switch After the switch reboots, enter the show run command. Example 30 (page 55) shows an example in which all configurations settings have been copied into the startup-config file except for the IPv6 address of VLAN 1 (2001:db8::214:c2ff:fe4c:e480) and the default IPv6 gateway (2001:db8:0:7::5), which were retained. If a switch received its IPv6 address from a DHCP server, the "ip address" field under "vlan 1" would display dhcp-bootp. Example 30 Configuration file with dedicated IP addressing HP Switch(config)# show run Running configuration: ; J8715A Configuration Editor; Created on release #K.14.01 hostname "HP Switch" module 1 type J8702A module 2 type J8705A trunk 11-12 Trk1 Trunk ip default-gateway 2001:db8:0:7::5 snmp-server community "public" Unrestricted vlan 1 name "DEFAULT_VLAN" untagged 1-10,13-24,1-24,Trk1 ip address 2001:db8::214:c2ff:fe4c:e480 exit spanning-tree Trk1 priority 4 password manager password operator

NOTE: Because the switch’s IPv6 address and default gateway were statically configured (not assigned by a DHCP server), when the switch boots up with the IP Preserve startup configuration file (see “How to enter IP preserve in a configuration file” (page 54)), its current IPv6 address, subnet mask, and default gateway are not changed. If a switch’s current IP address was acquired from a DHCP/Bootp server, the IP Preserve statement is ignored and the IP ddresses in the downloaded configuration file are implemented.

For more information on how to use the IP preserve feature, see the "Configuring IP Addressing" chapter in the current Basic Operation Guide .

View the neighbor cache ND occurs when there is communication between IPv6 devices on a VLAN. A neighbor destination is reachable from a given source address if a confirmation (neighbor solicitation) has been received View the neighbor cache

55

at the source verifying that traffic has been received at the destination. The neighbor cache retains data for a given neighbor until the entry times out. You can view and clear the contents of the neighbor cache using the commands described in this section. For more on this topic, see “Neighbor discovery” (page 32).

Clear the neighbor cache When there is an event such as a topology change or an address change, the neighbor cache may have too many entries to allow efficient use. Also, if an unauthorized client is answering DAD or normal neighbor solicitations with invalid replies, the neighbor cache may contain a large number of invalid entries and communication with some valid hosts may fail, the show ipv6 neighbors command output may become too cluttered to efficiently read, or both. In such cases, the fastest way to restore optimum traffic movement on a VLAN may be to statically clear the neighbor table instead of waiting for the unwanted entries to time-out.

SNTP and Timep About configuring (enabling or disabling) the SNTP mode The switch supports both Timepv6 and SNTPv6 time services. Software release K.13.01 and greater enables configuration of a global unicast address for IPv6 SNTP time server. This section lists the SNTP and related commands, including an example of using an IPv6 address. For the details of configuring SNTP on the switch, see the "Time Protocols" chapter in the Management and Configuration Guide for your switch. The following commands are available at the global config level for SNTP operation: Commands affecting SNTP

Function

show sntp

Display the current SNTP configuration.

timesync [ sntp | timep ]

Enable either SNTP or Timep as the time synchronization method on the switch without affecting the configuration of either.

[no] timesync

Enable time synchronization. (Requires a timesync method to also be enabled.) The no version disable time synchronization without affecting the configuration of the current time synchronization method.)

[no] sntp

Enables SNTP with the current SNTP configuration. The no version disables SNTP without changing the current SNTP configuration.

sntp [ unicast | broadcast ]

Configures the SNTP mode. Default: Broadcast

sntp 30 - 720

Changes the interval between time requests. Default: 720 seconds

About configuring an IPv6 address for an SNTP server To use a global unicast IPv6 address to configure an IPv6 SNTP time server on the switch, the switch must be receiving advertisements from an IPv6 router on a VLAN configured on the switch. To use a link-local IPv6 address to configure an IPv6 SNTP time server on the switch, it is necessary to append %vlan followed immediately (without spaces) by the VLAN ID of the VLAN on which the server address is available. (The VLAN must be configured on the switch.) For example: fe80::11:215%vlan10

56

IPv6 Management Features

TFTP file transfers over IPv6 You can use TFTP copy commands over IPv6 to upload or download files to and from a physically connected device or a remote TFTP server, including: •

Switch software



Software images



Switch configurations



ACL command files



Diagnostic data (crash data, crash log, and event log)

For information on how to configure TFTP file transfers between the switch and a TFTP server or other host device on the network, see the "File Transfers" appendix in the Management and Configuration Guide for your switch.

SNMP management for IPv6 As with SNMP for IPv4, you can manage a switch via SNMP from an IPv6-based network management station by using an application such as HP PCM or HP PCM+. (For more information on PCM and PCM+, go to the HP Switch Networking web site at http:www.hp.com/networking/ support.)

SNMP features supported The same SNMP for IPv4 features are supported over IPv6: •

Access to a switch using SNMP version 1, version 2c, or version 3



Enhanced security with the configuration of SNMP communities and SNMPv3 user-specific authentication password and privacy (encryption) settings



SNMP notifications, including: •

SNMP version 1 or SNMP version 2c traps



SNMPv2c informs



SNMPv3 notification process, including traps



Advanced RMON (remote monitoring) management



HP PCM or HP PCM+ management applications



Flow sampling using sFlow



Standard MIBs, such as the Bridge MIB (RFC 1493) and the Ethernet MAU MIB (RFC 1515)

TFTP file transfers over IPv6

57

3 IPv6 Management Security Features This chapter describes management security features that are IPv6 counterparts of IPv4 management security features on the switches. Table 4 Summary of Commands Command syntax

Description

Default

CLI page reference

ipv6 authorized-managers ipv6-addr

Configure authorized IP managers for IPv6. disabled

58

[no] ip ssh filetransfer

Enabling secure copy and secure FTP for IPv6. disabled

64

This chapter describes the following IPv6-enabled management security features: •

Authorized IP Managers for IPv6



Secure Shell for IPv6



Secure Copy and Secure FTP for IPv6

Configuring authorized IP managers for switch access To configure one or more IPv6-based management stations to access the switch using the authorized IP managers feature, enter the ipv6 authorized-managers command.

Syntax: [no] ipv6 authorized-managers ipv6-addr ipv6-mask [ access [ operator | manager ] ] access-method [ all | ssh | telnet | web | snmp | tftp ] Configures one or more authorized IPv6 addresses to access the switch, where: ipv6-mask Specifies the mask that is applied to an IPv6 address to determine authorized stations. For more information, see “About using a mask to configure authorized management stations” (page 66) Default: FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF. access [ operator | manager ] Specifies the level of access privilege granted to authorized stations. Applies only to access through Telnet, SSH, and SNMP (version 1, 2, and 3). Default: Manager access-method [ all | ssh | telnet | web | snmp | tftp ] Configures access levels by access method and IP address. Each management method can have its own set of authorized managers. Default: All

Configuring single station access To authorize only one IPv6-based station for access to the switch: Enter the IPv6 address of the station and set the mask to FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF.

58

IPv6 Management Security Features

NOTE: If you do not enter a value for the ipv6-mask parameter when you configure an authorized IPv6 address, the switch automatically uses FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF as the default mask. If you have ten or fewer management and/or operator stations for which you want to authorize access to the switch, it may be more efficient to configure them by entering each IPv6 address with the default mask in a separate ipv6 authorized-managers command. When used in a mask, "FFFF" specifies that each bit in the corresponding 16-bit (hexadecimal) block of an authorized station's IPv6 address must be identical to the same "on" or "off" setting in the IPv6 address entered in the ipv6 authorized-managers command. (The binary equivalent of FFFF is 1111 1111 1111 1111, where 1 requires the same "on" or "off" setting in an authorized address.) Example 31 Configuring single station access As shown in Table 5 (page 59), if you configure a link-local IPv6 address of FE80::202:B3FF:FE1E:8329 with a mask of FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF, only a station having an IPv6 address of FE80::202:B3FF:FE1E:8329 has management access to the switch. Table 5 Mask for configuring a single authorized IPv6 manager station 1st block

2nd block

3rd block

4th block

5th block

6th block

7th block

8th block

Manager- or operator-level access

IPv6 mask

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

IPv6 address

FE80

0000

0000

0000

202

B3FF

FE1E

8329

The “FFFF” in each hexadecimal block of the mask specifies that only the exact value of each bit in the corresponding block of the IPv6 address is allowed. This mask allows management access only to a station having an IPv6 address of FE80::202:B3FF:FE1E:8329.

Configuring multiple station access To authorize multiple stations to access the switch without having to re–enter the ipv6 authorized-managers command for each station, carefully select the IPv6 address of an authorized IPv6 manager and an associated mask to authorize a range of IPv6 addresses. For more information on this topic, see “About configuring multiple station access” (page 66).

Viewing an authorized IP managers configuration Use the show ipv6 authorized-managers command to list the IPv6 stations authorized to access the switch.

Configuring authorized IP managers for switch access

59

Example 32 show ipv6 authorized-managers HP Switch# show ipv6 authorized-managers IPv6 Authorized Managers --------------------------------------Address : 2001:db8:0:7::5 Mask : ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff Access : Manager Address : 2001:db8::a:1c:e3:3 Mask : ffff:ffff:ffff:ffff:ffff:ffff:ffff:fffe Access : Manager Address : 2001:db8::214:c2ff:fe4c:e480 Mask : ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff Access : Manager Address : 2001:db8::10 Mask : ffff:ffff:ffff:ffff:ffff:ffff:ffff:ff00 Access : Operator

By analyzing the masks displayed in Example 32 (page 60), the IPv6 stations shown in Table 6 (page 60) are granted access. Table 6 How masks determine authorized IPv6 manager addresses

Mask

Authorized IPv6 addresses

Number of authorized addresses

FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFC

2001:db8:0:7::4 through 2001:db8:0:7::7

4

FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFE

2001:db8::a:1c:e3:2 and 2001:db8::a:1c:e3:3

2

FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF

2001:db8::214:c2ff:fe4c:e480

1

FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FF00

2001:db8::0 through 2001:db8::FF

256

Authorizing manager access The following IPv6 commands authorize manager-level access for one link-local station at a time. When you enter a link-local IPv6 address with the ipv6 authorized-managers command, you must also enter a VLAN ID in the format: %vlanvlan-id. HP Switch(config)# ipv6 authorized-managers fe80::07be:44ff:fec5:c965%vlan2 HP Switch(config)# ipv6 authorized-managers fe80::070a:294ff:fea4:733d%vlan2 HP Switch(config)# ipv6 authorized-managers fe80::19af:2cff:fe34:b04a%vlan5 If you do not enter an ipv6-mask value when you configure an authorized IPv6 address, the switch automatically uses FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF as the default IPv6 mask. Also, if you do not specify an access value to grant either manager- or operator-level access, by default, the switch assigns manager access.

60

IPv6 Management Security Features

Example 33 Default IPv6 mask HP Switch# ipv6 authorized-managers 2001:db8::a8:1c:e3:69 HP Switch# show ipv6 authorized-managers IPv6 Authorized Managers -------------------------Address : 2001:db8::a8:1c:e3:69 Mask : ffff:ffff:ffff:ffff:ffff:ffff:ffff:ffff Access : Manager

NOTE: If you do not enter a value for ipv6-mask in the ipv6 authorized-managers command, the default mask of FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF: is applied. The default mask authorizes only the specified station (see “Configuring single station access” (page 58)). The next IPv6 command authorizes operator-level access for sixty-four IPv6 stations: thirty-two stations in the subnets defined by 0x0006 and 0x0007 in the fourth block of an authorized IPv6 address: HP Switch(config)# ipv6 authorized-managers 2001:db8:0000:0007:231:17ff:fec5:c967 ffff:ffff:ffff:fffe:ffff:ffff:ffff:ffe0 access operator The following ipv6 authorized-managers command authorizes a single, automatically generated (EUI-64) IPv6 address with manager-level access privilege: HP Switch(config)# ipv6 authorized-managers ::223:04ff:fe03:4501 ::ffff:ffff:ffff:ffff

Editing an existing authorized IP manager entry To change the mask or access level for an existing authorized IP manager entry, enter the IPv6 address with the new values. Any parameters not included in the command are reset to their default values. Example 34 Editing an existing authorized IP manager entry The following command replaces the existing mask and access level for IPv6 address 2001:DB8::231:17FF:FEC5:C967 with FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FF00 and operator: HP Switch(config)# ipv6 authorized-managers 2001:db8::231:17ff:fec5:c967 ffff:ffff:ffff:ffff:ffff:ffff:ffff:ff00 access operator The following command replaces the existing mask and access level for IPv6 address 2001:DB8::231:17FF:FEC5:3E61 with FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF and manager (the default values). It is not necessary to enter either of these parameters: HP Switch(config)# ipv6 authorized-managers 2001:db8::a05b:17ff:fec5:3f61

Deleting an authorized IP manager entry Enter only the IPv6 address of the configured authorized IP manager station that you want to delete with the no form of the command.

Configuring authorized IP managers for switch access

61

Example 35 Deleting an authorized IP manager entry HP Switch(config)# no ipv6 authorized-managers 2001:db8::231:17ff:fec5:3e61

Configuring SSH for IPv6 For more information on SSH configuration, see “Secure shell (SSH) for IPv6” (page 69). By default, SSH is automatically enabled for IPv4 and IPv6 connections on a switch. Use the ip ssh command options to reconfigure the default SSH settings used in SSH authentication for IPv4 and IPv6 connections: •

TCP port number



timeout period



file transfer



MAC type



cipher type



listening port(s)

Syntax: [no] ip ssh Enables SSH for on the switch for both IPv4 and IPv6, and activates the connection with a configured SSH server (RADIUS or TACACS+). The no form of the command disables SSH on the switch. cipher cipher-type Specify a cipher type to use for connection. Valid types are: •

aes128–cbc



3des-cbc



aes192–cbc



aes256–cbc



[email protected]



aes128–ctr



aes192–ctr



aes256–ctr

Default: All cipher types are available. Use the no form of the command to disable a cipher type. filetransfer Enables SSH on the switch to connect to an SCP or SFTP client application to transfer files to and from the switch over IPv4 or IPv6. Default: Disabled NOTE: Enabling filetransfer automatically disables TFTP client and TFTP server functionality. For more information, see “Performing secure file transfers to and from IPv4 and IPv6 client devices” (page 64).

62

IPv6 Management Security Features

mac MAC-type Allows configuration of the set of MACs that can be selected. Valid types are: •

hmac-md5



hmac-sha1



hmac-sha1–96



hmac-md5–96

Default: All MAC types are available. Use the no form of the command to disable a MAC type. port [ 1 - 65535 | default ] TCP port number used for SSH sessions in IPv4 and IPv6 connections Default: 22. Valid port numbers are from 1 to 65535, except for port numbers 23, 49, 80, 280, 443, 1506, 1513, and 9999, which are reserved for other subsystems. public-key [ manager | operator ] keystring Store a client-generated key for public-key authentication. manager Allows manager-level access using SSH public-key authentication. operator Allows operator-level access using SSH public-key authentication. keystring A legal SSHv2 (RSA or DSA) public key. The text string for the public key must be a single-quoted token. If the keystring contains double quotes, it can be quoted with single quotes ('key-string'). The following restrictions for a keystring apply:•

A keystring cannot contain both single and double quotes.



A keystring cannot have extra characters, such as a blank space or a new line. (To improve readability, you can add a backlash at the end of each line.)

For more information on configuring and using SSH public keys to authenticate SSH clients connecting to the switch, see chapter "Configuring Secure Shell" in the latest Access Security Guide for your switch. timeout 5 - 120 Timeout value allowed to complete an SSH authentication and login on the switch. Default: 120 seconds. listen [ oobm | data | both ] The listen parameter is available only on switches that have a separate OOBM port. Values for this parameter are: oobm Inbound SSH access is enabled only on the OOBM port. data Inbound SSH access is enabled only on the data ports. both Inbound SSH access is enabled on both the OOBM port and on the data ports. This is the default value. Configuring SSH for IPv6

63

For more information on OOBM, see the "Network Out-of-Band Management" Appendix in the Management and Configuration Guide. The listen parameter is not available on switches that do not have a separate OOBM port. NOTE: For both IPv4 and IPv6, the switch supports only SSH version 2. You cannot set up an SSH session with a client device running SSH version 1. For more information on how to configure SSH for encrypted, authenticated transactions between the switch and SSH-enabled client devices, see the "Configuring Secure Shell (SSH)" chapter in the latest Access Security Guide for your switch.

Displaying an SSH configuration To verify an SSH configuration and display all SSH sessions running on the switch, enter the show ip ssh command. Information on all current SSH sessions (IPv4 and IPv6) is displayed. Example 36 SSH configuration display With SSH running, the switch supports one console session and up to five other SSH and Telnet (IPv4 and IPv6) sessions. WebAgent sessions are also supported, but are not displayed in show ip ssh output. Source IPv6 IP addresses of SSH clients are displayed in hexadecimal format. HP Switch# show ip ssh SSH Enabled : Yes TCP Port Number : 22 Host Key Type : RSA

Secure Copy Enabled : No Timeout (sec) : 120 Host Key Size : 2048

Ciphers : aes128-cbc,3des-cbc,aes192-cbc,aes256-cbc, [email protected],aes128-ctr,aes192-ctr, aes256-ctr MACs : hmac-md5,hmac-sha1,hmac-sha1-96,hmac-md5-96 Ses --1 2 3 4 5 6

Type -------console ssh inactive inactive inactive inactive

| Source IP Port + ---------------------------------------- ----| | 10.168.31.114 1722 | | | |

Displays the current SSH configuration and status. The switch uses these five SSH settings internally for transactions with clients.

Performing secure file transfers to and from IPv4 and IPv6 client devices For more information, see “SCP and SFTP for IPv6” (page 70).

Syntax: [no] ip ssh filetransfer Enables SSH on the switch to connect to an SCP or SFTP client application to transfer files to and from the switch. Use the no ip ssh filetransfer command to disable the switch's ability to perform secure file transfers with an SCP or SFTP client, without disabling SSH on the switch. 64

IPv6 Management Security Features

Authorized IP managers for IPv6 The authorized IP managers feature uses IP addresses and masks to determine which stations (PCs or workstations) can access the switch through the network. This feature supports switch access through: •

Telnet and other terminal emulation applications



SNMP (with a correct community name)



SSH



TFTP

As with the configuration of IPv4 management stations, the authorized IP managers for IPv6 feature allows you to specify the IPv6-based stations that can access the switch. •







You can configure up to 100 authorized IPv4 and IPv6 manager addresses on a switch, where each address applies to either a single management station or a group of stations. Each authorized manager address consists of an IPv4 or IPv6 address and a mask that determines the individual management stations that are allowed access.



You configure authorized IPv4 manager addresses using the ip authorized-managers command. For more information, see "Using Authorized IP Managers" in the Access Security Guide.



You configure authorized IPv6 manager addresses using the ipv6 authorized-managers command. For more information, see “Configuring authorized IP managers for switch access” (page 58).

You can block all IPv4-based or all IPv6-based management stations from accessing the switch by entering the following commands:



To block access to all IPv4 manager addresses while allowing access to IPv6 manager addresses, enter the ip authorized-managers 0.0.0.0 command.



To block access to all IPv6 manager addresses while allowing access to IPv4 manager addresses, enter the ipv6 authorized-managers :: command. (The double colon represents an IPv6 address that consists of all zeros: 0:0:0:0:0:0:0:0.)

You configure each authorized manager address with manager- or operator-level privilege to access the switch. •

Manager privilege allows full access to all console interface screens for viewing, configuring, and all other operations available in these interfaces.



Operator privilege allows read-only access from the console interfaces.

When you configure station access to the switch using the authorized IP managers feature, the settings take precedence over the access configured with local passwords, TACACS+ servers, RADIUS-assigned settings, port-based (802.1X) authentication, and port security settings.

As a result, the IPv6 address of a networked management device must be configured with the authorized IP managers feature before the switch can authenticate the device using the configured settings from other access security features. If the authorized IP managers feature disallows access to the device, access is denied. Therefore, with authorized IP managers configured, logging in with the correct passwords is not sufficient to access a switch through the network unless the station requesting access is also authorized in the switch's authorized IP managers configuration.

Authorized IP managers for IPv6

65

About using a mask to configure authorized management stations The ipv6-mask parameter controls how the switch uses an IPv6 address to determine the IPv6 addresses of authorized manager stations on your network. For example, you can specify a mask that authorizes: •

Single station access



Multiple station access

NOTE: Mask configuration is a method for determining the valid IPv6 addresses that are authorized for management access to the switch. In the authorized IP managers feature, the mask serves a different purpose than an IPv6 subnet mask and is applied in a different manner.

About configuring multiple station access As shown in Table 7 (page 66), if a bit in any of the 4-bit binary representations of a hexadecimal value in a mask is "on" (set to 1), the corresponding bit in the IPv6 address of an authorized station must match the "on" or "off" setting of the same bit in the IPv6 address you enter with the ipv6 authorized-managers command. Conversely, in a mask, a "0" binary bit means that either the "on" or "off" setting of the corresponding IPv6 bit in an authorized address is valid and does not have to match the setting of the same bit in the specified IPv6 address. Table 7 (page 66) shows the binary expressions represented by individual hexadecimal values in an ipv6-mask parameter. Table 7 Hexadecimal mask values and binary equivalents

66

Hexadecimal value in an IPv6 mask

Binary equivalent

0

0000

1

0001

2

0010

3

0011

4

0100

5

0101

6

0110

7

0111

8

1000

9

1001

A

1010

B

1011

C

1100

D

1101

E

1110

F

1111

IPv6 Management Security Features

Example 37 Configuring multiple station access Table 8 (page 67) shows an example in which a mask that authorizes switch access to four management stations is applied to the IPv6 address: 2001:DB8:0000:0000:244:17FF:FEB6:D37D. The mask is: FFFF:FFFF:FFFF:FFF8:FFFF:FFFF:FFFF:FFFC. Table 8 Mask for configuring a single authorized IPv6 manager station 1st block

2nd block

3rd block

4th block

5th block

6th block

7th block

8th block

Manager- or operator-level access

IPv6 mask

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

IPv6 address

2001

DB8

0000

0000

244

17FF

FEB6

D37D

The "F" value in the first 124 bits of the mask specifies that only the exact value of each corresponding bit in an authorized IPv6 address is allowed. However, the "C" value in the last four bits of the mask allows four possible combinations (D37C, D37D, D37E, and D37F) in the last block of an authorized IPv6 address.

As shown in Table 9 (page 67), if you use a mask of FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFF:FFFC with an IPv6 address, you can authorize four IPv6-based stations to access the switch. In this mask, all bits except the last two are set to 1 (“on”); the binary equivalent of hexadecimal C is 1100. Table 9 How a mask determines four authorized IPv6 manager addresses Last block in mask: FFFC Last block in IPv6 address: D37D Bit numbers

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

Bit value

F

F

F

C

FFFC: Last block in mask

1

1

1

1

1

1

1

1

1

1

1

1

1

1

0

0

D37D:Last block in IPv6 address

1

1

0

0

1

0

0

1

1

0

1

1

1

1

0

1

Bit setting:

1 = On

0 = Off

Therefore, this mask requires the first corresponding 126 bits in an authorized IPv6 address to be the same as in the specified IPv6 address: 2001:DB8:0000:0000:244:17FF:FEB6:D37C. However, the last two bits are set to 0 ("off") and allow the corresponding bits in an authorized IPv6 address to be either "on" or "off". As a result, only the four IPv6 addresses shown in Table 10 (page 67) are allowed access. Table 10 Mask for configuring a single authorized IPv6 manager station

IPv6 mask

1st block

2nd block

3rd block

4th block

5th block

6th block

7th block

8th block

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

DB8

0000

0000

244

17FF

FEB6

D37D

IPv6 address entered with the ipv6 authorized-managers command 2001

About using a mask to configure authorized management stations

67

Table 10 Mask for configuring a single authorized IPv6 manager station (continued)

Other authorized IPv6 addresses

1st block

2nd block

3rd block

4th block

5th block

6th block

7th block

8th block

2001

DB8

0000

0000

244

17FF

FEB6

D37C

2001

DB8

0000

0000

244

17FF

FEB6

D37E

2001

DB8

0000

0000

244

17FF

FEB6

D37F

Table 5 (page 59) shows an example in which a mask is applied to the IPv6 address: 2001:DB8:0000:0000:244:17FF:FEB6:D37D/64. The specified mask FFFF:FFFF:FFFF:FFF8:FFFF:FFFF:FFFF:FFFF configures eight management stations as authorized IP manager stations. In this example, the IPv6 mask is applied as follows: •

Eight management stations in different subnets are authorized by the value of the fourth block (FFF8) in the 64-bit prefix ID (FFFF:FFFF:FFFF:FFF8) of the mask. (The fourth block of the prefix ID is often used to define subnets in an IPv6 network.) The binary equivalent of FFF8 that is used to specify valid subnet IDs in the IPv6 addresses of authorized stations is 1111 1111 1111 1000. The three "off" bits (1000) in the last part of the this block (FFF8) of the mask allow for eight possible authorized IPv6 stations: 2001:DB8:0000:0000:244:17FF:FEB6:D37D 2001:DB8:0000:0001:244:17FF:FEB6:D37D 2001:DB8:0000:0002:244:17FF:FEB6:D37D 2001:DB8:0000:0003:244:17FF:FEB6:D37D 2001:DB8:0000:0004:244:17FF:FEB6:D37D 2001:DB8:0000:0005:244:17FF:FEB6:D37D 2001:DB8:0000:0006:244:17FF:FEB6:D37D 2001:DB8:0000:0007:244:17FF:FEB6:D37D



Each authorized station has the same 64-bit device ID (244:17FF:FEB6:D37D), because the value of the last four blocks in the mask is FFFF (binary value 1111 1111).

FFFF requires all bits in each corresponding block of an authorized IPv6 address to have the same "on" or "off" setting as the device ID in the specified IPv6 address. In this case, each bit in the device ID (last four blocks) in an authorized IPv6 address is fixed and can be only one value: 244:17FF:FEB6:D37D. Table 11 Mask for Configuring Authorized IPv6 Manager Stations in Different Subnets

68

1st block

2nd block

3rd block

4th block

5th block

6th block

7th block

8th block

Manager- or operator-level access

IPv6 mask

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

FFFF

IPv6 address

2001

DB8

0000

0000

244

17FF

FEB6

D37D

In this example, the IPv6 mask allows up to four stations in different subnets to access the switch. This authorized IP manager configuration is useful if only management stations are specified by the authorized IPv6 addresses. For how the bitmap of the IPv6 mask determines authorized IP manager stations, see fix this — Example of How an ACL Filters Packets —

IPv6 Management Security Features

Table 12 (page 69) shows the bits in the fourth block of the mask that determine the valid subnets in which authorized stations with an IPv6 device ID of 244:17FF:FEB6:D37D reside. Table 12 How a mask determines authorized IPv6 manager addresses by subnet Fourth block in mask: FFF8 Last block in IPv6 address: D37DFourth block in prefix ID of IPv6 address: 0000 Bit numbers

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

15

14

13

12

11

10

9

8

7

6

5

4

3

2

1

0

Bit value

F

F

F

8

FFF8: Fourth block in mask

1

1

1

1

1

1

1

1

1

1

1

1

1

0

0

0

0000: Fourth block in IPv6 address

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

Bit setting:

1 = On

0 = Off

FFF8 in the fourth block of the mask means that bits 3 to 15 of the block are fixed and, in an authorized IPv6 address, must correspond to the "on" and "off" settings shown for the binary equivalent 0000 in the fourth block of the IPv6 address. Conversely, bits 0 to 2 are variable and, in an authorized IPv6 address, may be either "on" (1) or "off" (0). As a result, assuming that the seventh and eighth bytes (fourth hexadecimal block) of an IPv6 address are used as the subnet ID, only the following binary expressions and hexadecimal subnet IDs are supported in this authorized IPv6 manager configuration: Table 13 Binary equivalents of authorized subnet IDs (in hexadecimal) Authorized subnet ID in fourth hexadecimal block of IPv6 address

Binary equivalent

0000

0000 0000

0001

0000 0001

0002

0000 0010

0003

0000 0011

0004

0000 0100

0005

0000 0101

0006

0000 0110

0007

0000 0111

Secure shell (SSH) for IPv6 Beginning with software release K.14.01, SSH for IPv4 and IPv6 operate simultaneously with the same command set. Both are enabled in the default configuration, and are controlled together by the same command set.

Secure shell (SSH) for IPv6

69

SSH for IPv6 provides the same Telnet-like functions through encrypted, authenticated transactions as SSH for IPv4. SSH for IPv6 provides CLI (console) access and secure file transfer functionality. The following types of transactions are supported: •

Client public-key authentication Public keys from SSH clients are stored on the switch. Access to the switch is granted only to a client whose private key matches a stored public key.



Password-only client authentication The switch is SSH-enabled but is not configured with the login method that authenticates a client's public-key. Instead, after the switch authenticates itself to a client, users connected to the client authenticate themselves to the switch by providing a valid password that matches the operator- and/or manager-level password configured and stored locally on the switch or on a RADIUS or TACACS+ server.



Secure Copy (SCP) and Secure FTP (SFTP) client applications You can use either one SCP session or one SFTP session at a given time to perform secure file transfers to and from the switch.

SCP and SFTP for IPv6 You can take advantage of the SCP and SFTP client applications to provide a secure alternative to TFTP for transferring sensitive switch information, such as configuration files and login information, between the switch and an administrator workstation. Because SCP and SFTP run over an encrypted SSH session, you can use a secure SSH tunnel to: •

Transfer files and update HP Switch software images.



Distribute new software images with automated scripts that make it easier to upgrade multiple switches simultaneously and securely. You can perform secure file transfers to and from IPv4 and IPv6 client devices by entering the ip ssh filetransfer command. After an IPv6 client running SCP/SFTP successfully authenticates and opens an SSH session on the switch, you can perform secure file transfers to and from IPv4 and IPv6 client devices by entering the ip ssh filetransfer command. For instructions on how to use this command, see “Performing secure file transfers to and from IPv4 and IPv6 client devices” (page 64). For information on the file transfer commands and software utilities to use, see the documentation that comes with an SCP or SFTP client application. NOTE:

Enabling SSH file transfer disables TFTP and auto-TFTP operation.

The switch supports one SFTP session or one SCP session at a time. All files on the switch have read-write permission. However, several SFTP commands, such as create or remove, are not supported and return an error. For complete information on how to configure SCP or SFTP in an SSH session to copy files to and from the switch, see the "File Transfers" appendix in the Management and Configuration Guide for your switch.

70

IPv6 Management Security Features

4 Multicast Listener Discovery (MLD) Snooping Table 14 Summary of commands Command syntax

Description

Default

CLI page reference

[no] ipv6 mld [ enable | disable ]

Enables MLD Disabled snooping on a VLAN.

72

ipv6 mld [ auto port-list | blocked port-list | forward port-list ]

Sets per-port traffic Auto filters, which specify how each port should handle MLD traffic.

73

[no] ipv6 mld querier

Enables the switch to act as querier on a VLAN.

Enabled

74

[no] ipv6 mld fastleave port-list

Enables the fast leave Enabled function on the specified ports in a VLAN.

76

[no] ipv6 mld fastleave port-list

Enables the forced Disabled fast leave function on the specified ports in a VLAN.

77

show ipv6 mld

Displays MLD status information for all VLANs on the switch that have MLD configured.



77

show ipv6 mld vlan vid

Displays MLD status for the specified VLAN vid —VLAN ID.



77

show ipv6 mld config

Displays current global MLD configuration for all MLD-enabled VLANS on the switch.



78

show ipv6 vlan vid [ config ]

Displays current MLD configuration for the specified VLAN.



78

show ipv6 mld vlan vid group

Lists the ports currently joined for all IPv6 multicast group addresses in the specified VLAN.



79

show ipv6 mld vlan vid group ipv6-addr

Lists the ports currently joined for the specified IPv6 multicast group address in the specified VLAN.



79

show ipv6 mld statistics

Shows MLD statistics for all MLD-enabled VLANs.



80

71

Table 14 Summary of commands (continued) Command syntax

Description

show ipv6 mld vlan vid statistics

Shows MLD statistics for the specified VLAN.



80

show ipv6 mld vlan vid

Displays MLD counters for the specified VLAN.



82

counters

Default

CLI page reference

Overview Multicast addressing allows one-to-many or many-to-many communication among hosts on a network. Typical applications of multicast communication include audio and video streaming, desktop conferencing, collaborative computing, and similar applications. MLD is an IPv6 protocol used on a local link for multicast group management. MLD operates in a manner similar to IGMP in IPv4 networks. MLD is enabled per VLAN and is analogous to the IPv4 IGMP protocol. In the factory default state(MLD disabled), the switch floods all IPv6 multicast traffic it receives on a given VLAN through all ports on that VLAN except the port receiving the inbound multicast traffic. Enabling MLD imposes management controls on IPv6 multicast traffic to reduce unnecessary bandwidth usage. MLD is configured per-VLAN. MLD snooping is a subset of the MLD protocol that operates at the port level and conserves network bandwidth by reducing the flooding of multicast IPv6 packets. MLD can be configured using version 1 (MLDv1) or version 2 (MLDv2). MLDv2 introduces source-specific multicast in which the only packets delivered to the receiver are those that originate from a specified source address requested by the receiver. The receiver indicates interest in receiving traffic to a multicast address and additionally can indicate interest in receiving traffic from only one specified source sending to that multicast address. This reduces the amount of multicast routing information that needs to be maintained. These options are available for MLDv1 and MLDv2: •

Query interval—the time interval between general queries sent by the querier.



Query Max Response Time—the amount of time to wait for a response to a query.



Last Member Query Interval—the amount of time the querier waits to receive a response from members to a group-specific query message. It also specifies the amount of time between successive group-specific query messages.



Robustness—the number of times to retry a query.



Fast Learn—enables the port to learn group information when there is a topology change.

This chapter describes MLD snooping and the CLI commands available for configuring it and for viewing its status. For introductory information on MLD snooping, see “MLD snooping” (page 83).

Enabling or Disabling MLD Snooping on a VLAN Several CLI commands are available for configuring MLDv1 and MLDv2 parameters on a switch. To enable or disable MLD on a VLAN, enter the appropriate command.

Syntax: [no] ipv6 mld [ enable | disable ] The no form disables MLD snooping on a VLAN. 72

Multicast Listener Discovery (MLD) Snooping

NOTE:

This command must be issued in a VLAN context.

The command no ipv6 mld deletes the configuration on the VLAN. The command no ipv6 mld option will remove that option from the configuration and apply the default if one exists. MLDv2 is disabled by default. enable: Enables MLDv2 on a VLAN. disable: Disables MLDv2 on a VLAN. The last-saved or the default MLD configuration is saved, whichever is most recent. Example 38 Enabling or Disabling MLD Snooping on a VLAN To enable MLD snooping on VLAN 8: HP HP HP HP

Switch(vlan-8)# ipv6 mld enable Switch# config Switch(config)# vlan 8 Switch(vlan-8)# ipv6 mld

To disable MLD snooping on VLAN 8: HP Switch(vlan-8)# no ipv6 mld

Setting the MLD Version You can specify the MLD version you wish to use with this command.

Syntax [no] ipv6 mld version 1 - 2 strict Note: This command must be issued in a VLAN context. Sets the MLD protocol version to use. The no version of the command resets the version to the default, version 2. Default: 2 strict: Only the packets for the selected version are processed. Example 39 To set MLD to version 1 for VLAN 8 and version 2 for VLAN 9 HP HP HP HP

Switch(vlan-8)# Switch(vlan-8)# Switch(config)# Switch(vlan-9)#

ipv6 mld version 1 exit vlan 9 ipv6 mld version 2

Configuring per-port MLD traffic filters Syntax: ipv6 mld [ auto port-list | blocked port-list | forward port-list ] Sets per-port traffic filters, which specify how each port should handle MLD traffic. Allowed settings are: auto Follows MLD snooping rules: packets are forwarded for joined groups.

Setting the MLD Version

73

blocked All multicast packets are dropped, except that packets for well-known addresses are forwarded. forward All multicast packets are forwarded. port-list Specifies the affected port or range of ports. The default value of the filter is auto. NOTE:

This command must be issued in a VLAN context.

Example 40 Configuring per-port MLD traffic filters HP Switch(vlan-8)# ipv6 mld forward 16-18 HP Switch(vlan-8)# ipv6 mld blocked 19-21 HP Switch(vlan-8)# show ipv6 mld vlan 8 config MLD Service Vlan Config VLAN ID : 8 VLAN NAME : VLAN8 MLD Enabled [No] : Yes Querier Allowed [Yes] : Yes Port ---13 14 15 16 17 18 19 20 21 22 23 24

Type --------100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T 100/1000T

| + | | | | | | | | | | | |

Port Mode --------auto auto auto forward forward forward blocked blocked blocked auto auto auto

Forced Fast Leave ----------------No No No No No No No No No No No No

Fast Leave ---------Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Configuring the querier For more information on queriers, see “Queries” (page 86).

Syntax: [no] ipv6 mld querier Enables the switch to act as querier on a VLAN. The no form disables the switch from acting as querier on a VLAN. The querier function is enabled by default. If another switch or a multicast router is acting as the MLD querier on the VLAN, this switch defers to that device. If an acting querier stops performing the querier function, all querier-enabled switches and multicast routers on the VLAN enter an election to determine the next device to act as querier. NOTE:

74

This command must be issued in a VLAN context.

Multicast Listener Discovery (MLD) Snooping

Example 41 Configuring the querier To disable the switch from acting as querier on VLAN 8: HP Switch(vlan-8)# no ipv6 mld querier To enable the switch to act as querier on VLAN 8: HP Switch(vlan-8)# ipv6 mld querier

Configuring the Query Interval To specify the number of seconds between membership queries, enter this command with the desired interval.

Syntax [no] ipv6 mld query-interval 60 - 31744 NOTE:

This command must be issued in a VLAN context.

Specifies the number of seconds between membership queries. The no form of the command sets the interval to the default of 125 seconds. Default: 125 seconds. Example 42 To set the query-interval to 300 seconds on ports in VLAN 8: HP Switch(vlan-8)# ipv6 mld query-interval 300

Configuring the Query Maximum Response Time To specify the maximum amount of time to wait for a response to a query, enter this command.

Syntax [no] ipv6 mld query-max-response-time 10 - 128 NOTE:

This command must be issued in a VLAN context.

Specifies the number of seconds to wait for a response to a query. The no form of the command sets the interval to the default of 10 seconds. Default: 10 seconds. Example 43 To set the query-max-response-time to 30 seconds on ports on VLAN 8: HP Switch(vlan-8)# ipv6 mld query-max-response-time 30

Configuring the Number of Times to Retry a Query To specify the number of times to retry a query, enter this command.

Syntax [no] ipv6 mld robustness 1 - 8 NOTE:

This command must be issued in a VLAN context.

Specifies the number of times to retry a query. The no form of the command sets the interval to the default of 2. Configuring the Query Interval

75

Default: 2 Example 44 To set the number of times to retry a query to 4 on ports on VLAN 8: HP Switch(vlan-8)# ipv6 mld robustness 4

Configuring the Last Member Query Interval You can specify the amount of time that the querier waits to receive a response from members to a group-specific query message by entering this command.

Syntax [no] ipv6 mld last-member-query-interval 1 - 2 NOTE:

This command must be issued in a VLAN context.

Sets the amount of time that the querier waits to receive a response from members to a group-specific query message. It also specifies the amount of time between successive groupspecific query messages. The no form of the command sets the interval to the default of 1 second. Default: 1 second. Example 45 To set the amount of time the querier waits to 20 on VLAN 8: HP Switch(vlan-8)# ipv6 mld last-member-query-interval 2

Configuring Fast Learn The Fast Learn option allows fast convergence of multicast traffic after a topology change. When a new port joins or moves to a forwarding state, MLD sends joins for the groups it maintains. For MLDv1, a join is transmitted for each group if the switch is a non-querier. If the switch is a querier, an MLDv1 query is sent to learn the group on that port. For MLDv2, an IS_EX report is sent when the switch is a non-querier. If the switch is a querier, an MLDv2 query is sent on the port to learn the group. This command is executed in the global config context

Syntax [no] ipv6 mld fastlearn [ port-list | all ] This command enables fast learn on the specified ports in a VLAN. The no form of the command disables the fast learn function on the specified ports. The all option enabled or disables all ports. Default: Disabled Example 46 To enable fastlearn on ports 5 and 6: HP Switch(config)# ipv6 mld fastlearn 5 - 6

Configuring fastleave For more information on fast leave, see “Fast leaves and forced fast leaves” (page 87).

76

Multicast Listener Discovery (MLD) Snooping

Syntax: [no] ipv6 mld fastleave port-list Enables the fast leave function on the specified ports in a VLAN. The no form disables the fast leave function on the specified ports in a VLAN. Default: Enabled NOTE:

This command must be issued in a VLAN context.

Example 47 Configuring fastleave To disable fast leave on ports in VLAN 8: HP Switch(vlan-8)# no ipv6 mld fastleave 14-15 To enable fast leave on ports in VLAN 8: HP Switch(vlan-8)# ipv6 mld fastleave 14-15

Configuring forced fastleave For more information on forced fast leave, see “Fast leaves and forced fast leaves” (page 87).

Syntax: [no] ipv6 mld fastleave port-list Enables the forced fast leave function on the specified ports in a VLAN. The no form disables the forced fast leave function on the specified ports in a VLAN. Default: Disabled NOTE:

This command must be issued in a VLAN context.

Example 48 Configuring forced fastleave To enable forced fast leave on ports in VLAN 8: HP Switch(vlan-8)# ipv6 mld forcedfastleave 19-20 To disable forced fast leave on ports in VLAN 8: HP Switch(vlan-8)# no ipv6 mld forcedfastleave 19-20

Viewing the current MLD status For more information, see “Current MLD status” (page 87).

Syntax: show ipv6 mld Displays MLD status information for all VLANs on the switch that have MLD configured. show ipv6 mld vlan vid Displays MLD status for the specified VLAN. vid VLAN ID

Configuring forced fastleave

77

Example 49 Displaying the MLD Configuration for a VLAN on the Switch, Version 2: HP Switch# show ipv6 mld vlan 8 MLD Service Protocol Info VLAN ID : 8 MLD Version

Name : VLAN8 : 2

MLD Interface State : Querier Querier Address : fe80::218:71ff:fec4:2f00 [this switch] Version : 2 Up Time : 0h:5m:3s Expires : 1h:14m:55s Ports with multicast routers : Active Group Addresses -------------------------ff3e:30:2001:db8:8:0:7:101 ff3e:30:2001:db8:8:0:7:102

Tracking -------Filtered Standard

Vers ---2 2

Mode ---EXC INC

Uptime ------20h 55m 18h 18m

Expires ------22h 50m 12m 10s

Example 50 Displaying the MLD Configuration for a VLAN on the Switch, Version 1 HP Switch# show ipv6 mld vlan 8 MLD Service Protocol Info VLAN ID : 8 MLD Version

Name: VLAN8 : 1

MLD Interface State : Querier Querier Address : fe80::218:71ff:fec4:2f00 [this switch] Version : 1 Up Time : 0h:5m:3s Expires : 1h:14m:55s Ports with multicast routers : Active Group Addresses ------------------------------ff3e:30:2001:db8:8:0:7:101 ff3e:30:2001:db8:8:0:7:102

Tracking -------Filtered Standard

Uptime -------20h 55m 18h 18m

Expires ------22h 50m 12m 10s

Configuring the current MLD For more information, see “Current MLD configuration” (page 88).

Syntax: show ipv6 mld config Displays current global MLD configuration for all MLD-enabled VLANS on the switch. show ipv6 vlan vid [ config ] Displays current MLD configuration for the specified VLAN, including per-port configuration information. vid VLAN ID

78

Multicast Listener Discovery (MLD) Snooping

Example 51 Configuring the current MLD The general form of the command might look like this: HP Switch# show ipv6 mld config MLD Service Config Control Unknown Multicast : Yes Forced Fast Leave Timeout (deci-seconds) : 4 deci-seconds VLAN ID ------8 9

VLAN NAME ---------VLAN8 VLAN9

MLD Enabled ----------Yes Yes

Querier Allowed --------------Yes Yes

MLD Version ----------2 1

Listing ports currently joined Syntax: show ipv6 mld vlan vid group Lists the ports currently joined for all IPv6 multicast group addresses in the specified VLAN. vid VLAN ID show ipv6 mld vlan vid group ipv6-addr Lists the ports currently joined for the specified IPv6 multicast group address in the specified VLAN. vid VLAN ID ipv6-addr address of the IPv6 multicast group for which you want information. show ipv6 mld vlan vid group port-num Shows a list of all the MLD groups on the specified port. show ipv6 mld vlan vid group ipv6-addr source ipv6-addr Only for MLDv2. Specify the source IPv6 address.

Listing ports currently joined

79

Example 52 Ports Joined to Multicast Groups in a Specific VLAN, Version 1 The general form of the command is:. HP Switch# show ipv6 mld vlan 9 group ff33::00 MDL Service Protocol Group Info VLAN ID : 9 VLAN Name : VLAN9 Group Address : ff33:: Last Reporter : fe80::7061:4b38:dbea:2c4f Group Type : Filtered Port ---3 5

Uptime --------1h 45m 1h 9m

Expires --------4m 34s 4m 34s

. . . Group Address : ff33:: Last Reporter : fe80::7061:4b38:dbea:2c4f Group Type : Filtered Port ---6 8

Uptime Expires --------- --------1h 45m 4m 34s 1h 9m 4m 34s

Example 53 Group Information for a VLAN, Version 2 The specific form of the command is similar, except that it lists the port information for only the specified group. HP Switch# show ipv6 vlan 8 group ff33:: MLD Service Protocol Group Info VLAN ID : 8

Name: VLAN8

Group Address : ff33:30:2001:db8:8:0:7:101 Last Reporter : fe80::7061:4b38:dbea:2c4f Group Type : Filtered

Port ---5 6 7 8

Vers ---2 2 1 1

Mode ---INC EXC EXC EXC

Uptime ------20h 55m 8h 18m 1h 18m 18h 18m

V1 Timer ------19h 22m 19h 21m

Expires ------22h 50m 12m 10s 12m 10s 12m 10s

Filter Timer ------20h 30m 22h 23m 0 0

Group Address : ff33:30:2001:db8:8:0:7:102 Source Address : fe80::c0a8:100 Source Type : Filtered Port ---5 6

80

Mode ---EXC INC

Uptime --------10m 34s 10m 34s

Multicast Listener Discovery (MLD) Snooping

Expires -------10m 34s 10m 34s

Configured Mode ---------------Auto Forward

Sources Forwarded --------2 3 0 0

Sources Blocked ------0 2 0 0

Viewing MLD statistics Syntax: show ipv6 mld statistics Shows MLD statistics for all MLD-enabled VLANs show ipv6 mld vlan vid statistics Shows MLD statistics for the specified VLAN. vid VLAN ID The general form the of the command shows the total number of MLD-enabled VLANs and a count of multicast groups currently joined. Both forms of the command show VLAN IDs and names, as well as the number of filtered and standard multicast groups and the total number of multicast groups.

Viewing MLD statistics

81

Example 54 MLD Statistics for all VLANs Configured, Version 2 HP Switch# show ipv6 mld statistics MLD Service Statistics Mulitcast Groups Joined :4

(EXCLUDE Mode : 2

INCLUDE Mode : 2)

MLD Joined Groups Statistics VLAN ID ------8 9

VLAN NAME ---------VLAN8 VLAN9

filtered -------2 2

standard --------0 0

total -----2 2

Example 55 MLD Statistics for a Single VLAN Version 2 HP Switch# show ipv6 mld vlan 8 statistics MLD Statistics VLAN ID : 8

VLAN NAME: VLAN8

Number of Filtered Groups : 4 Number of Standard Groups : 0 Total Multicast Groups Joined : 4 Mode -------Filtered Standard Total

EXCLUDE ------3 0 3

INCLUDE ------1 0 1

Example 56 MLD Statistics for a Single VLAN Version 1 HP Switch# show ipv6 mld vlan 8 statistics MLD Statistics VLAN ID : 8

VLAN NAME : VLAN8

Number of Filtered Groups : 4 Number of Standard Groups : 0 Total Multicast Groups Joined : 4

Viewing MLD counters For more information, see “Counters” (page 90).

Syntax: show ipv6 mld vlan vid counters Displays MLD counters for the specified VLAN. vid VLAN ID

82

Multicast Listener Discovery (MLD) Snooping

EXCLUDE -------1 1

INCLUDE -------1 1

Example 57 MLD counters for a single VLAN HP Switch# show ipv6 mld vlan 8 counters MLD Service Vlan Counters VLAN ID : 8 Name : VLAN8

V1 All Hosts Query V2 All Hosts Query V1 Group Specific Query V2 Group Specific Query Group and Source Specific Query V2 Member Report V1 Member Join V1 Member Leave Forward to Routers Forward to VLAN

Rx ---55 0 0 0 0 15 15 30 83 48

Tx ---888 0 0 0 0 0 0 0 0 0

Errors: Unknown MLD Type 2 Unknown Packet 3 Malformed Packet 0 Bad Checksum 0 Martian Source 0 Packet Received on MLD-disabled Interface 0 Interface Wrong Version Query 0 Port Counters: Fast Leave : 4 Forced Fast Leave : 0 Membership Timeout : 8

MLD snooping There are several roles that network devices may play in an IPv6 multicast environment: MLD host A network node that uses MLD to "join" (subscribe to) one or more multicast groups. Multicast router A router that routes multicast traffic between subnets. Querier A switch or multicast router that identifies MLD hosts by sending out MLD queries to which the MLD hosts respond. A network node that acts as a source of IPv6 multicast traffic is only an indirect participant in MLD snooping—it just provides multicast traffic, and MLD does not interact with it. (However, in an application like desktop conferencing a network node may act as both a source and an MLD host, but MLD interacts with that node only in its role as an MLD host.) A source node creates multicast traffic by sending packets to a multicast address. In IPv6, addresses with the first eight bits set (that is, "FF" as the first two characters of the address) are multicast addresses, and any node that listens to such an address will receive the traffic sent to that address. Application software running on the source and destination systems cooperates to determine what multicast address to use. (This is a function of the application software, not of MLD.) MLD snooping

83

For example, if several employees engage in a desktop conference across the network, they all need application software on their computers. At the start of the conference, the software on all the computers determines a multicast address of, for example, FF3E:30:2001:DB8::101 for the conference. Then any traffic sent to that address can be received by all computers listening on that address.

General operation Multicast communication can take place without MLD, and by default, MLD is disabled. In that case, if a switch receives a packet with a multicast destination address, it floods the packet to all ports in the same VLAN (except the port that it came in on), as shown in Figure 1 (page 84). Any network nodes that are listening to that multicast address will see the packet; all other hosts ignore the packet. Figure 1 Without MLD, multicast traffic is flooded to all ports

When MLD snooping is enabled on a VLAN, the switch acts to minimize unnecessary multicast traffic. If the switch receives multicast traffic destined for a given multicast address, it forwards that traffic only to ports on the VLAN that have MLD hosts for that address, as shown in Figure 2 (page 85). It drops that traffic for ports on the VLAN that have no MLD hosts (except for a few special cases explained below).

84

Multicast Listener Discovery (MLD) Snooping

Figure 2 With MLD snooping, traffic is sent to MLD hosts

MLD snooping enabled Listener (MLD host)

Switch Source Listener (MLD host) MLD snooping operates on a single VLAN (though there can be multiple VLANs, each running MLD snooping). Cross-VLAN traffic is handled by a multicast router.

Forwarding in MLD snooping When MLD snooping is active, a multicast packet is handled by the switch as shown in the following list. The packet is: •

Forwarded to ports that have nodes that have joined the packet's multicast address (that is, MLD hosts on that address packet)



Forwarded toward the querier—If the switch is not the querier, the packet is forwarded out the port that leads to the querier.



Forwarded toward any multicast routers—If there are multicast routers on the VLAN, the packet is forwarded out any port that leads to a router.



Forwarded out administratively forwarded ports—The packet is forwarded through all ports set administratively to forward mode. (See the description of forward modes, below.)



Dropped for all other ports.

Each individual port's forwarding behavior can be explicitly set using a CLI command to one of these modes: Auto (the default mode) The switch forwards packets through this port based on the MLD rules and the packet's multicast address. In most cases, this means that the switch forwards the packet only if the port connects to a node that is joined to the packet's multicast address (that is, to an MLD host). There is seldom any reason to use a mode other than "auto" in normal operation (though some diagnostics may make use of "forward" or "block" mode). MLD snooping

85

Forward The switch forwards all IPv6 multicast packets through the port. This includes IPv6 multicast data and MLD protocol packets. Block The switch drops all MLD packets received by the port and blocks all outgoing IPv6 multicast packets through the port, except those packets destined for well-known IPv6 multicast addresses. This has the effect of preventing IPv6 multicast traffic from moving through the port. The switch floods all packets with "well-known" IPv6 multicast destination addresses through all ports. Well-known addresses are permanent addresses defined by the Internet Assigned Numbers Authority (www.iana.org). IPv6 standards define any address beginning with FF0x/12 (binary 1111 1111 0000) as a well-known address.

Listeners and joins The "snooping" part of MLD snooping arises because a switch must keep track of which ports have network nodes that are MLD hosts for any given multicast address. It does this by keeping track of "joins" on a per-port basis. A network node establishes itself as an MLD host by issuing a multicast "join" request (also called a multicast "report") for a specific multicast address when it starts an application that listens to multicast traffic. The switch to which the node is connected sees the join request and forwards traffic for that multicast address to the node's port.

Queries The querier is a multicast router or a switch that periodically asks MLD hosts on the network to verify their multicast join requests. There is one querier for each VLAN, and all switches on the VLAN listen to the responses of MLD hosts to multicast queries and forward or block multicast traffic accordingly. All of the HP switches have the querier function enabled by default. If there is another device on the VLAN that is already acting as querier, the switch defers to that querier. If there is no device acting as querier, the switch enters an election state and negotiates with other devices on the network (if any) to determine which one will act as the querier. The querier periodically sends general queries to MLD hosts on each multicast address that is active on the VLAN. The time that the querier waits between sending general queries is known as the query interval; the MLD standard sets the default query interval to 125 seconds. Network nodes that wish to remain active as MLD hosts respond to the queries with join requests; in this way they continue to assert their presence as MLD hosts. The switch through which any given MLD host connects to the VLAN sees the join requests and continues forwarding traffic for that multicast address to the MLD host's port.

Leaves A node acting as an MLD host can be disconnected from a multicast address in two ways: •

It can stop sending join requests to the querier. This might happen if the multicast application quits or the node is removed from the network. If the switch goes for slightly more than two query intervals without seeing a join request from the MLD host, it stops sending multicast traffic for that multicast address to the MLD host's port.



It can issue a "leave" request. This is done by the application software running on the MLD host. If the MLD host is the only node connected to its switch port, the switch sees the leave request and stops sending multicast packets for that multicast address to that port. (If there is more than one node connected to the port the situation is somewhat more complicated, as explained under “Fast leaves and forced fast leaves” (page 87).)

86

Multicast Listener Discovery (MLD) Snooping

Fast leaves and forced fast leaves The fast leave and forced fast leave functions can help to prune unnecessary multicast traffic when an MLD host issues a leave request from a multicast address. Fast leave is enabled by default, and forced fast leave is disabled by default. Both functions are applied to individual ports. Which function to use depends on whether a port has more than one node attached to it, as follows: •

If a port has only one node attached to it, when the switch sees a leave request from that node (an MLD host) it knows that it does not need to send any more multicast traffic for that multicast address to the host's port.



If fast leave is enabled (the default setting), the switch stops sending the multicast traffic immediately.



If fast leave is disabled, the switch continues to look for join requests from the host in response to group-specific queries sent to the port.



The interval during which the switch looks for join requests is brief and depends on the forced fast leave setting:



If forced fast leave is enabled for the port, it is equal to the "forced fast leave interval" (typically several seconds or less).



If forced fast leave is disabled for the port, the period is about 10 seconds (governed by the MLD standard).



When this process has completed, the multicast traffic for the group will stop (unless the switch sees a new join request).



If a single port has multiple nodes attached to it, a leave request from one of those nodes (an MLD host) does not provide enough information for the switch to stop sending multicast traffic to the port. In this situation, the fast leave function does not operate. The switch continues to look for join requests from any MLD hosts connected to the port in response to group-specific queries sent to the port. As in the case described above for a single-node port that is not enabled for fast leave, the interval during which the switch looks for join requests is brief and depends on the forced fast leave setting:



If forced fast leave is enabled for the port, it is equal to the "forced fast leave interval" (typically several seconds or less).



If forced fast leave is disabled for the port, the period is about 10 seconds (governed by the MLD standard).



When this process has completed, the multicast traffic for the group will stop unless the switch sees a new join request. This reduces the number of multicast packets forwarded unnecessarily.

Current MLD status The following information is shown for each VLAN that has MLD snooping enabled: •

VLAN ID number and name



Querier address IPv6 address of the device acting as querier for the VLAN.



Querier up time Length of time in seconds that the querier has been acting as querier.



Querier expiry time If this switch is the querier, this is the amount of time until the switch sends the next general query. If this switch is not the querier, this is the amount of time in seconds until the current querier is considered inactive (after which a new querier election is held). Current MLD status

87



Ports with multicast routers Ports on the VLAN that lead toward multicast routers (if any).



Multicast group address information For each active group on the VLAN, including:





Multicast group address.



Type of tracking for multicast joins: standard or filtered. •

If MLD snooping is enabled, port-level tracking results in filtered groups.



If MLD snooping is not enabled, joins result in standard groups being tracked by this device.



In addition, if hardware resources for multicast filtering are exhausted, new joins may result in standard groups even though MLD snooping is enabled.



MLD version number (MLDv2 display only)



Mode—INCLUDE or EXCLUDE (MLDv2 only): when INCLUDE is displayed, the host has requested specific source/group pairs. When EXCLUDE is displayed, the host has requested all sources for a group except for a specified list of sources to exclude.

Uptime The length of time the group has been joined



Expiry time Time until the group expires if no joins are seen.



The ports that have joined the multicast group.

The group addresses that are listed typically result from several network functions. In the example Example 119 (page 189), several of the addresses at the top of the list for each VLAN are IANA well-known addresses (see www.iana.org/assignments/ipv6-multicast-addresses); the addresses in the form of ff02::1:ffxx:xxxx are solicited-node multicast addresses (used in IPv6 ND); and the addresses beginning with ff3e are group addresses used by listeners to streaming video feeds.

Current MLD configuration The following information applies to all MLD-enabled VLANs: •

Control unknown multicast If this is set to YES, any IPv6 multicast packets that are not joined by an MLD host are sent only to ports that have detected a multicast router or ports that are administratively forwarded. If this is set to NO (or if MLD snooping is disabled), unjoined IPv6 multicast packets are flooded out all ports in the VLAN.



Forced fast leave timeout Interval between an address-specific query and a forced fast leave (assuming no response), in tenths of seconds.



88

For each VLAN that has MLD enabled: •

Whether MLD is enabled on the VLAN (default NO, but the VLAN will not show up on this list unless MLD is enabled).



Whether the switch can act as querier for the VLAN (default YES).



the MLD version (1 or 2)

Multicast Listener Discovery (MLD) Snooping

Example 58 MLD configuration for a specific VLAN HP Switch# show ipv6 mld vlan 8 config MLD Service Vlan Config VLAN ID : VLAN NAME : MLD Enable : Querier Allowed : MLD Version : Strict Mode : Last Member Query Interval(seconds): Query Interval(seconds) : Query Max. Response Time(seconds) : Robustness-Count : Port ---13 14 15 16

Port Type --------100/1000T 100/1000T 100/1000T 100/1000T

Port Mode --------auto auto auto auto

8 VLAN8 Yes Yes 2 Yes 4 seconds 135 seconds 10 seconds 2

Forced Fast Leave ----------------No No No No

Fast Leave ---------Yes Yes Yes Yes

Fast Learn ---------Yes Yes Yes Yes



VLAN ID and Name



MLD enabled: whether MLD is enabled on the VLAN (default NO, but the information for this VLAN will be listed only if MLD is enabled)



Querier Allowed: whether the switch is allowed to act as querier on the VLAN



MLD version



Strict Mode: whether strict mode is enabled



Last Member Query Interval: showing the amount of time the querier waits for a response from members, in seconds



Query Interval showing the length of time between membership queries, in seconds



Query Max. Response Time displaying the number of seconds to wait for a response to a query, in seconds



Robustness-Count displaying the number of times to retry a query



Port information for each IPv6 multicast group address in the VLAN (general group command) or for the specified IPv6 multicast group address (specific group command): •

Group multicast address.



Last reporter: Last MLD host to send a join to the group address.



Group expiry: Time until the group expires if no further joins are seen.



Port name for each port.



Port type for each port: Ethernet connection type.



Port mode for each port: •

auto (follows MLD snooping rules, that is, packets are forwarded for joined groups)



forward (all multicast packets are forwarded to this group)



blocked (all multicast packets are dropped, except that packets for well-known addresses are forwarded)



Expiry time for each port: Amount of time until this port is aged out of the multicast address group, unless a join is received. Current MLD configuration

89



whether Forced Fast Leave is enabled or disabled



whether Fast Leave is enabled or disabled



whether Fast Learn is enabled or disabled - not in sw commands

Counters The following information is shown: •

VLAN number and name



For each VLAN, number of:





90



general queries(MLDv1) received and sent



general queries (MLDv2) received and sent



version 1 group-specific queries received and sent



version 2 group-specific queries received and sent



group and source-specific queries received and sent



MLD version2 member reports (joins) received



MLD version 1 member reports (joins) received



version 1 leaves received and sent



packets forwarded to routers on this VLAN received and sent



packets forwarded to all ports on this VLAN received and sent

Errors, number of:



MLD packets of unknown type received • packets of unknown type received



malformed packets received



packets with bad checksums



packets from a martian source (the wrong subnet on an interface)



packets received on an MLD-disabled interface



queries—when a VLAN is configured as MLDv2 and an MLDv1 query is received from another switch for that VLAN, this counter is incremented. The reverse also applies.

Port Counters, number of:



fast leaves that have occurred



forced fast leaves that have occurred



times a join has timed out on this VLAN

Multicast Listener Discovery (MLD) Snooping

5 IPv6 Access Control Lists (ACLs) Table 15 Summary of commands Command syntax

Description

Default

CLI page reference

ipv6 access-list name-str

Adds or inserts an ACE to the end of an ACL.

-

120

ipv6 access-list ascii-str

Prerequisite for entering or editing ACEs in an ACL.

-

122

[ deny | permit ] [ ipv6 | ipv6-protocol | ipv6-protocol-nbr ] [ any | hostSA | SA/prefix-length ] [ any | hostDA | DA/prefix-length ] [ dscp tos-bits | precedence ] [ log ]

Appends an ACE to the end of the list of ACEs in the current ACL.

-

122

TCP

You can optionally apply comparison operators specifying TCP or UDP source and/or destination port numbers or ranges of numbers to further define the criteria for a match.

-

125

Allows configuring an ACE to selectively permit some types of ICMP traffic, while denying other types.

-

127

[no] vlan vid ipv6 access-group identifier [ in | out Assigns an ACL to a VLAN as an RACL to ] filter routed IP traffic entering or leaving the switch on that VLAN.

-

128

[no] vlan vid ipv6 access-group identifier vlan

Assigns an ACL as a VACL to a VLAN to filter switched or routed IPv6 traffic entering the switch on that VLAN.

-

130

[no] interface [ port-list | trkx ] ipv6 access-group identifier in

Assigns an ACL as a static port ACL to a port, port list, or static trunk to filter switched or routed IPv6 traffic entering the switch on that interface.

-

131

[no] ipv6 access-list identifier

Used in the global config context to remove the specified IPv6 ACL from the

-

132

[ deny | permit ] tcp SA [ comparison-operator tcp-src-port ] DA [ comparison-operator tcp-dest-port ] UDP [ deny | permit ] udp SA [ comparison-operator udp-src-port ] DA [ comparison-operator udp-dest-port ] [ deny | permit ] icmp [ deny | permit ] icmp

SA SA

DA icmp-type icmp-code DA icmp-type-name

91

Table 15 Summary of commands (continued) Command syntax

Description

Default

CLI page reference

switch's running-config file.

92

1 - 2147483647 [ permit | deny ] ipv6-ACE-criteria

Used in the context of a given ACL, this command inserts an ACE into the ACL.

-

132

[no] 1 - 2147483647 no [ permit | deny ] ipv6-ACE-criteria

The first command option deletes the ACE assigned to the specified sequence number. The second command option deletes the ACE having the syntax specified by ipv6-ACE-criteria.

-

134

ipv6 access-list resequence identifier starting-seq-# Resets the sequence numbers for all ACEs interval in the ACL.

-

135

show access-list

Lists a summary table of the name, type, and application status of all ACLs (IPv4 and IPv6) configured on the switch.

-

140

show access-list config

Lists the configured syntax for all IPv4 and IPv6 ACLs currently configured on the switch.

-

141

show access-list vlan vid

Lists the current IPv4 and IPv6 ACL assignments to the specified VLAN (in the running config file).

-

142

show access-list ports [ all | port-list ]

Lists the current static port ACL assignments for ports and trunks in the running config file.

-

144

show access-list identifier [ config ]

Displays detailed information on the content of a specific ACL configured in the running-config file.

-

145

IPv6 Access Control Lists (ACLs)

Table 15 Summary of commands (continued) Command syntax

Description

[ show | clear ] statistics aclv4 acl-name-str port port-# aclv4 acl-name-str vlan vid [ in | out | vlan ] aclv6 acl-name-str port port-# aclv6 acl-name-str vlan vid vlan [ in | out | vlan ]

aclv4 acl-name-str port aclv4 acl-name-str vlan aclv6 acl-name-str port aclv6 acl-name-str vlan ] aclv6 acl-name-str tunnel tunnel-id [ in |

port-# vid [ in | out | vlan ] port-# vid vlan [ in | out | vlan

Default

CLI page reference

Monitors ACL performance by using counters to display the current number of matches the switch has detected for each ACE in an ACL assigned to a switch interface.

-

156

Displays the current match (hit) count per ACE for the specified IPv4 or IPv6 static ACL assignment on a specific interface.

-

156

out ]

Introduction An access control list (ACL) contains one or more access control entries (ACEs) specifying the criteria the switch uses to either permit (forward) or deny (drop) IP packets traversing the switch's interfaces. This chapter describes how to configure, apply, and edit static IPv6 ACLs for filtering IPv6 traffic in a network populated with the switches and how to monitor IPv6 ACL actions. NOTE: Because the switches operate in an IPv4/IPv6 dual stack mode, IPv6 and IPv4 ACLs can operate simultaneously in these switches. However: •

Static IPv6 ACLs and IPv4 ACLs do not filter each other's traffic.



IPv6 and IPv4 ACEs cannot be configured in the same static ACL.



RADIUS-assigned ACLs can be configured to filter either IPv4 traffic only, or both IPv4 and IPv6 traffic. See “RADIUS-assigned ACLs” (page 160).

In this chapter, unless otherwise noted: •

The term "ACL" refers to IPv6 ACLs.



Descriptions of ACL operation apply only to IPv6 traffic.

For information on configuring and applying static IPv4 ACLs, see chapter "IPv4 Access Control Lists (ACLs)" in the Access Security Guide for your switch.

Default

CLI page reference

None

115

Filtering routed IPv6 traffic

n/a

128

Viewing ACL Configuration Data

n/a

140

Delete an ACL

n/a

132

Inserting an ACE in an existing ACL

n/a

132

Creating or Editing ACLs Offline

n/a

150

Enable ACL Logging

n/a

164

Feature Configure IPv6 ACLs

Introduction

93

IPv6 traffic filtering with ACLs can help to improve network performance and restrict network use by creating policies for: •

Switch management access Permits or denies in-band management access. This includes limiting and/or preventing the use of designated protocols that run on top of IPv6, such as TCP, UDP, ICMP, and others. Also included are the use of DSCP criteria and control for application transactions based on source and destination IPv6 addresses and transport layer port numbers.



Application access security Eliminates unwanted IPv6 traffic in a path by filtering IPv6 packets where they enter or leave the switch on specific VLAN interfaces.

The ACLs described in this chapter can filter IPv6 traffic to or from a host, a group of contiguous hosts, or entire subnets. CAUTION: The ACLs described in this chapter can enhance network security by blocking selected IPv6 traffic and can serve as part of your network security program. However, because ACLs do not provide user or device authentication or protection from malicious manipulation of data carried in IPv6 packet transmissions, they should not be relied upon for a complete security solution. Static IPv6 ACLs on the switches do not screen non-IPv6 traffic such as IPv4, AppleTalk, and IPX packets. For option information, see “Options for applying IPv6 ACLs on the switch” (page 159).

Command Summary for Configuring ACLs Create an IPv6 ACL or

HP Switch(config)# ipv6 access-list name-str HP Switch(config-ipv6-acl)# deny | permit

Add an ACE to the End of an Existing IPv6 ACL

ipv6 | esp | ah | sctp | ipv6-protocol-nbr

161

any | host SA | SA/prefix-length any | host DA | DA/prefix-length tcp | udp any | host SA | SA/prefix-length [ comparison-operator value ] any | host DA | DA/prefix-length [ comparison-operator value ] [ established ] 1

[ ack ] [ fin ] [ rst ] [ syn ] 2

icmp any | host SA | SA/prefix-length any | host DA | DA/prefix-length [ 0 - 255 [ 0 - 255 ] | icmp-message ] [dscp [log]

precedence | codepoint

]

3

Insert an ACE or a remark by Assigning a Sequence Number

HP Switch(config)# ipv6 access-list name-str HP Switch(config-ipv6-acl)# seq-# deny | permit | remark The deny and permit keywords use the options shown above for "Create an IPv6 ACL".

94

IPv6 Access Control Lists (ACLs)

132

Delete an ACE or a Remark (or both) by Sequence Number

135

HP Switch(config)# ipv6 access-listname-str HP Switch(config-ipv6-acl)# no seq-# [remark] NOTE: You can also delete an ACE by entering no permit|deny followed by the settings explicitly configured for that ACE.

Resequence the ACEs in an ACL

HP Switch(config)# ipv6 access-list resequence name-str 135 starting-# increment

1

TCP only

2

TCP flag (control bit) options for destination TCP

3

The log function applies to both “deny” and “permit” ACLs, and generates a message when there is either a “deny” match or a “permit” match.

Action

Command(s)

Page

Enter a Remark Remove a Remark:

HP HP HP HP

Switch(config)# ipv6 access-list name-str Switch(config-ipv6-acl)# remark remark-str Switch(config-ipv6-acl)# no remark Switch(config-ipv6-acl)# no seq-# remark

136

HP Switch(config)# no ipv6 access-list name-str

132

• Immediately After Entry • After entry of an ACE Delete an IPv6 ACL

138

Command Summary for Enabling, Disabling, and Displaying ACLs Enable or Disable an IPv6 RACL

HP Switch(config)# [no] vlan vid ipv6 access-group name-str in | out | vlan

128

Enable or Disable an IPv6 VACL

HP Switch(config)# [no] vlan vid ipv6 access-group name-str vlan

130

Enable or Disable a Static Port ACL HP Switch(config)# [no] interface port-list | trkx ipv6 access-group name-str in HP Switch (eth- port-list) | trkx)# [no] ipv6 access-group name-str in

131

Displaying ACL Data

139

HP HP HP HP HP HP HP

Switch# Switch# Switch# Switch# Switch# Switch# Switch#

show show show show show show show

access-list access-list access-list access-list access-list access-list access-list

acl-name-str [ config ] config ports port-list | trkx vlan vid radius port-list | all resources

Displaying or Clearing ACL Statistics HP Switch# show | clear statistics aclv6 acl-name-str 156 port port-# HP Switch# show clear statistics aclv6 acl-name-str vlan vid in | out | vlan

IPv6 ACL Terminology Access Control Entry (ACE)

A policy consisting of criteria and an action (permit or deny) to execute on a packet if it meets the criteria. For IPv6 ACEs, the elements composing the criteria include: •

source IPv6 address and prefix length

• •

destination IPv6 address and prefix length either of the following:

◦ ◦

all IPv6 traffic IPv6 traffic of a specific IPv6 protocol (For TCP, UDP, and ICMP, the criteria can include either a specific sub-type within the protocol or all traffic of the protocol type.) Introduction

95

Access Control List (ACL)



option to log packet matches with deny ACEs



optional use of DSCP (precedence and ToS settings)

A list (or set) consisting of one or more explicitly configured Access Control Entries (ACEs) and terminating with an implicit deny ipv6 any any ACE. Each ACE in an IPv6 ACL includes layer- 3 IPv6 source and destination criteria and IPv6 protocol-specific criteria. IPv6 ACLs can be applied in any of the following ways: RACL An ACL assigned to filter routed IPv6 traffic entering or leaving the switch on a VLAN or tunnel. (Separate assignments are required for inbound and outbound IPv6 traffic.) VACL An ACL assigned to filter inbound IPv6 traffic on a specific VLAN configured on the switch Static Port ACL An ACL assigned to filter inbound IPv6 traffic on a specific switch port RADIUS-Assigned ACL Dynamic ACL assigned to a port by a RADIUS server to filter inbound IPv4 and IPv6 traffic from an authenticated client on that port. See the chapter titled “Configuring RADIUS Server Support for Switch Services” in the latest Access Security Guide for your switch. Static ACLs are configured in switch memory with an alphanumeric name, and can be assigned to an IP routing interface as an RACL or VACL (or both), and to a port list (or static trunk). (RADIUS-assigned ACLs are configured on a RADIUS server, and are identified by the associated client credentials instead of an alphanumeric name.)

ACE

See Access Control Entry (ACE).

ACL

See Access Control List (ACL).

ACL ID

An alphanumeric string used to identify an ACL. See also identifier and name-str. NOTE: RADIUS-assigned ACLs are identified by client authentication data and do not use the ACL ID strings described in this chapter.

96

ACL Prefix

Follows any IPv6 address listed in an IPv6 ACE. Analogous to the ACL mask used with IPv4 ACEs. Specifies the number of leftmost, contiguous bits in a packet’s corresponding IPv6 addressing that must exactly match the IPv6 addressing in the ACE, and which bits need not match (wildcards), (see “How an ACE uses a prefix to screen packets for SA and DA matches” (page 113)).

Address Family

Used in this manual to refer to the version of the IP protocol running on the switch; IPv4 and IPv6.

CIDR

The acronym for Classless Inter-Domain Routing. In IPv6 ACEs, CIDR notation is used to specify the prefix length for SA and DA address criteria. For example, the length of the following prefix includes the first 48 bits of an address: 2001:db8:101::/48

DA

The acronym for Destination Address. In an IPv6 packet, this is the destination IPv6 address carried in the header, and identifies the packet’s destination. This is the second of two IPv6 addresses used in an ACE to determine whether there is a match between an IPv6 packet and the ACE. See also SA.

Deny

An ACE configured with this action causes the switch to drop an IPv6 packet for which there is a match within an applicable ACL.

IPv6 Access Control Lists (ACLs)

Empty ACL

An ACL that is not populated with any explicit ACEs, and functions only as a placeholder. An ACL exists in this state if any one of the following occurs: •

An ACL identifier has been created in the running config file with the ipv6 access-list [ name-str ] command, but no explicit ACEs exist in the ACL.



An ACL identifier has been assigned to an interface without first populating the ACL with ACEs. If the empty ACL did not already exist in the running config file, assigning the identifier to an interface automatically creates the empty ACL in the running config file.



An ACL configured with one or more explicit ACEs has been deleted from the running config file while the ACL is still assigned to an interface.

Note that an empty ACL does not include an Implicit Deny and does not filter traffic. However, if you configure any ACE in an empty ACL that is already assigned to an interface, the ACL immediately begins filtering traffic, which includes application of the Implicit Deny. identifier

A term used in ACL syntax statements to represent the alphanumeric name by which the ACL can be accessed. An identifier can have up to 64 characters. NOTE: RADIUS-assigned ACLs are identified by client authentication criteria and do not use the identifiers described in this chapter. See also name-str.

Implicit Deny

If the switch finds no matches between an IPv6 packet and the configured criteria in an applicable ACL, then the switch denies (drops) the packet with an implicit deny ipv6 any any function. You can preempt the Implicit Deny in a given ACL by configuring a permit ipv6 any any as the last explicit ACE in the ACL. Doing so permits any packet that is not explicitly permitted or denied by other ACEs configured sequentially earlier in the ACL. NOTE: Beginning with software release K.14.01, any dynamically created ACL will include an implicit deny for both Ipv4 and IPv6 traffic, regardless of the address family capabilities of the server (see “RADIUS-assigned ACLs” (page 160)).

Inbound Traffic

For the purpose of defining where the switch applies IPv6 ACLs to filter traffic, inbound traffic is a packet that meets one of the following criteria: Routed ACL (RACL) Inbound traffic is a packet entering the switch on an IP routing interface (or a subnet in a multinetted VLAN) with a destination IPv6 address (DA) that is for any of the following: •

an external device on a different IP routing interface than the interface on which it arrived



an IPv6 address configured on the switch itself Inbound traffic having a destination IPv6 address on the routing switch itself will be screened by an IPv6 RACL that is configured to screen inbound traffic, regardless of whether IPv6 routing is enabled. ACLs do not screen outbound traffic generated by the routing switch itself.

VLAN ACL (VACL) Inbound traffic is a packet entering the switch on a VLAN interface (or a subnet in a multinetted VLAN). Static Port ACL Inbound traffic is a packet entering the switch on the port. RADIUS-Assigned ACL Where a RADIUS server has authenticated a client and assigned an ACL to the port to filter the client’s IPv6 traffic, inbound traffic is a packet entering the switch from that client. (Note that IPv4 traffic-filtering is automatically included in a RADIUSassigned ACL configured to filter IPv6 traffic.) Introduction

97

98

name-str

The term used in ACL syntax statements to represent the “name string”; the alphanumeric string used to identify the ACL. A name string allows up to 64 alphanumeric characters. See also identifier, ACL ID.

Outbound Traffic

For defining the points where the switch applies an RACL (Routed ACL) to filter traffic, outbound traffic is routed traffic leaving the switch through an IP routing interface (or a subnet in a multinetted VLAN). “Outbound traffic” can also apply to switched traffic leaving the switch on an IP routing interface, but outbound, switched traffic is not filtered by ACLs (see also “IPv6 ACL applications” (page 99)).

Permit

An ACE configured with this action allows the switch to forward an IPv6 packet for which there is a match.

Permit Any Forwarding

An ACE configured with this action causes the switch to forward IPv6 packets that have not been permitted or denied by earlier ACEs in the list. (This has no effect on packets that are not filtered by the applicable ACL, such as switched packets entering or leaving the switch on an IP routing interface that is configured with an RACL.)

Prefix Length

In an IPv6 ACE, a network prefix is used to specify the leftmost contiguous bits in a packet’s SA and DA that must match the bit settings defined in the SA and DA configured in the ACE. The prefix length is specified (in CIDR format) by /nn immediately following the specified SA or DA address. For example, if the SA prefix in an ACE is 2001:db8:127::/48, then the first 48 bits in the SA of a packet being compared to that ACE must be the same to allow a match. In this case, bits 49 through 128 are not compared and are termed a “wildcard”. See also Wildcard.

RACL

See Routed ACL.

RADIUS-Assigned ACL

An ACL assigned by a RADIUS server to a port to filter inbound IP traffic from a client authenticated by the server for that port. A RADIUS-assigned ACL can be configured (on a RADIUS server) to filter inbound IPv4 and IPv6 traffic (or just IPv4 traffic), regardless of whether it is switched or routed. When the client session ends, the RADIUS-assigned ACL for that client is removed from the port. See also Implicit Deny.

remark-str

The term used in ACL syntax statements to represent the variable “remark string”; a set of alphanumeric characters you can include as a remark in an ACL. A remark string allows up to 100 characters and must be delimited by single or double quotes if any spaces are included in the string.

Routed ACL (RACL)

An ACL applied to routed IPv6 traffic that is entering or leaving the switch on a given IP routing interface. See also Access Control List (ACL).

SA

The acronym for Source Address. In an IPv6 packet, this is the source IPv6 address carried in the header, and identifies the packet’s sender. This is the first of two IPv6 addresses used in an ACE to determine whether there is a match between a packet and the ACE. See also DA.

seq-#

The term used in ACL syntax statements to represent the sequence number variable used to insert an ACE within an existing list. The range allowed for sequence numbers is 1 - 2147483647.

Static Port ACL

An ACL statically configured on a specific port, group of ports, or trunk. A static port ACL filters incoming IPv6 traffic on the port, regardless of whether it is switched or routed.

VACL

See VLAN ACL (VACL).

VLAN ACL (VACL)

An ACL applied to all IPv6 traffic entering the switch on a given VLAN interface. See also Access Control List.

Wildcard

The bits in an SA or DA of a packet that are ignored when determining whether the packet is a match for a given ACE. That is, when the switch is comparing the address bits in a packet header with the address bits specified in a given IPv6 ACE, only the address bits included in the prefix length in the ACE are significant. The remaining

IPv6 Access Control Lists (ACLs)

bits—those to the right of the bits specified by the prefix length—comprise a wildcard and can be either on or off. See also Prefix Length.

Overview Types of IPv6 ACLs A permit or deny policy for IPv6 traffic you want to filter is based on source and destination IPv6 address, plus other IPv6 protocol factors such as TCP/UDP, ICMP, and DSCP.

Concurrent IPv4 and IPv6 ACLs The switches support concurrent configuration and operation of IPv4 and IPv6 ACLs. For information on IPv4 ACLs, see the Access Security Guide for your switch.

IPv6 ACL applications ACL filtering is applied to IPv6 traffic as follows: •

Routed ACL (RACL)—on a VLAN configured with an RACL, filters: •

Routed IPv6 traffic entering or leaving the switch. (Routing can be between different VLANs or between different subnets in the same VLAN. IPv6 routing must be enabled.)



Routed IPv6 traffic having a DA on the switch itself. In Figure 3 (page 100), this is any of the IPv6 addresses shown in VLANs "A", "B", and "C". (IPv6 routing need not be enabled.)



Filters outbound traffic generated by the switch itself if the ACL is applied to outbound traffic.



VLAN ACL (VACL): On a VLAN configured with a VACL, filters inbound IPv6 traffic, regardless of whether it is switched or routed.On a multinetted VLAN, this includes inbound IPv6 traffic from any subnet.



Static port ACL: Filters inbound IPv6 traffic on the port.



RADIUS-assigned ACL: On a port having an ACL assigned by a RADIUS server to filter an authenticated client's traffic, filters inbound IPv4 and IPv6 traffic (or IPv4-only traffic) from that client For information on RADIUS-assigned ACLs, see chapter "Configuring RADIUS Server Support for Switch Services" in the latest Access Security Guide for your switch.

RACL applications RACLs filter routed IPv6 traffic entering or leaving the switch on VLANs configured with the "in" and/or "out" ACL option: vlan vid ipv6 access-group identifier [ in | out | vlan ] interface tunnel tunnel-id ipv6 access-group identifier [ in | out ]

Overview

99

Example 59 RACL filter applications on routed IPv6 Traffic In Figure 3 (page 100): •

You would assign either an inbound ACL on VLAN 1 or an outbound ACL on VLAN 2 to filter a packet routed between subnets on different VLANs, that is, a packet sent from the workstation 2001:db8:0:111::2 on VLAN 1 to the server at 2001:db8:0:222::25 on VLAN 2. (An outbound ACL on VLAN 1 or an inbound ACL on VLAN 2 would not filter the packet.)



Where multiple subnets are configured on the same VLAN, you can use either inbound or outbound ACLs to filter routed IPv6 traffic between the subnets on the VLAN if the traffic source and destination IP addresses are on devices external to the switch.

Figure 3 RACL filter applications on routed IPv6 Traffic

NOTE: The switch allows one inbound IPv6 RACL assignment and one outbound IPv6 RACL assignment configured per IP routing interface. This is in addition to any other IPv6 ACL assigned to the IP routing interface or to any ports on the VLAN. You can use the same RACL or different RACLs to filter inbound and outbound routed IPv6 traffic on an IP routing interface. IPv6 RACLs do not filter traffic that remains in the same subnet from source to destination (switched traffic) unless the destination address (DA) or source address (SA) is on the switch itself.

VACL applications IPv6 VACLs filter traffic entering the switch on a VLAN configured with the "VLAN" ACL option: vlan vid ipv6 access-group vacl-identifier vlan

100 IPv6 Access Control Lists (ACLs)

Example 60 VACL filter applications on IPv6 traffic In Figure 4 (page 101) ,you would assign a VACL to VLAN 2 to filter all inbound switched or routed IPv6 traffic received from clients on the 2001:db8 :0:222:: network. In this instance, routed IPv6 traffic received on VLAN 2 from VLANs 1 or 3 would not be filtered by the VACL on VLAN 2. Figure 4 Example of VACL filter applications on IPv6 traffic entering the switch

NOTE: The switch allows one IPv6 VACL assignment configured per VLAN. This is in addition to any other IPv6 ACL applications assigned to the IP routing interface or to ports in the VLAN.

IPv6 static port ACL applications An IPv6 static port ACL filters IPv6 traffic inbound on the designated ports, regardless of whether the traffic is switched or routed. An IPv6 static port ACL filters IPv6 traffic inbound on the designated ports.

RADIUS-assigned (dynamic) port ACL applications NOTE: Beginning with software release K.14.01, IPv6 support is available for RADIUS-assigned port ACLs configured to filter inbound IPv4 and IPv6 traffic from an authenticated client. Also, the implicit deny in RADIUS-assigned ACLs applies to both IPv4 and IPv6 traffic inbound from the client. For information on enabling RADIUS-assigned ACLs, see chapter "Configuring RADIUS Support for Switch Services" in this guide. Dynamic (RADIUS-assigned) port ACLs are configured on RADIUS servers and can be configured to filter IPv4 and IPv6 traffic inbound from clients authenticated by such servers. For example, in Figure 4 (page 101), client "A" connects to a given port and is authenticated by a RADIUS server. Because the server is configured to assign a dynamic ACL to the port, the IPv4 and IPv6 traffic inbound on the port from client "A" is filtered. See also “Operating notes for IPv6 applications” (page 102). Effect of RADIUS-assigned ACLs when multiple clients are using the same port Some network configurations may allow multiple clients to authenticate through a single port where a RADIUS server assigns a separate, RADIUS-assigned ACL in response to each client's authentication on that port. In such cases, a given client's inbound traffic is allowed only if the Overview

101

RADIUS authentication response for that client includes a RADIUS-assigned ACL. Clients authenticating without receiving a RADIUS-assigned ACL are immediately de-authenticated. For example, in Figure 5 (page 102), clients A through D authenticate through the same port (B1) on an HP switch running software release K.14.01 or greater. Figure 5 Example of Multiple Clients Authenticating Through a Single Port HP Switch Running K.14.01 or Greater

LAN

RADIUS Server

Port B1 Unmanaged Switch

Client A

Client D Client B

Client C

In this case, the RADIUS server must be configured to assign an ACL to port B1 for any of the authorized clients authenticating on the port. 802.1X user-based and port-based applications User-Based 802.1X access control allows up to 32 individually authenticated clients on a given port. Port-Based access control does not set a client limit and requires only one authenticated client to open a given port (and is recommended for applications where only one client at a time can connect to the port). •

If you configure 802.1X user-based security on a port and the RADIUS response includes a RADIUS-assigned ACL for at least one authenticated client, the RADIUS response for all other clients authenticated on the ports must also include a RADIUS-assigned ACL. Inbound IP traffic on the port from a client that authenticates without receiving a RADIUS-assigned ACL is dropped and the client de-authenticated.



Using 802.1X port-based security on a port where the RADIUS response to a client authenticating includes a RADIUS-assigned ACL, different results can occur, depending on whether any additional clients attempt to use the port and whether these other clients initiate an authentication attempt. This option is recommended for applications where only one client at a time can connect to the port, and not recommended for instances where multiple clients may access the same port at the same time. For more information, see "802.1X Port-Based Access Control" in the chapter titled "Configuring Port-Based and User-Based Access Control (802.1X)" in the latest Access Security Guide for your switch.

Operating notes for IPv6 applications •

For RADIUS ACL applications using software release K.14.01 or greater, the switch operates in a dual-stack mode, and a RADIUS-assigned ACL filters both IPv4 and IPv6 traffic. At a minimum, a RADIUS-assigned ACL automatically includes the implicit deny for both IPv4 and IPv6 traffic. Thus, an ACL configured on a RADIUS server to filter IPv4 traffic also denies inbound IPv6 traffic from an authenticated client unless the ACL includes ACEs that permit the desired IPv6 traffic. The reverse is true for a dynamic ACL configured on RADIUS server to filter IPv6 traffic. (ACLs are based on the MAC address of the authenticating client.) See

102 IPv6 Access Control Lists (ACLs)

chapter "Configuring RADIUS Server Support for Switch Services" in the latest Access Security Guide for your switch. •

To support authentication of IPv6 clients: •

The VLAN to which the port belongs must be configured with an IPv6 address.



Connection to an IPv6-capable RADIUS server must be supported.



For 802.1X or MAC authentication methods, clients can authenticate regardless of their IP version (IPv4 or IPv6).



For the web authentication method, clients must authenticate using IPv4. However, this does not prevent the client from using a dual stack, or the port receiving a RADIUS-assigned ACL configured with ACEs to filter IPv6 traffic.



The RADIUS server must support IPv4 and have an IPv4 address. RADIUS clients can be dual stack, IPv6-only, or IPv4-only.



802.1X rules for client access apply to both IPv6 and IPv4 clients for RADIUS-assigned ACLs. See “802.1X user-based and port-based applications” (page 102).

Multiple ACL assignments on an interface The switch simultaneously supports IPv6, IPv4, and RADIUS-assigned ACLs on the same interface (subject to internal resource availability). This means that traffic on a port belonging to a given VLAN "X" can simultaneously be subject to all of the ACLs listed in Table 16 (page 103). Table 16 Per-interface multiple ACL assignments ACL type

ACL application

RADIUS-assigned (dynamic) ACLs

One port-based ACL (for first client to authenticate on the port) or up to 32 user-based ACLs (one per authenticated client) NOTE: If one or more user-based, RADIUS-assigned ACLs are assigned to a port, the only traffic allowed inbound on the port is from authenticated clients.

IPv6 static ACLs

One static VACL for IPv6 traffic for VLAN "X" entering the switch through the port. One static port ACL for IPv6 traffic entering the switch on the port. One inbound and one outbound RACL filtering routed IPv6 traffic moving through the port for VLAN "X." (Also applies to inbound, switched traffic on VLAN "X" that has a destination on the switch itself.)

IPv4 static ACLs

One static VACL for IPv4 traffic for VLAN "X" entering the switch through the port. One static port ACL for any IPv4 traffic entering the switch on the port. One connection-rate ACL for inbound IPv4 traffic for VLAN "X" on the port (if the port is configured for connection-rate filtering). One inbound and one outbound RACL filtering routed IPv4 traffic moving through the port for VLAN "X". (Also applies to inbound, switched traffic on VLAN "X" that has a destination on the switch itself.)

About filtering inbound traffic with multiple ACLS When traffic inbound on a port is subject to multiple ACL assignments, and a RADIUS-assigned, user-based ACL is present, this traffic must satisfy the following conditions to be permitted on the switch: 1

Originate with an authenticated client associated with the RADIUS-assigned ACL (if present).

2

Be permitted by the RADIUS-assigned ACL (if present). Includes both IPv4 and IPv6 traffic—unless the ACL is configured to exclude (drop) IPv6 traffic.

3

For IPv4-only traffic, be permitted by connection-rate ACL filtering.

Overview 103

4

Be permitted by a VACL configured on a VLAN to which the port is assigned.1

5

Be permitted by a PACL assigned to the port.1

6

For IPv4 traffic only, be permitted by a RACL assigned inbound to the port, if the traffic is subject to RACL rules.Be permitted by a RACL assigned inbound to the port, if the traffic is subject to RACL rules.

1

IPv4 VACLs and PACLs ignore IPv6 traffic, and the reverse.

Filtering outbound traffic Outbound IPv4 traffic can be filtered only by a RACL assigned outbound on the port, and only if the traffic is subject to RACL rules. (Software version K.14.01 does not support IPv6 RACLs.)

Permitting traffic filtered through multiple ACLs On a given interface where multiple ACLs apply to the same traffic, a packet having a match with a deny ACE in any applicable ACL on the interface (including an implicit deny any any) is dropped. For example, suppose the following is true: •

Ports 10 and 12 belong to VLAN 100.



A static port ACL filtering inbound IPv6 traffic is configured on port 10.



A VACL (with a different set of ACEs) is configured on VLAN 100.



An RACL is also configured for inbound, routed traffic on VLAN 100.

An inbound, switched packet entering on port 10, with a destination on port 12, will be screened first by the VACL and then by the static port ACL and the RACL. A match with a deny action (including an implicit deny) in any of the applicable ACLs causes the switch to drop the packet. If the packet has a match with explicit deny ACEs in multiple ACLs and the log option is included in these ACEs, a log event for that denied packet occurs in each ACL where there is an applicable "deny" ACE. Note that logging can also be enabled for matches with "permit" ACEs. However, in this case, suppose that VLAN 2 in Figure 6 (page 104) is configured with the following: •

A VACL permitting IPv6 traffic having a destination on the 2001:db8:0:101:: subnet



An RACL that denies inbound IPv6 traffic having a destination on the 2001:db8:0:101:: subnet

In this case, no routed IPv6 traffic received on the switch from clients on the 2001:db8:0:105:: subnet will reach the 2001:db8:0:101:: subnet, even though the VACL allows such traffic. This is because the RACL is configured with a deny ACE that causes the switch to drop the traffic regardless of whether the VACL permits the traffic. Figure 6 Order of application for multiple ACLs on an interface Subnet Mask: /64 • •

RACL on VLAN2 denies IPv6 traffic having a destination on the 2001:db8:0:101:: subnet. VACL on VLAN2 permits IPv6 traffic having a destination on the 2001:db8:0:101:: subnet.

A 2001:db8:0:101:5

B

Because the RACL on VLAN 2 2001:db8:0:105:88 denies traffic entering the switch for the 2001:db8:0:101:: subnet destination, no IPv6 traffic received D inbound from clients on the 2001:db8:0:105:: subnet will reach the 2001:db8:0:101:: subnet, even 2001:db8:0:125::22 though the VACL permits this traffic.

104 IPv6 Access Control Lists (ACLs)

Switch with IPv6 Routing

VLAN 1 2001:db8:0:101::1 (One Subnet) VLAN 2 with a VACL and an RACL 2001:db8:0:105::1 VLAN 3 (Multiple Subnets) 2001:db8:0:120::1 2001:db8:0:125::1

C

2001:db8:0:105:99

E

2001:db8:0:120:33

NOTE: Software release K.15.01 supports connection-rate ACLs for inbound IPv4 traffic, but not for IPv6 traffic. In cases where an RACL and any type of port or VLAN ACL are filtering traffic entering the switch, the switched traffic explicitly permitted by the port or VLAN ACL is not filtered by the RACL (except when the traffic has a destination on the switch itself). However, routed traffic explicitly permitted by the port or VLAN ACL (and switched traffic having a destination on the switch itself) must also be explicitly permitted by the RACL, or it will be dropped. A switched packet is not affected by an outbound RACL assigned to the VLAN on which the packet exits from the switch. Beginning with software release K.14.01, static ACL mirroring and static ACL rate-limiting are deprecated in favor of classifier-based mirroring and rate-limiting features that do not use ACLs. If ACL mirroring or ACL rate-limiting are already configured in a switch running software version K.13.xx, downloading and booting from release K.14.01 or greater automatically modifies the deprecated configuration to conform to theclassifier-based mirroring and rate-limiting supported in release K.14.01 or greater. For more information on this topic, see chapter "Classifier-Based Software Configuration" in the latest Advanced Traffic Management Guide for your switch. For information on traffic mirroring, see the "Monitoring and Analyzing Switch Operation" appendix in the latest Management and Configuration Guide for your switch.

Features common to all ACL applications •

Any ACL can have multiple entries (ACEs).



You can apply any one ACL to multiple interfaces.



All ACEs in an ACL configured on the switch are automatically sequenced (numbered). For an existing ACL, entering an ACE without specifying a sequence number automatically places the ACE at the end of the list. Specifying a sequence number inserts the ACE into the list at the specified sequential location. •

Automatic sequence numbering begins with "10" and increases in increments of 10. You can renumber the ACEs in an ACL and also change the sequence increment between ACEs.



The CLI remark command option allows you to enter a separate comment for each ACE.



A source or destination IPv6 address and a prefix length, together, can define a single host, a range of hosts, or all hosts.



Every ACL populated with one or more explicit ACEs automatically includes an implicit deny as the last entry in the list. The switch applies this action to packets that do not match other criteria in the ACL.



In any ACL, you can apply an ACL log function to ACEs that have an explicit "deny" or “permit” action. (The logging occurs when there is a match on a "deny" or “permit” ACE that includes the log keyword.) The switch sends ACL logging output to syslog, if configured, and optionally, to a console session.

You can create ACLs for the switch configuration using either the CLI or a text editor. The text-editor method is recommended when you plan to create or modify an ACL that has more entries than you can easily enter or edit using the CLI alone (see “Creating or editing an ACL offline” (page 150)).

Overview 105

General steps for planning and configuring ACLs 1.

Identify the ACL action to apply. Determine the best points at which to apply specific ACL controls. For example, you can improve network performance by filtering unwanted IPv6 traffic at the edge of the network instead of in the core. Also, on the switch itself, you can improve performance by filtering unwanted IPv6 traffic where it is inbound to the switch instead of outbound.

2.

3.

4. 5. 6. 7.

Traffic source

ACL application

IPv6 traffic from a specific, authenticated client

RADIUS-assigned ACL for inbound IPv6 traffic from an authenticated client on a port For more information, see chapter "Configuring RADIUS Server Support for Switch Services" in the latest version of the Access Security Guide for your switch. See also the documentation for your RADIUS server.

IPv6 traffic entering the switch on a specific port

Static port ACL (static-port assigned) for inbound IPv6 traffic on a port from any source

Switched or routed IPv6 traffic entering the switch on a specific VLAN

VACL (VLAN ACL)

Routed IPv6 traffic entering or leaving the switch on a specific VLAN

RACL (routed ACL)

Identify the IPv6 traffic types to filter: •

The SA and/or the DA of IPv6 traffic you want to permit or deny; this can be a single host, a group of hosts, a subnet, or all hosts.



IPv6 traffic of a specific protocol type (0 to 255).



TCP traffic (only) for a specific TCP port or range of ports, including optional control of connection traffic based on whether the initial request should be allowed.



UDP traffic (only) or UDP traffic for a specific UDP port.



ICMP traffic (only) or ICMP traffic of a specific type and code.



Any of the above with specific DSCP settings.

Design the ACLs for the control points (interfaces) you have selected. Where you are using explicit "deny" or “permit” ACEs, you can optionally use the ACL logging feature for notification that the switch is denying unwanted packets, or permitting packets that you want to go through. Configure the ACLs on the selected switches. Assign the ACLs to the interfaces you want to filter, using the ACL application (static port ACL or VACL) appropriate for each assignment. For RADIUS-assigned ACLs, see page 106. If you are using a routed ACL (RACL), ensure that IPv6 routing is enabled on the switch. Test for desired results.

For more details on ACL planning considerations, see “Configuring and assigning an IPv6 ACL” (page 115).

106 IPv6 Access Control Lists (ACLs)

IPv6 routing: To activate an IPv6 RACL to screen inbound traffic for routing between subnets, assign the RACL to the statically configured VLAN on which the traffic enters the switch. Also, ensure that IPv6 routing is enabled. Similarly, to activate an IPv6 RACL to screen routed, outbound traffic, assign the RACL to the statically configured VLAN on which the traffic exits from the switch. Inbound traffic having a destination IPv6 address on the routing switch itself are screened by an IPv6 RACL that is configured to screen inbound traffic, regardless of whether IPv6 routing is enabled. ACLs do not screen outbound traffic generated by the routing switch itself.

IPv6 ACL operation An ACL is a list of one or more ACEs, where each ACE consists of a matching criteria and an action (permit or deny). An ACL applies only to the switch in which it is configured. ACLs operate on assigned interfaces, and offer these traffic filtering options: •

IPv6 traffic inbound on a port.



IPv6 traffic inbound on a VLAN.



Routed IPv6 traffic entering or leaving the switch on a VLAN. (ACLs do not screen traffic at the internal point where traffic moves between VLANs or subnets within the switch.) See “IPv6 ACL applications” (page 99).

The following table lists the range of interface options: Interface

ACL application

Application point

Port

Static port ACL (switch configured)

Inbound on the switch port Inbound IPv6 traffic

RADIUS-assigned ACL This chapter describes ACLs statically configured on the switch. For information on RADIUS-assigned ACLs, see chapter "Configuring RADIUS Server Support for Switch Services"in the latest version of the Access Security Guide for your switch.

Inbound on the switch port Inbound IPv6 traffic from used by authenticated the authenticated client client

VLAN

VACL

Entering the switch on the VLAN

Inbound IPv6 traffic

IP routing interface (VLAN or tunnel)

RACL

Entering the switch on the VLAN

Routed IPv6 traffic entering the switch and IPv6 traffic with a destination on the switch itself

Exiting from the switch on the VLAN

Routed IPv6 traffic exiting from the switch

Supports one inbound and/or one outbound RACL. When both are used, one RACL can be assigned to filter both inbound and outbound, or different RACLs can be assigned to filter inbound and outbound.

Filter action

NOTE: After you assign an ACL to an interface, the default action on the interface is to implicitly deny any IPv6 traffic that is not specifically permitted by the ACL. (This applies only in the direction of traffic flow filtered by the ACL.)

IPv6 ACL operation 107

Packet-filtering process When an ACL filters a packet, it sequentially compares each ACE's filtering criteria to the corresponding data in the packet until it finds a match. The action indicated by the matching ACE (deny or permit) is then performed on the packet. Figure 7 Packet-filtering process in an ACL with N entries (ACEs) Test a packet against criteria in first ACE.

Is there a match?

1. If a match is not found with the first ACE in an ACL, the switch proceeds to the next ACE and so on. Yes

Perform action (permit or deny).

End

No Test the packet against criteria in second ACE

Is there a match?

Yes

Perform action (permit or deny).

End

No

Test packet against criteria in Nth ACE.

Is there a match?

Yes

Perform action (permit or deny).

End

2. If a match with an explicit ACE is subsequently found, the packet is either permitted (forwarded) or denied (dropped), depending on the action specified in the matching ACE. In this case the switch ignores all subsequent ACEs in the ACL. 3. If a match is not found with any explicit ACE in the ACL, the switch invokes the Implicit Deny at the end of every ACL, and drops the packet. Note: If the list includes an ACE configured with Permit Any forwarding, no packets can reach the Implicit Deny at the end of the list. Also, placing an ACE with Permit Any forwarding at any point in an ACL defeats the purpose of any subsequent ACEs in the list.

No Deny the packet (invoke an Implicit Deny).

End

NOTE: The order in which an ACE occurs in an ACL is significant. For example, if an ACL contains six ACEs, but the first ACE allows "Permit Any" forwarding, the ACL permits all IPv6 traffic, and the remaining ACEs in the list do not apply, even if they have a match with any traffic permitted by the first ACE. Packet-filtering Suppose you want to configure an ACL (with an ID of "Test-02") to invoke these policies for IPv6 traffic entering the switch on VLAN 100: 1. Permit inbound IPv6 traffic from 2001:db8:0:fb::11:42. 2. Deny only the inbound Telnet traffic from 2001:db8:0:fb::11:101. 3. Permit inbound IPv6 traffic from 2001:db8:0:fb::11:101. 4. Permit only inbound Telnet traffic from 2001:db8:0:fb::11:33. 5. Deny any other inbound IPv6 traffic.

108 IPv6 Access Control Lists (ACLs)

The following ACL, when assigned to filter inbound traffic on VLAN 100, supports the above case: Example 61 How an ACL filters packets ipv6 access-list "Test-02" 10 permit ipv6 2001:db8:0:fb::11:42/128 ::/0 20 deny tcp 2001:db8:0:fb::11:101/128 eq 23 ::/0 30 permit ipv6 2001:db8:0:fb::11:101/128 ::/0 40 permit tcp 2001:db8:0:fb::11:33/128 ::/0 eq 23 Implicit Deny Any Any

Line 10 Permits IPv6 traffic from 2001:db8:0:fb::11:42. Packets matching this criterion are permitted and will not be compared to any later ACE in the list. Packets not matching this criterion will be compared to the next entry in the list. Line 20 Denies IPv6 Telnet traffic from 2001:db8:0:fb::11:101. Packets matching this criterion are dropped and are not compared to later criteria in the list. Packets not matching this criterion are compared to the next entry in the list. Line 30 Permits IPv6 traffic from 2001:db8:0:fb::11:101. Packets matching this criterion will be permitted and will not be compared to any later criteria in the list. Because this entry comes after the entry blocking Telnet traffic from this same address, there will not be any Telnet packets to compare with this entry; they have already been dropped as a result of matching the preceding entry. Line 40 Permits IPv6 Telnet traffic from 2001:db8:0:fb::11:33. Packets matching this criterion are permitted and are not compared to any later criteria in the list. Packets not matching this criterion are compared to the next entry in the list. “Implicit Deny Any Any” This entry does not appear in an actual ACL, but is implicit as the last entry in every IPv6 ACL. Any IPv6 packets that do not match any of the criteria in the preceding ACL entries will be denied (dropped) from the VLAN. It is important to remember that ACLs configurable on the switch include an implicit deny ipv6 any any. That is, IPv6 packets that the ACL does not explicitly permit or deny will be implicitly denied, and therefore dropped instead of forwarded on the interface. If you want to preempt the implicit deny so that packets not explicitly denied by other ACEs in the ACL will be permitted, insert an explicit permit ipv6 any any as the last ACE in the ACL. Doing so permits any packet not explicitly denied by earlier entries. (Note that this solution would not apply in the preceding example, where the intention is for the switch to forward only the explicitly permitted packets entering the switch on VLAN 100.)(Note that this solution does not apply in the preceding example, where the intention is for the switch to forward only explicitly permitted packets routed on VLAN 12.)

Planning an ACL application Before creating and implementing ACLs, define the policies you want your ACLs to enforce and understand how the ACL assignments will impact your network users.

Planning an ACL application 109

NOTE: IPv6 traffic entering the switch on a given interface is filtered by the ACLs configured for inbound traffic on that interface. For this reason, an inbound packet is denied (dropped) if it has a match with an implicit (or explicit) deny ipv6 any any in any of the inbound ACLs applied to the interface. (This does not apply to IPv6 traffic leaving the switch, because only one type of ACL—RACL—can be applied to outbound traffic, and only to routed IPv6 traffic.) See “Multiple ACL assignments on an interface” (page 103).

IPv6 traffic management and improved network performance You can use ACLs to block IPv6 traffic from individual hosts, workgroups, or subnets, and to block access to VLANs, subnets, devices, and services. Traffic criteria for ACLs include: •

Switched IPv6 traffic



Switched and/or routed IPv6 traffic



IPv6 traffic of a specific protocol type (0 to 255)



TCP traffic (only) for a specific TCP port or range of ports, including optional control of connection traffic based on whether the initial request should be allowed



UDP traffic (only) or UDP traffic for a specific UDP port



ICMP traffic (only) or ICMP traffic of a specific type and code



Any of the above with specific precedence and/or ToS settings

Depending on the source, destination, or both of a given IPv6 traffic type, you must also determine the ACL applications (VACL or static port ACL) needed to filter the traffic on the applicable switch interfaces. Depending on the source and/or destination of a given IPv6 traffic type, you must also determine the ACL applications (RACL, VACL, or static port ACL) needed to filter the traffic on the applicable switch interfaces. Answering the following questions can help you to design and properly position ACLs for optimum network usage: •

What are the logical points for minimizing unwanted IPv6 traffic, and what ACL applications should be used? In many cases, it makes sense to prevent unwanted IPv6 traffic from reaching the core of your network by configuring ACLs to drop unwanted IPv6 traffic at or close to the edge of the network. (The earlier in the network path you can deny unwanted traffic, the greater the benefit for network performance.)



From where is the traffic coming? The source and destination of IPv6 traffic you want to filter determines the ACL application to use (VACL, static port ACL, and RADIUS-assigned ACL). The source and destination of IPv6 traffic you want to filter determines the ACL application to use (RACL, VACL, static port ACL, and RADIUS-assigned ACL).



What IPv6 traffic should you explicitly deny? Depending on your network size and the access requirements of individual hosts, this can involve creating a large number of ACEs in a given ACL (or a large number of ACLs), which increases the complexity of your solution.



What IPv6 traffic can you implicitly deny by taking advantage of the implicit deny ipv6 any any to deny IPv6 traffic that you have not explicitly permitted? This can reduce the number of entries needed in an ACL.



What IPv6 traffic should you permit? In some cases, you will need to explicitly identify permitted IPv6 traffic. In other cases, depending on your policies, you can insert an ACE with "permit any" forwarding at the end

110

IPv6 Access Control Lists (ACLs)

of an ACL. This means that IPv6 traffic not specifically matched by earlier entries in the list will be permitted.

Security ACLs can enhance security by blocking IPv6 traffic carrying an unauthorized source IPv6 address. This can include: •

Blocking access from specific devices or interfaces (port or VLAN)



Blocking access to or from subnets in your network



Blocking access to or from the internet



Blocking access to sensitive data storage or restricted equipment



Preventing specific TCP, UDP, and ICMP traffic types, including unauthorized access using functions such as Telnet and SSH

You can also enhance switch management security by using ACLs to block IPv6 traffic that has the switch itself as the DA. CAUTION: ACLs can enhance network security by denying selected IPv6 traffic, and they can serve as one aspect of maintaining network security. However, because ACLs do not provide user or device authentication, or protection from malicious manipulation of data carried in IPv6 packet transmissions, they should not be relied upon for a complete security solution. NOTE:

ACLs in the switches do not filter non-IPv6 traffic such as IPv4, AppleTalk, and IPX packets.

Guidelines for planning the structure of an ACL After determining the ACL application (VACL or static port ACL) to use at a particular point in your network, determine the order in which to apply individual ACEs to filter IPv6 traffic. After determining the ACL application (RACL, VACL, or static port ACL) to use at a particular point in your network, determine the order in which to apply individual ACEs to filter IPv6 traffic. For information on ACL applications, see “IPv6 ACL applications” (page 99). •

The sequence of ACEs is significant. When the switch uses an ACL to determine whether to permit or deny a packet on a particular VLAN, it compares the packet to the criteria specified in the individual ACEs in the ACL, beginning with the first ACE in the list and proceeding sequentially until a match is found. When a match is found, the switch applies the indicated action (permit or deny) to the packet.



The first match in an ACL dictates the action on a packet. Subsequent matches in the same ACL are ignored. However, if a packet is permitted by one ACL assigned to an interface, but denied by another ACL assigned to the same interface, the packet will be denied on the interface.



On any ACL, the switch implicitly denies IPv6 packets that are not explicitly permitted or denied by the ACEs configured in the ACL. If you want the switch to forward a packet for which there is not a match in an ACL, append an ACE that enables permit any forwarding as the last ACE in an ACL. This ensures that no packets reach the implicit deny case for that ACL.



Generally, you should list ACEs from the most specific (individual hosts) to the most general (subnets or groups of subnets), unless doing so permits IPv6 traffic that you want dropped. For example, an ACE allowing a series of workstations to use a specialized printer should occur earlier in an ACL than an entry used to block widespread access to the same printer.

Planning an ACL application

111

ACL configuration and operating rules RACLs and routed IPv6 traffic Except for IPv6 traffic with a DA on the switch itself, RACLs filter only routed IPv6 traffic that is entering or leaving the switch on a given VLAN. Thus, if routing is not enabled on the switch, there is no routed IPv6 traffic for RACLs to filter. For more information on IPv6 routing, see the following: •

“IPv6 Routing Basics” (page 173)



“IPv6 Static Routing” (page 187)



“IPv6 Router Advertisements” (page 193)



“DHCPv6-Relay” (page 211)



“OSPFv3 Routing” (page 218)



“IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels” (page 273)



“IPv6 Diagnostic and Troubleshooting” (page 288)

VACLs and switched or routed IPv6 traffic A VACL filters IPv6 traffic entering the switch on the VLANs to which it is assigned. VACLs A VACL filters IPv6 traffic entering the switch on the VLANs to which it is assigned. Static port ACLs A static port ACL filters IPv6 traffic entering the switch on the ports or trunks to which it is assigned. Per switch ACL limits for all ACL types At a minimum, an ACL must have one, explicit "permit" or "deny" ACE. You can configure up to 2048 ACLs (IPv4 and IPv6 combined). Total ACEs in all ACLs depends on the combined resource usage by ACL and other features. For more information on this topic, see “Monitoring shared resources” (page 171). Implicit deny In any static ACL, the switch implicitly (automatically) applies an implicit deny ipv6 any any that does not appear in show listings. This means that the ACL denies any packet it encounters that does not have a match with an entry in the ACL. Thus, if you want an ACL to permit any IPv6 packets that you have not expressly denied, you must enter a permit ipv6 any any as the last ACE in an ACL. Because, for a given packet, the switch sequentially applies the ACEs in an ACL until it finds a match, any packet that reaches a permit ipv6 any any entry is permitted and does not encounter the implicit "Deny" ACE the switch automatically includes at the end of the ACL. For an example, see Example 65 (page 119). For implicit deny operation in RADIUS-assigned (dynamic) ACLs, see chapter "Configuring RADIUS Server Support for Switch Services" in the latest Access Security Guide for your switch. Explicitly permitting IPv6 traffic Entering a permit ipv6 any any ACE in an ACL permits the IPv6 traffic not previously permitted or denied by that ACL. Any ACEs listed after that point do not have any effect. Explicitly denying IPv6 traffic Entering a deny ipv6 any any ACE in an ACL denies IPv6 traffic not previously permitted or denied by that ACL. Any ACEs listed after that point have no effect. Replacing one ACL with another of the same type For a specific interface, the most recent ACL assignment using a given application replaces any previous ACL assignment using the same application on the same interface. For example, if you assign a VACL named "Test-01" to filter inbound IPv6 traffic on VLAN 20, but later you 112

IPv6 Access Control Lists (ACLs)

assign another VACL named "Test-02" to filter inbound IPv6 traffic on this same VLAN, VACL "Test-02" replaces VACL "Test-01" as the ACL to use. For example, if you assign an RACL named "Test-01" to filter inbound routed IPv6 traffic on VLAN 20, but later you assign another RACL named "Test-02" to filter inbound routed IPv6 traffic on this same VLAN, RACL "Test-02" replaces RACL "Test-01" as the ACL to use. Static port ACLs These are applied per port, per port list, or per static trunk. Adding a port to a trunk applies the trunk's ACL configuration to the new member. If a port is configured with an ACL, the ACL must be removed before the port is added to the trunk. In addition, removing a port from an ACL-configured trunk removes the ACL configuration from that port. VACLs These filter IPv6 traffic entering the switch through any port belonging to the designated VLAN. VACLs do not filter IPv6 traffic leaving the switch or being routed from another VLAN. VACLs operate on static VLANs You can assign an ACL to any VLAN that is statically configured on the switch. ACLs do not operate with dynamic VLANs. VACLs and RACLs operate on static VLANs You can assign an ACL to any VLAN that is statically configured on the switch. ACLs do not operate with dynamic VLANs. A VACL affects all physical ports in a static VLAN A VACL assigned to a VLAN applies to all physical ports on the switch belonging to that VLAN, including ports that have dynamically joined the VLAN. A VACL or RACL affects all physical ports in a static VLAN A VACL or RACL assigned to a VLAN applies to all physical ports on the switch belonging to that VLAN, including ports that have dynamically joined the VLAN. RACLs screen routed IPv6 traffic entering or leaving the switch on a given VLAN interface This means that the following traffic is subject to ACL filtering: •

IPv6 traffic arriving on the switch through one VLAN and leaving the switch through another VLAN.



IPv6 traffic arriving on the switch through one subnet and leaving the switch through another subnet within the same, multinetted VLAN. Filtering the desired, routed IPv6 traffic requires assigning an RACL to screen IPv6 traffic inbound or outbound on the appropriate VLANs. In the case of a multinetted VLAN, it means that IPv6 traffic inbound from different subnets in the same VLAN is screened by the same inbound RACL, and IPv6 traffic outbound from different subnets is screened by the same outbound RACL. See Figure 3 (page 100).

RACLs do not filter switched IPv6 traffic unless the switch itself is the SA or DA RACLs do not filter IPv6 traffic moving between ports belonging to the same VLAN or subnet (in the case of a subnetted VLAN). (IPv6 traffic moving between ports in different subnets of the same VLAN can be filtered by a RACL.) NOTE:

RACLs do filter routed or switched IPv6 traffic having an SA or DA on the switch itself.

How an ACE uses a prefix to screen packets for SA and DA matches For an IPv6 ACL, a match with a packet occurs when both the protocol and the SA/DA configured in a given ACE within the ACL are a match with the same criteria in a packet being filtered by the ACL. In IPv6 ACEs, prefixes define how many leading bits in the SA and DA to use for determining a match. That is, the switch uses IPv6 prefixes in CIDR format to specify how many leading bits in a Planning an ACL application

113

packet's SA and DA must be an exact match with the same bits in an ACE. The bits to the right of the prefix are "wildcards" and are not used to determine a match. Prefix

Range of applicable addresses

Examples

/0

Any IPv6 host

::/0

/ 1 — /127

All IPv6 hosts within the range defined by the number of bits in the prefix

2001:db8::/48 2001:db8::/64

/128

One IPv6 host

2001:db8::218:71ff:fec4:2f00/128

Example 62 SA/DA prefix lengths The following ACE applies to Telnet packets from a source address where the leading bits are set to 2001:db8:10:1 and any destination address where the leading bits are set to 2001:db8:10:1:218:71ff:fec. permit tcp 2001:db8:10:1::/64 eq 23 2001:db8:10:1:218:71ff:fec4::/112

“::/64” Prefix Defining the Mask for the Leading Bits in the Source Address “::/112” Prefix Defining the Mask for the Leading Bits in the Destination Address Thus, in the above example, if an IPv6 Telnet packet has an SA match with the ACE's leftmost 64 bits and a DA match with the ACE's leftmost 112 bits, there is a match and the packet is permitted. In this case, the source and destination addresses allowed are: Address

Prefix

Range of unicast addresses

Source (SA)

2001:db8:10:1

prefix ::0 to prefix :FFFF:FFFF:FFFF:FFFF

Destination (DA)

2001:db8:10:1:218:71ff:fec4

prefix :0 to prefix :FFFF

To summarize, when the switch compares an IPv6 packet to an ACE in an ACL, it uses the subnet prefixes configured with the SA and DA in the ACE to determine how many leftmost, contiguous bits in the ACE's SA and DA must be matched by the same bits in the SA and DA carried by the packet. Thus, the subnet prefixes specified with the SA and DA in an ACE determine the ranges of source and destination addresses acceptable for a match between the ACE and a packet being filtered.

114

IPv6 Access Control Lists (ACLs)

Prefix usage differences between ACLs and other IPv6 addressing For ACLs, the prefix is used to specify the leftmost bits in an address that are meaningful for a packet match. In other IPv6 usage, the prefix separates network and subnet values from the device identifier in an address. Prefix usage

Examples

Notes

For an SA or DA in the ACE belonging to an IPv6 ACL, the associated prefix specifies how many consecutive, leading bits in the address are used to define a match with the corresponding bits in the SA or DA of a packet being filtered.

2620:0:a03:e102:215:60ff:fe7a:adc0/128 All bits. Used for a specific SA or DA. 2620:0:a03:e102:215/80 The first 80 bits. Used for an SA or DA having 2620:0:a03:e102:215 in the leftmost 80 bits of an address.

For the IPv6 address assigned to a given device, the prefix defines the type of address and the network and subnet in which the address resides. In this case, the bits to the right of the prefix comprise the device identifier.

fe80::215:60ff:fe7a:adc0/64 Link-Local address with a prefix of 64 bits and a device ID of 64 bits. 2620:0:a03:e102:215:60ff:fe7a:adc0/64 Global unicast address with a prefix of 64 bits and a device ID of 64 bits.

Zero bits. Used to allow a match ::/0 with "any" SA or DA.

For an RA, the included prefix defines the network or range of networks and the subnets the router is advertising.

2620:0:a03::/48 An RA with a 48-bit prefix 2620:0:a03:e102::/64 an RA with a 64-bit prefix

For more information on RAs, see “IPv6 Router Advertisements” (page 193).

Configuring and assigning an IPv6 ACL ACL Feature

Page

Adding or Removing an ACL

128

Enabling or Disabling ACL Filtering

132

Implementing IPv6 ACLs For more information on configuring ACLs, see “Configuring and assigning an IPv6 ACL” (page 115). 1. Configure one or more ACLs. This creates and stores the ACLs in the switch configuration. 2.

3.

Assign an ACL to an interface using one of the following applications: •

RACL (routed IPv6 traffic entering or leaving the switch on a given VLAN)



VACL (IPv6 traffic entering the switch on a given VLAN)



Static port ACL (IPv6 traffic entering the switch on a given port, port list, or static trunk)

If the ACL is applied as an RACL, IPv6 routing must be enabled. Except for instances where the switch is the traffic source or destination, assigned RACLs filter IPv6 traffic only when IPv6 routing is enabled on the switch.

Configuring and assigning an IPv6 ACL

115

Permit/deny options You can use the following criteria as options for permitting or denying a packet: •

Source IPv6 address



Destination IPv6 address



IPv6 protocol options: •

All IPv6 traffic



IPv6 traffic of a specific protocol type (0 to 255)



IPv6 traffic for a specific TCP port or range of ports, including: •

Optional control of connection (established) traffic based on whether the initial request should be allowed



TCP flag (control bit) options



IPv6 traffic for a specific UDP port or range of ports



IPv6 traffic for a specific ICMP type and code



Any of the above with specific DSCP precedence or ToS settings

Carefully plan ACL applications before configuring specific ACLs. For more information on this topic, see “Configuring and assigning an IPv6 ACL” (page 115).

Overriding an implicit deny If a packet does not have a match with the criteria in any of the ACEs in the ACL, the ACL denies (drops) the packet. If you need to override the implicit deny so that a packet that does not have a match will be permitted, configure permit ipv6 any any as the last ACE in the ACL. This directs the ACL to permit (forward) packets that do not have a match with any earlier ACE listed in the ACL and prevents these packets from being filtered by the implicit deny ipv6 any any. Example 63 Overriding an implicit deny Suppose the following ACL with five ACEs is assigned to filter the IPv6 traffic from an authenticated client on a given port in the switch: 10 permit ipv6 ::/0 fe80::136:24/128 20 permit ipv6 ::/0 fe80::156:7/128 30 deny ipv6 ::/0 fe80::156:3/128 40 deny tcp ::/0 ::/0 eq 23 50 permit ipv6 ::/0 ::/0 (deny ipv6 ::/0 ::/0)

For an inbound packet with a destination IP address of FE80::156:3, the ACL: 1. Compares the packet to the first ACE first (line 10). 2. Since there is not a match with the first ACE, the ACL compares the packet to the second ACE, where there is also not a match (line 20). 3. The ACL compares the packet to the third ACE. There is an exact match, so the ACL denies (drops) the packet (line 30). 4. The packet is not compared to the fourth ACE (line 40). 5. The last line demonstrates the "deny any any" ACE implicit in every IPv6 ACL. Inbound IPv6 traffic from an authenticated client that does not have a match with any of the five explicit ACEs in this ACL will be denied by the implicit "deny any any". As shown above, the ACL tries to apply the first ACE in the list. If there is not a match, it tries the second ACE, and so on. When a match is found, the ACL invokes the configured action for that 116

IPv6 Access Control Lists (ACLs)

entry (permit or drop the packet) and no further comparisons of the packet are made with the remaining ACEs in the list. This means that when an ACE whose criteria matches a packet is found, the action configured for that ACE is invoked, and any remaining ACEs in the ACL are ignored. Because of this sequential processing, successfully implementing an ACL depends in part on configuring ACEs in the correct order for the overall policy you want the ACL to enforce. To see a flow diagram of the packet-filtering process in an ACL, see Example 63 (page 116).

ACL configuration After you enter an ACL command, you may want to inspect the resulting configuration. This is especially true where you are entering multiple ACEs into an ACL. Also, it is helpful to understand the configuration structure when using later sections in this chapter. The basic ACL structure includes four elements: 1. ACL identity This is a string of up to 64 characters specifying the ACL name. 2. 3.

Optional remark entries. One or more deny/permit list entries (ACEs): One entry per line. Element

Notes

Identifier

Alphanumeric; up to 64 characters, including spaces

Remark

Allows up to 100 alphanumeric characters, including blank spaces. (If any spaces are used, the remark must be enclosed in a pair of single or double quotes.) A remark is associated with a particular ACE and has the same sequence number as the ACE. (One remark is allowed per ACE.) See “Attaching a remark to an ACE” (page 136).

Maximum ACEs per switch

4.

The maximum number of ACEs supported by the switch is up to 3072 for IPv6 ACEs and up to 3072 for IPv4 ACEs. The maximum number of ACEs applied to a VLAN or port depends on the concurrent resource usage by multiple configured features. For more information, use the show qos|access-list resources command and/or see “Monitoring shared resources” (page 171).

Implicit deny Where an ACL is applied to an interface, it denies any packets that do not have a match with any of the ACEs explicitly configured in the list. The implicit deny does not appear in ACL configuration listings, but always functions when the switch uses an ACL to filter packets. (You cannot delete the implicit deny, but you can supersede it with a permit ipv6 any any ACE.)

ACL Configuration Structure Individual ACEs in an IPv6 ACL include: •

Optional remark statements



A permit/deny statement



Source and destination IPv6 addressing



Choice of IPv6 criteria



Optional ACL log command (for deny or permit entries)

General structure options for an IPv6 ACL ipv6 access-list identifier [ seq-# ]

Configuring and assigning an IPv6 ACL

117

[ remark remark-str ] permit | deny 0 - 255 esp ah sctp icmp SA [operator value ] DA [operator value ] [type [code] | icmp-msg ] [dscp | precedence ] ipv6 tcp SA [operator value ] DA [operator value ] [dscp codepoint | precedence] [established] [ack | fin | rst | syn] udp SA [operator value ] DA [operator value ] [log] (Allowed only with “deny” or "permit" ACEs.) . . . Implicit Deny Any Any exit

codepoint

The ACL in Example 64 (page 118) filters traffic for individual hosts in some instances and all hosts in others: Example 64 Displayed ACL configuration HP Switch# show run . . . ipv6 access-list "Sample-List-1" 10 permit ipv6 2001:db8:0:130::55/128 2001:db8:0:130::240/128 20 permit tcp ::/0 ::/0 eq 23 30 remark "ALLOWS HTTP FROM SINGLE HOST." 30 permit tcp 2001:db8:0:140::14/128 eq 80 ::/0 eq 3871 40 remark "DENIES HTTP FROM ANY TO ANY." 40 deny tcp ::/0 ::/0 eq 80 log 50 deny udp 2001:db8:0:150::44/128 eq 69 2001:db8:0:120::19/128 range 3680 3690 log 60 deny udp ::/0 2001:db8:0:150::121/128 log 70 permit ipv6 2001:db8:0:01::/56 ::/0 exit

Line

Action

10

Permits all IPv6 traffic from the host at 2001:db8:0:130::55 to the host at 2001:db8:0:130::240.

20

Permits all Telnet traffic from any source to any destination.

30

Includes a remark and permits TCP port 80 traffic received at any destination as port 3871 traffic.

40

Includes a remark and denies TCP port 80 traffic received at any destination, and causes a log message to be generated when a match occurs.

50

Denies UDP port 69 (TFTP) traffic sent from the host at 2001:db8:0:150::44 to the host at 2001:db8:0:120::19 with a destination port number in the range of 3680 to 3690 and causes a log message to be generated when a match occurs.

60

Denies UDP traffic from any source to the host at 2001:db8:0:150::121 and causes a log message to be generated when a match occurs.

70

Permits all IPv6 traffic with an SA prefix of 2001:db8:0:01/56 that is not already permitted or denied by the preceding ACEs in the ACL.

NOTE: An implicit deny IPv6 any any is automatically applied following the last line (70, in this case) and denies all IPv6 traffic not already permitted or denied by the ACEs in lines 10 through 70.

118

IPv6 Access Control Lists (ACLs)

ACL configuration factors The sequence of entries in an ACL is significant When the switch uses an ACL to determine whether to permit or deny a packet, it compares the packet to the criteria specified in the individual ACEs in the ACL, beginning with the first ACE in the list and proceeding sequentially until a match is found. When a match is found, the switch applies the indicated action (permit or deny) to the packet. This is significant because, once a match is found for a packet, subsequent ACEs in the same ACL are not applied to that packet, regardless of whether they match the packet. Example 65 ACE that permits all IPv6 traffic not implicitly denied Suppose that you have applied the ACL shown in Example 65 (page 119) to inbound IPv6 traffic on VLAN 1 (the default VLAN): ipv6 access-list "Sample-List-2" 10 deny ipv6 2001:db8::235:10 1 /128 2

::/0 3

20 deny ipv6 2001:db8::245:89/128 ::/0 30 permit tcp 2001:db8::18:100/128 2001:db8::237:1/128 40 deny tcp 2001:db8::18:100/128 ::/0 50 permit ipv6 ::/0 ::/0 4 (Implicit deny ipv6 any any) 5 exit 1 2 3 4 5

Source Address Prefix length Destination Address and Prefix Length (Specifies Any IPv6 Destination) After the last explicit ACE there is always an Implicit Deny. However, in this case it will not be used because the last permit ipv6 ACL allows all IPv6 packets that earlier ACEs have not already permitted or denied.

Line # N/A

Action Shows IPtype (IPv6) and ID (Sample-List-2).

10

A packet from source address 2001:db8:235:10 will be denied (dropped). This ACE filters out all packets received from 2001:db8:235:10. As a result, IPv6 traffic from that device will not be allowed, and packets from that device will not be compared against any later entries in the list.

20

A packet from IPv6 source address 2001:db8::245:89 will be denied (dropped). This ACE filters out all packets received from 2001:db8::245:89. As the result, IPv6 traffic from that device will not be allowed, and packets from that device will not be compared against any later entries in the list.

30

A TCP packet from SA 2001:db8::18:100 with a DA of 2001:db8::237:1 will be permitted (forwarded). Since no earlier ACEs in the list have filtered TCP packets from 2001:db8::18:100 with a destination of 2001:db8::237:1, the switch will use this ACE to evaluate such packets. Any packets that meet this criteria will be forwarded. (Any packets that do not meet this TCP source-destination criteria are not affected by this ACE.)

40

A TCP packet from source address 2001:db8::18:100 to any destination address will be denied (dropped). Since, in this example, the intent is to block TCP traffic from 2001:db8::18:100 to any destination except the destination stated in the ACE at line 30, this ACE must follow the ACE at line 30. (If their relative positions were exchanged, all TCP traffic from 2001:db8::18:100 would be dropped, including the traffic for the 2001:db8::237:1 destination.)

Configuring and assigning an IPv6 ACL

119

Line #

Action

50

Any packet from any IPv6 source address to any IPv6 destination address will be permitted (forwarded). The only traffic filtered by this ACE will be packets not specifically permitted or denied by the earlier ACEs.

N/A

The implicit deny (deny ipv6 any any) is a function the switch automatically adds as the last action in all IPv6 ACLs. It denies (drops) traffic from any source to any destination that has not found a match with earlier entries in the ACL. In this example, the ACE at line 50 permits (forwards) any traffic not already permitted or denied by the earlier entries in the list, so there is no traffic remaining for action by the implicit deny function.

exit

Defines the end of the ACL.

Implied deny function In any ACL having one or more ACEs, there is always a packet match. This is because the switch automatically applies the implicit deny as the last ACE in any ACL. This function is not visible in ACL listings, but is always present; see Example 65 (page 119). This means that if you configure the switch to use an ACL for filtering either inbound or outbound traffic on a VLAN, any IPv6 packets not specifically permitted or denied by the explicit entries you create is denied by the implicit deny action. If you want to preempt the implicit deny (so that IPv6 traffic not specifically addressed by earlier ACEs in a given ACL is permitted), insert an explicit permit ipv6 any any as the last explicit ACE in the ACL.

Assignment of an ACL to an interface The switch stores ACLs in the configuration file. Until you actually assign an ACL to an interface, it is present in the configuration, but not used (and does not use any of the monitored resources described in the appendix "Monitoring Resources" in the latest version of the Management and Configuration Guide for your switch.)

Assignment of an ACL name to an interface In this case, if you subsequently create an ACL with that name, the switch automatically applies each ACE as soon as you enter it in the running-config file. Similarly, if you modify an existing ACE in an ACL you already applied to an interface, the switch automatically implements the new ACE as soon as you enter it. See “General ACL operating notes” (page 170). The switch allows up to 2048 ACLs each for IPv4 and IPv6. For example, if you configure two ACLs, but assign only one of them to a VLAN, the ACL total is two, for the two unique ACL names. If you then assign the name of an empty ACL to a VLAN, the new ACL total is three, because the switch now has three unique ACL names in its configuration. (RADIUS-based ACL resources are drawn from the IPv4 allocation). For information on switch resource use, see “Monitoring shared resources” (page 171).

Creating an ACL using the CLI You can use either the switch CLI or an offline text editor to create an ACL. This section describes the CLI method, which is recommended for creating short ACLs. To use the offline method, see “Creating or editing an ACL offline” (page 150)

General ACE rules These rules apply to all ACEs you create or edit using the CLI.

Adding or inserting an ACE in an ACL To add an ACE to the end of an ACL: 1. Use the ipv6 access-list name-str command to enter the context for a specific IPv6 ACL. (If the ACL does not already exist in the switch configuration, this command creates it.)

120 IPv6 Access Control Lists (ACLs)

2.

Enter the text of the ACE without specifying a sequence number. For example, the following pair of commands enter the context of an ACL named "List-1" and add a "permit" ACE to the end of the list. This new ACE permits the IPv6 traffic from the device at 2001:db8:0:a9:8d:100 to go to all destinations. HP Switch(config)# ipv6 access-list List-1 HP Switch(config-ipv6-acl)# permit host 2001:db8:0:a9::8d:100 any

To insert an ACE anywhere in an existing ACL: Enter the context of the ACL and specify a sequence number. For example, to insert a new ACE as line 15 between lines 10 and 20 in an existing ACL named "List-2" to deny traffic from the device at 2001:db8:0:a9::8d:77: HP Switch(config)# ipv6 access-list List-2 HP Switch(config-ipv6-acl)# deny host 2001:db8:0:a9::8d:77 any

Deleting an ACE Enter the ACL context and delete the sequence number for the unwanted ACE. (To view the sequence numbers of the ACEs in a list, use show access-list acl-name-str config.) For example, to delete the ACE at line 40 in an ACL named "List-2", enter the following commands: HP Switch(config)# ipv6 access-list List-2 config HP Switch(config-ipv6-acl)# no 40

Duplicate ACE sequence numbers Duplicate sequence numbering for ACEs are not allowed in the same ACL. Attempting to enter a duplicate ACE displays the Duplicate sequence number message.

Configuration Commands Creating an ACL and/or entering the IPv6 ACL (ipv6-acl) context This command is a prerequisite for entering or editing ACEs in an ACL. For general information about ACL commands, see “Overview of IPv6 ACLs” (page 160). For information specific to this command, see “Commands to create, enter, and configure an ACL” (page 161).

Syntax: ipv6 access-list ascii-str Places the CLI in the IPv6 ACL (ipv6-acl) context specified by the ascii-str alphanumeric identifier. This enables entry of individual ACEs in the specified ACL. If the ACL does not already exist, this command creates it. ascii-str Specifies an alphanumeric identifier for the ACL and consists of an alphanumeric string of up to 64 case-sensitive characters. If you include spaces in the string, you must enclose the string in single or double quotes. For example: "Accounting ACL". You can also use this command to access an existing ACL; see “General editing rules” (page 162).

Configuration Commands

121

Example 66 Entering the ACL context HP Switch(config)# ip access-list Sample-List HP Switch(config-ipv6-acl)#

Configuring ACEs in an ACL Configuring ACEs is done after using the ipv6 access-list ascii-str command described on page 121 to enter the IPv6 ACL (ipv6_acl) context of an ACL.

Syntax: (ipv6 acl context) [ [ [ [

deny | permit ] [ ipv6 | ipv6-protocol | ipv6-protocol-nbr ] any | hostSA | SA/prefix-length ] any | hostDA | DA/prefix-length ] dscp tos-bits | precedence ] [ log ] Appends an ACE to the end of the list of ACEs in the current ACL. In the default configuration, ACEs are automatically assigned consecutive sequence numbers in increments of 10 and can be renumbered using resequence, page 135). NOTE: To insert a new ACE between two existing ACEs in an ACL, precede deny or permit with an appropriate sequence number. See “Inserting an ACE in an existing ACL” (page 132). For a match to occur, a packet must have the source and destination IPv6 addressing criteria specified in the ACE, as well as: •

The protocol-specific criteria configured in the ACE, including any optional elements (described later in this section)



Any (optional) DSCP settings configured in the ACE

[ deny | permit ] These keywords are used in the IPv6 ACL (ipv6-acl) context to specify whether the ACE denies or permits a packet matching the criteria in the ACE, as described below. [ ipv6 | ipv6-protocol | ipv6-protocol-nbr ] Used after deny or permit to specify the packet protocol type required for a match. An ACL must include one of the following: ipv6 Any IPv6 packet. ipv6-protocol Any one of the following IPv6 protocol names: esp ah sctp icmp1tcp1udp1 ipv6-protocol-nbr The protocol number of an IPv6 packet type, such as "8" for Exterior Gateway Protocol or 121 for Simple Message Protocol. Range: 0 to 255 (For a listing of IPv6 protocol numbers and their corresponding protocol names, see the IANA protocol number assignments at www.iana.com.) [ any | host | SA | SA prefix-length ] This is the first instance of IPv6 addressing in an ACE. It follows the protocol specifier and defines the source IPv6 address (SA) a packet must carry for a match with the ACE. any Allows IPv6 packets from any IPv6 SA. 1. For TCP, UDP, and ICMP, additional (optional) criteria can be specified, as described on 125through 128.

122

IPv6 Access Control Lists (ACLs)

host SA Specifies only packets having a single address as the SA. Use this criterion when you want to match only the IPv6 packets from a single SA. SA

prefix-length Specifies packets received from one or more contiguous subnets or contiguous addresses within a single subnet. The prefix length is in CIDR format and defines the number of leftmost bits to use in determining a match. See “Using CIDR notation to enter the IPv6 ACL prefix length” (page 160). In a given ACE, the SA prefix length defines how many leftmost bits in a packet's SA must exactly match the SA configured in the ACE. Examples of prefix-length applications: •

2001:db8:0:e102::10:100/120 matches any IPv6 address in the range of 2001:db8:0:e102::10:<0100 - 01FF>



2001:db8:a0:e102::/64 matches any IPv6 address having a prefix of 2001:db8:a0:e102.



FE80::/16 matches any link-local address on an interface.

NOTE: For more information on how prefix lengths are used in IPv6 ACLs, see “How an ACE uses a prefix to screen packets for SA and DA matches” (page 113). [ any | host DA | DA/prefix-length ] This is the second instance of addressing in an IPv6 ACE. It follows the first (SA) instance, described earlier in this section, and defines the destination IPv6 address (DA) that a packet must carry to have a match with the ACE. any Allows IPv6 packets to any IPv6 DA. host DA Specifies only packets having DA as the destination address. Use this criterion when you want to match only the IPv6 packets for a single DA. DA/prefix-length Specifies packets intended for one or more contiguous subnets or contiguous addresses within a single subnet. The prefix length is in CIDR format and defines the number of leftmost bits to use in determining a match. See “Using CIDR notation to enter the IPv6 ACL prefix length” (page 160). In a given ACE, the DA prefix-length defines how many leftmost bits in a packet's DA must exactly match the DA configured in the ACE. For examples, see “Examples of prefix-length applications” (page ?). [ dscp codepoint/precedence ] This option follows the DA to include a DSCP codepoint or precedence as a matching criteria. codepoint: Supports these codepoint selection options: 0 - 63 Select a specific DSCP codepoint by entering its decimal equivalent. See “DSCP codepoints with decimal equivalents” (page 124). Assured Forwarding (AF) codepoint matches: AF

DSCPMatch

af11

001010

af12

001100 Configuration Commands

123

AF

DSCPMatch

af13

001110

af21

010010

af22

010100

af23

010110

af31

011010

af32

011100

af33

011110

af41

100010

af42

100100

af43

100110

Default Matches with the 000000 (default) DSCP. ef Expedited forwarding (EF; 000000) DSCP match. precedence Supports selection of a precedence setting in the DSCP. Option

Precedence Bits

Name

cs1

001

priority

cs2

010

immediate

cs3

011

flash

cs4

100

flash-override

cs5

101

critical

cs6

110

internet (for internetwork control)

cs7

111

network (for network control)

NOTE: The precedence criteria described in this section are applied in addition to any other selection criteria configured in the same ACE. Also, where dscp is configured in a given ACE, the established keyword and the optional TCP control bits cannot be configured. [log] This option can be used after the DA to generate an Event Log message if: •

The action is deny or permit.



There is a match.



ACL logging is enabled. See “Enabling ACL logging on the switch” (page 152).

For a given ACE, if log is used, it must be the last keyword entered. Table 17 DSCP codepoints with decimal equivalents

124

DSCP bits

Decimal

DSCP bits

Decimal

DSCP bits

Decimal

000000

0 (default)

010110

22 (32)

101011

43

000001

1

010111

23

101100

44

000010

2

011000

24

101101

45

IPv6 Access Control Lists (ACLs)

Table 17 DSCP codepoints with decimal equivalents (continued) DSCP bits

Decimal

DSCP bits

Decimal

DSCP bits

Decimal

000011

3

011001

25

101110

46 (71)

000100

4

011010

26 (42)

101111

47

000101

5

011011

27

110000

48

000110

6

011100

28 (42)

110001

49

000111

7

011101

29

110010

50

2

001000

8

011110

30 (5 )

110011

51

001001

9

011111

31

110100

52

001010

10 (12)

100000

32

110101

53

001011

11

100001

33

110110

54

001100

12 (12)

100010

34 (62)

110111

55

001101

13

100011

35

111000

56

001110

14 (22)

100100

36 (62)

111001

57

001111

15

100101

37

111010

58

010000

16

100110

38 (72)

111011

59

010001

17

100111

39

111100

60

2

010010

18 (0 )

101000

40

111101

61

010011

19

101001

41

111110

62

010100

20 (0 2)

101010

41

111111

63

1

Expedited forwarding codepoint configured by default.

2

Assured forwarding codepoint and 802.1p precedence.

Configuring TCP and UDP traffic in IPv6 ACLs An ACE designed to permit or deny TCP or UDP traffic can optionally include port number criteria for either the source, the destination, or both. Use of TCP criteria also allows the established option for controlling TCP connection traffic.

TCP Syntax: [ deny | permit ] tcp SA [ comparison-operator tcp-src-port ] DA [ comparison-operator tcp-dest-port ] [ established ] [ ack ] [ fin ] [ rst ] [ syn ]

UDP Syntax: [ deny | permit ] udp SA [ comparison-operator udp-src-port ] DA [ comparison-operator udp-dest-port ] In an IPv6 ACL using either tcp or udp as the IP packet protocol type, you can optionally apply comparison operators specifying TCP or UDP source and/or destination port numbers or ranges of numbers to further define the criteria for a match. For example: Configuration Commands

125

#deny tcp host fe80::119 eq 23 host fe80::155 established #permit tcp host 2001:db8::10.100 host 2001:db8::15:12 eq telnet #deny udp 2001:db8::ad5:1f4 host 2001:db8::ad0:ff3 range 161 162

[ comparison-operator tcp/udp-src-port ] To specify a TCP or UDP source port number in an ACE: 1. Select a comparison operator from the following list. 2. Enter the port number or a well-known port name. Comparison operators: eq tcp/udp-port-nbr "Equal To" — to have a match with the ACE entry, the TCP or UDP source port number in a packet must be equal to tcp/udp-port-nbr. gt tcp/udp-port-nbr "Greater Than" — to have a match with the ACE entry, the TCP or UDP source port number in a packet must be greater than tcp/udp-port-nbr. lt tcp/udp-port-nbr "Less Than" — to have a match with the ACE entry, the TCP or UDP source port number in a packet must be less than tcp/udp-port-nbr. neq tcp/udp-port-nbr "Not Equal" — to have a match with the ACE entry, the TCP or UDP source port number in a packet must not be equal to tcp/udp-port-nbr. range start-port-nbr end-port-nbr For a match with the ACE entry, the TCP or UDP source-port number in a packet must be in the range start-port-nbr end-port-nbr . Port number or well-known port name: Use the TCP or UDP port number required by your application. The switch also accepts these well-known TCP or UDP port names as an alternative to their port numbers: TCP bgp, dns, ftp, http, imap4, ldap, nntp, pop2, pop3, smtp, ssl, telnet UDP bootpc, bootps, dns, ntp, radius, radius-old, rip, snmp, snmp-trap, tftp To list the above names, press the [Shift]- [?] key combination after entering an operator. For a comprehensive listing of port numbers, see www.iana.org/ assignments/port-numbers. [ comparison-operator tcp-dest-port ] [ established ] [ comparison-operator udp-dest-port ] This option, if used, is entered immediately after the DA entry. To specify a TCP or UDP port number: 1. Select a comparison operator. 2. Enter the port number or a well-known port name.

126

IPv6 Access Control Lists (ACLs)

These are the same as those used with the TCP/UDP source-port options and are listed earlier in this command description. Comparison operators and well-known port names [ established ] This option applies only where TCP is the configured IPv6 protocol type. It blocks the synchronizing packet associated with establishing a new TCP connection, while allowing all other IPv6 traffic for existing connections. For example, a Telnet connect requires TCP traffic to move both ways between a host and the target device. Simply applying a deny to inbound Telnet traffic on a VLAN prevents Telnet sessions in either direction, because responses to outbound requests are blocked. However, by using the established option, inbound Telnet traffic arriving in response to outbound Telnet requests are permitted, but inbound Telnet traffic trying to establish a new connection is denied. The established and dscp options are mutually exclusive in a given ACE. Configuring established and any combination of TCP control bits in the same ACE is supported, but established must precede any TCP control bits configured in the ACE. TCP control bits In a given ACE for filtering TCP traffic you can configure one or more of these options: [ ack ] Acknowledgment [ fin ] Sender finished [ rst ] Connection reset [ syn ] TCP control bit: sequence number synchronize For more information on using TCP control bits, see RFC 793.

Filtering ICMP traffic This option allows configuring an ACE to selectively permit some types of ICMP traffic, while denying other types. An ACE designed to permit or deny ICMP traffic can optionally include an ICMP type and code value to permit or deny an individual type of ICMP packet, while not addressing other ICMP traffic types in the same ACE. As a further option, the ACE can include the name of an ICMP packet type.

Syntax: [ deny | permit ] icmp SA DA icmp-type icmp-code [ deny | permit ] icmp SA DA icmp-type-name Using icmp as the packet protocol type, you can optionally specify an individual ICMP packet type or packet type/code pair to further define the criteria for a match. This option, if used, is entered immediately after the destination IP address (DA) entry.

Configuration Commands

127

Example 67 Showing two ACEs entered in an ACL context: #permit icmp any any 1 3 #permit icmp any any destination-unreachable

[ icmp-type [ icmp-code ]] This option identifies an individual ICMP packet type as criteria for permitting or denying that type of ICMP traffic in an ACE. •

icmp-type—This value is in the range of 0 to 255 and corresponds to an ICMP packet type.



icmp-code—This value corresponds to an ICMP code for an ICMP packet type. It is optional and needed only when a particular ICMP subtype is needed as a filtering criterion. Range: 0 to 255

For example, the following ACE specifies "destination unreachable" (ICMP type 1) where "address unreachable" (3; a subtype of "destination unreachable") is the specific code. #permit icmp any any 1 3

For more information on ICMP types and codes, see the Internet Assigned Numbers Authority (IANA) website at www.iana.com and see "Internet Control Message Protocol version 6 (ICMPv6) Type Numbers." [ icmp-type-name ] These name options are an alternative to the [ icmp-type [ icmp-code ] ] methodology described above. For more information, visit the IANA website cited above. cert-path-advertise

mobile-advertise

cert-path-solicit

mobile-solicit

destination-unreachable

nd-na

echo-reply

nd-ns

echo-request

node-info

home-agent-reply

node-query

home-agent-request

packet-too-big

cert-path-advertise

mobile-advertise

inv-nd-na

parameter-problem

inv-nd-ns

redirect

mcast-router-advertise

router-advertisement

mcast-router-solicit

router-renum

mcast-router-terminate

router-solicitation

mld-done

time-exceeded

mld-query

ver2-mld-report

mld-report

Filtering routed IPv6 traffic Adding or Removing an ACL Assignment On an Interface

128

IPv6 Access Control Lists (ACLs)

You can assign the same ACL to filter both inbound and outbound routed traffic, and to filter traffic on multiple VLANs. For limits and operating rules, see “ACL configuration and operating rules” (page 112).

Syntax: [no] vlan vid ipv6 access-group identifier [ in | out ] Assigns an ACL to a VLAN as an RACL to filter routed IP traffic entering or leaving the switch on that VLAN. You can use either the global configuration level or the VLAN context level to assign or remove an RACL. vid VLAN identification number tunnel tunnel-id Tunnel Identification identifier The alphanumeric name by which the ACL can be accessed. An identifier can have up to 64 characters in Keyword for assigning the ACL to filter routed traffic entering the switch on the specified VLAN out Keyword for assigning the ACL to filter routed traffic leaving the switch on the specified VLAN NOTE: The switch allows you to assign an "empty" ACL to a VLAN. In this case, if you later populate the empty ACL with one or more ACEs for that same identifier, the ACL automatically becomes active on the assigned VLAN. Also, where a given ACL is assigned to an interface, if you delete the ACL from the running configuration without also using the no form of this command to remove the assignment to the interface, the ACL becomes "empty," but remains assigned to the interface and continues to exist (as an empty ACL) in the running configuration. In this case, if you later repopulate the ACL with an explicit ACE, the ACL immediately reactivates and begins filtering traffic (which includes use of the implicit deny). See also "Notes" under “Deleting an ACL” (page 132).

Configuration Commands

129

Example 68 Methods for enabling and disabling RACLs HP Switch(config)# vlan 20 ipv6 access-group List-001 in 1 HP Switch(config)# vlan 20 HP Switch(vlan-20)# ipv6 access-group List-005 out 2 HP Switch(vlan-20)# exit HP Switch(config)# no vlan 20 ipv6 access-group List-001 in 3 HP Switch(config)# vlan 20 HP Switch(vlan-20)# no ipv6 access-group List-005 out 4 HP Switch(vlan-20)# exit 1 2 3 4

Enables an RACL from the Global Configuration Level Enables an RACL from a VLAN Context Disables an RACL from the Global Configuration Level Disabling an RACL from a VLAN Context

Filtering routed or switched IPv6 traffic inbound on a VLAN For a given port, port list, or static port trunk, you can assign an ACL as a static port ACL to filter switched or routed IPv6 traffic entering the switch on that interface. You can use the same ACL for assignment to multiple VLANs. For limits and operating rules, see “ACL configuration and operating rules” (page 112).

Syntax: [no] vlan vid ipv6 access-group identifier vlan Assigns an ACL as a VACL to a VLAN to filter switched or routed IPv6 traffic entering the switch on that VLAN. You can use either the global configuration level or the VLAN context level to assign or remove a VACL. vid VLAN identification number. identifier The alphanumeric name by which the ACL can be accessed. An identifier can have up to 64 characters. The no form of the command removes the ACL assignment from the interface. NOTE: The switch allows you to assign an "empty" ACL identifier to a VLAN. In this case, if you later populate the ACL with ACEs, the new ACEs automatically become active on the assigned VLAN as they are created. Also, if you delete an assigned ACL from the switch without also using the no form of this command to remove the assignment to a VLAN, the ACL assignment remains as an "empty" ACL. For more information on "empty" ACLs, see the notes under “Deleting an ACL” (page 132).

130 IPv6 Access Control Lists (ACLs)

Example 69 Methods for enabling and disabling VACLs HP Switch(config)# vlan 20 ipv6 access-group List-010 vlan 1 HP Switch(config)# vlan 20 HP Switch(vlan-20)# ipv6 access-group List-015 vlan 2 HP Switch(vlan-20)# exit HP Switch(config)# no vlan 20 ipv6 access-group List-010 vlan 3 HP Switch(config)# vlan 20 HP Switch(vlan-20)# no ipv6 access-group 015 vlan 4 HP Switch(vlan-20)# exit 1 2 3 4

Enables a VACL from the Global Configuration Level Enables a VACL from a VLAN Context Disables a VACL from the Global Configuration Level Disables a VACL from a VLAN Context

Filtering inbound IPv6 traffic per port and trunk You can use the same ACL for assignment to multiple interfaces. For limits and operating rules, see “ACL configuration and operating rules” (page 112). Syntax: [no] interface [ port-list | trkx ] ipv6 access-group identifier in Assigns an ACL as a static port ACL to a port, port list, or static trunk to filter switched or routed IPv6 traffic entering the switch on that interface. You can use either the global configuration level or the interface context level to assign or remove a static port ACL. identifier The alphanumeric name by which the ACL can be accessed. An identifier can have up to 64 characters. [ port-list | trkx ] The port, trunk, or list of ports and/or trunks on which to assign or remove the specified ACL. NOTE: The switch allows you to assign an "empty" ACL identifier to an interface. If you later populate the empty ACL with one or more ACEs, it automatically becomes active on the assigned interfaces. Also, if you delete an assigned ACL from the running config file without also using the no form of this command to remove the assignment to an interface, the ACL assignment remains and automatically activates any new ACL you create with the same identifier.

Configuration Commands

131

Example 70 Methods for enabling and disabling ACLs HP Switch(config)# interface b10 ipv6 access-group List-1 in 1 HP Switch(config)# interface b10 HP Switch(eth-b10)# ipv6 access-group List-4 in 2 HP Switch(eth-b10)# exit HP Switch(config)# no interface b10 ipv6 access-group List-1 in 3 HP Switch(config)# interface b10 HP Switch(eth-b10)# no ipv6 access-group List-4 in 4 HP Switch(eth-b10)# exit 1 2 3 4

Enables a static port ACL from the Global Configuration level Enables a static port ACL from a port Disables a static port ACL from the Global Configuration level Uses a VLAN context to disable a static port

Deleting an ACL Syntax: no ipv6 access-list identifier Used in the global config context to remove the specified IPv6 ACL from the switch's running-config file. identifier The alphanumeric name assigned to an ACL.

Notes: If an ACL name is assigned to an interface before the ACL itself has been created, the switch creates an "empty" version of the ACL in the running configuration and assigns the empty ACL to the interface. Later adding explicit ACEs to the empty ACL causes the switch to automatically activate the ACEs as they are created and to implement the implicit deny at the end of the ACL. Deleting an ACL from the running configuration while the ACL is currently assigned on an interface results in an "empty" version of the ACL in the running configuration and on the interface. Later removing the ACL from the interface also removes the empty ACL from the running configuration. If you need to remove an ACL identifier assignment on an interface, see “Filtering routed IPv6 traffic” (page 128).

Inserting an ACE in an existing ACL This action uses a sequence number to specify where to insert a new ACE into an existing sequence of ACEs in an ACL. For information on editing ACLs, see “Editing an existing ACL” (page 162).

Syntax: 1 - 2147483647 [ permit | deny ] ipv6-ACE-criteria Used in the context of a given ACL, this command inserts an ACE into the ACL. 1 - 2147483647 The range of valid sequence numbers for an ACL. ipv6-ACE-criteria The various traffic selection options described earlier in this chapter. 132

IPv6 Access Control Lists (ACLs)

NOTE: Entering an ACE that would result in an out-of-range sequence number is not allowed. Use the resequence command to free up ACE numbering availability in the ACL. See “Resequencing the ACEs in an IPv6 ACL” (page 135). (For details on these options, see “Command Summary for Configuring ACLs” (page 94).) Example 71 Inserting a New ACE in an Existing ACL From the global configuration context: 1.

Insert a new ACE with a sequence number of 45 between the ACEs numbered 40 and 50 in “Appending an ACE to an existing list” (page 164). Inserting an ACE in an existing ACL HP Switch(Config)# ipv6 access-list My-list 1 HP Switch(config-ipv6-acl)# 45 permit icmp host 2001:db8:0:5ad::33 ::/0 2 HP Switch(config-ipv6-acl)# show run . . . ipv6 access-list "My-list" 10 permit ipv6 2001:db8:0:5ad::25/128 ::/0 20 permit ipv6 2001:db8:0:5ad::111/128 ::/0 30 permit icmp 2001:db8:0:5ad::115/128 ::/0 40 permit icmp 2001:db8:0:5ad::/64 ::/0 45 permit icmp 2001:db8:0:5ad::33 ::/0 3 50 permit icmp 2001:db8:0:5ad::19/128 ::/0 60 permit ipv6 ::/0 2001:db8:0:5ad::1/128 70 deny ipv6 2001:db8:0:5ad::/64 ::/0 80 permit ipv6 ::/0 ::/0 exit 1

Enters the Named-ACL context for “My-list”

2 3

Inserts a new ACE assigned to line 45

Configuration Commands

133

2.

From within the context of an IPv6 ACL named "List-01", insert a new ACE between two existing ACEs. In this example, the first command creates a new ACL and enters the ACL context. The next two ACEs entered become lines 10 and 20 in the list. The third ACE entered is inserted between lines 10 and 20 by using the sequence command with a sequence number of 11. Inserting an ACE into an existing sequence HP Switch(config)# Port_1_5400(config)# ipv6 access-list List-01 1 HP Switch(config-ipv6-acl)# permit ipv6 host fe80::100 host fe80::200 2 HP Switch(config-ipv6-acl)# permit ipv6 host fe80::103 any HP Switch(config-ipv6-acl)# 11 permit ipv6 host fe80::110 host fe80:: 3 HP Switch(config-ipv6-acl)# show Running configuration: . . . ipv6 access-list "List-01" 10 permit ipv6 fe80::100/128 11 permit ipv6 fe80::110/128 20 permit ipv6 fe80::103/128 exit 1 2 3 4

run

fe80::200/128 fe80::210/128 ::/0 4

Becomes Line 10 Becomes Line 20 Lines 10 and 20 were automatically numbered according to their order of entry in the list. Line 11 was explicitly numbered by the 11 permit command and was inserted in its proper place in the list.

Deleting an ACE from an existing ACL Syntax: no 1 - 2147483647 no [ permit | deny ] ipv6-ACE-criteria Both command options require entering the configuration context of the ACL containing the ACE you want to delete. The first command option deletes the ACE assigned to the specified sequence number. The second command option deletes the ACE having the syntax specified by ipv6-ACE-criteria. 1 - 2147483647 The range of valid sequence numbers for an ACL. ipv6-ACE-criteria The traffic selection options included in the ACE. To use this method to delete an ACE, the criteria specified in the command must match the criteria specified in the actual ACE you want to delete. 1. 2. 3.

To find the sequence number of the ACE you want to delete, use show access-list identifier or show access-list config to view the ACL. Use ipv6 access-list identifier config to enter the IPv6 ACL (config-ipv6-acl) context of the specified ACE. In the IPv6 ACL (config-ipv6-acl) context, type no and enter the sequence number of the ACE you want to delete.

Example 72 (page 135) illustrates the process for deleting an ACE from a list:

134

IPv6 Access Control Lists (ACLs)

Example 72 Deleting an ACE from an IPv6 ACL HP Switch(config)# show access-list My-List config 1

ipv6 access-list "My-List" 10 permit ipv6 fe80::100/128 ::/0 20 deny ipv6 fe80::110/128 fe80::/124 30 deny ipv6 fe80::111/128 fe80::/124 40 permit ipv6 ::/0 ::/0 exit HP Switch(config)# ipv6 access-list My-List 2 HP Switch(config-ipv6-acl)# no 30 3 HP Switch(config-ipv6-acl)# show access-list My-List config 4

ipv6 10 20 40 exit

access-list "My-List" permit ipv6 fe80::100/128 ::/0 deny ipv6 fe80::110/128 fe80::/124 permit ipv6 ::/0 ::/0 5

1 2 3 4 5

ACL Before Deleting an ACE Enters the IPv6 ACL (config-ipv6-acl) context for “My-List” This command deletes the ACE at line 30 ACL After Deleting the ACE at Line 20 The ACE at line 30 has been removed

Resequencing the ACEs in an IPv6 ACL This action reconfigures the starting sequence number for ACEs in an IPv6 ACL and resets the numeric interval between sequence numbers for ACEs configured in the ACL.

Syntax: ipv6 access-list resequence identifier starting-seq-# interval Resets the sequence numbers for all ACEs in the ACL. starting-seq-# Specifies the sequence number for the first ACE in the list. Default: 10; Range: 1 – 2147483647 interval Specifies the interval between consecutive sequence numbers for the ACEs in the list. Default: 10; Range: 1 – 2147483647 1. 2.

To view the current sequence numbering in an ACE, use show access-list config or show access-list identifier config. Use the command syntax (above) to change the sequence numbering.

Resequencing the ACEs in an IPv6 ACL

135

Example 73 Viewing and Resequencing an ACL This example resequences the "My-List" ACL at the bottom of Example 72 (page 135) so that the list begins with line 100 and uses a sequence interval of 100. HP Switch(config)# show access-list My-List config ipv6 access-list "My-List" 10 permit ipv6 fe80::100/128 ::/0 20 deny ipv6 fe80::110/128 fe80::/124 40 permit ipv6 ::/0 ::/0 exit HP Switch(config)# ipv6 access-list resequence My-List 100 100 HP Switch(config)# show access-list config ipv6 access-list "My-List" 100 permit ipv6 fe80::100/128 ::/0 200 deny ipv6 fe80::110/128 fe80::/124 300 permit ipv6 ::/0 ::/0 exit

Attaching a remark to an ACE A remark is numbered in the same way as an ACE and uses the same sequence number as the ACE to which it refers. This operation requires that the remark for a given ACE be entered prior to entering the ACE itself.

Syntax: remark remark-str 1 - 2147483647 remark remark-str no seq-# remark These commands are used in the ACL context to enter a comment related to an adjacent ACE. To associate a remark with a specific ACE, do one of the following: •

Enter the remark first (without a sequence number) and immediately follow it with the ACE (also without a sequence number). The remark and the following ACE will have the same (automatically generated) sequence number.



Enter the ACE with or without a sequence number, then use 1 - 2147483647 remark remark-str to enter the remark, where a number in the range of 1 - 2147483647 matches the sequence number of the related ACE. This method is useful when you want to enter a remark at some time after you have entered the related ACE.

remark-str The text of the remark. If spaces are included in the remark, the remark string must be delimited by either single quotes or double quotes. For example: remark Permits_Telnet_from_2001:db8:0:1ab_subnet remark "Permits Telnet from 2001:db8:0:1ab subnet" remark 'Permits Telnet from 2001:db8:0:1ab subnet'

1 - 2147483647 The range of valid sequence numbers for an ACL. For example, if the sequence number of the last ACE entered is "30", and sequence numbering is set to the (default) interval of 10, entering a remark and another ACE without specifying any sequence numbers results in a sequence number of "40" for both the remark and the ACE that follows it. 136

IPv6 Access Control Lists (ACLs)

The no form of the command deletes the indicated remark, but does not affect the related ACE.

Appending remarks and related ACEs to the end of an ACL To include a remark for an ACE that will be appended to the end of the current ACL: 1. Enter the remark first. 2. Then enter the related ACE. This results in the remark and the subsequent ACE having the same sequence number. Example 74 Appending remarks and related ACEs to the end of an ACL To append an ACE with an associated remark to the end of an ACL named "List-100," enter remarks from the CLI context for the desired ACL: HP Switch(config)# ipv6 access-list List-100 HP Switch(config-ipv6-acl)# permit tcp host 2001:db8:0:b::100:17 eq telnet any HP Switch(config-ipv6-acl)# permit tcp host 2001:db8:0:b::100:23 eq telnet any HP Switch(config-ipv6-acl)# remark “BLOCKS UNAUTH TELNET TRAFFIC FROM SUBNET B” 1 HP Switch(config-ipv6-acll)# deny tcp 2001:db8:0:a::/64 eq telnet any HP Switch(config-ipv6-acl)# show access-list List-100 config ipv6 access-list "List-100" 10 remark "TEXT" 10 permit tcp 2001:db8:0:b::100:17/128 eq 23 ::/0 20 permit tcp 2001:db8:0:b::100:23/128 eq 23 ::/0 30 remark "BLOCKS UNAUTH TELNET TRAFFIC FROM SUBNET B" 2 30 deny tcp 2001:db8:0:b::/64 eq 23 ::/0 exit HP Switch(config-ipv6-acl)# 1 2

The remark is assigned the same number as the immediately following ACE (“30” in this example) is assigned when it is automatically appended to the end of the list. This operation applies where new remarks and ACEs are appended to the end of the ACL and are automatically assigned a sequence number.

Inserting remarks and related ACEs within an existing list To insert an ACE with a remark within an ACL by specifying a sequence number: 1. Insert the numbered remark first 2. Then, using the same sequence number, insert the ACE. HP Switch(config-ipv6-acl)# 15 remark "PERMIT HTTP; STATION 23; SUBNET 1D" HP Switch(config-ipv6-acl)# 15 permit tcp host 2001:db8:0:1d::23 eq 80 2001:db8:0:2f::/64

HP Switch(config-ipv6-acl)# show access config . . . ipv6 access-list "List-105" 10 permit tcp 2001:db8:0:1f::/64 eq 80 2001:db8:0:2f::/64 15 remark "PERMIT HTTP; STATION 23; SUBNET 1D" 1 15 permit tcp 2001:db8:0:1d::23/128 eq 80 2001:db8:0:2f::/64 2 20 deny tcp 2001:db8:0:1d::/64 eq 80 2001:db8:0:2f::/64 exit . . .

Appending remarks and related ACEs to the end of an ACL

137

1 2

The above two commands insert a remark with its corresponding ACE (same sequence number) between two previously configured ACEs

Inserting a remark for an ACE that already exists in an ACL If an ACE already exists in a given ACL, you can insert a remark for that ACE by simply configuring the remark to have the same sequence number as the ACE.

Replacing an existing remark 1. 2.

Use ipv6 access-list identifier to enter the desired ACL context. Configure the replacement remark with the same sequence number as the remark you want to replace. This step overwrites the former remark text with the new remark text.

Example 75 Replacing an existing remark To change the text of the remark at line 15 in Section (page 137) to "PERMIT HTTP FROM ONE STATION", use the following command: HP Switch(config)# ipv6 access-list List-105 HP Switch(config-ipv6-acl)# 15 remark "PERMIT HTTP FROM ONE STATION"

Removing a remark from an existing ACE If you want to remove a remark, but want to retain the ACE: 1. Use ipv6 access-list identifier to enter the desired ACL context. 2. Use no 1 - 2147483647 remark. Using the no 1 - 2147483647 command without the remark keyword deletes both the remark and the ACE to which it is attached.

Operating notes for remarks •

An "orphan" remark is a remark that does not have an ACE counterpart with the same sequence number. The resequence command renumbers an orphan remark as a sequential, stand-alone entry without a permit or deny ACE counterpart. ipv6 access-list "XYZ" 10 remark "Permits HTTP" 10 permit tcp 2001:db8::2:1/120 eq 80 ::/0 12 remark "Denies HTTP from subnet 1." 18 remark "Denies pop3 from 1:157." 18 deny tcp 2001:db8::1:157/128 eq 110 ::/0 log 50 permit ipv6 ::/0 ::/0 exit HP Switch# ipv6 access-list resequence XYZ 100 10 HP Switch# show access-list XYZ config ipv6 access-list "XYZ" 100 remark "Permits HTTP" 100 permit tcp 2001:db8::2:1/120 eq 80 ::/0 110 remark "Denies HTTP from subnet 1." 120 remark "Denies pop3 from 1:157." 120 deny tcp 2001:db8::1:157/128 eq 110 ::/0 log

138

IPv6 Access Control Lists (ACLs)

130 permit ipv6 ::/0 ::/0 exit



Entering either an unnumbered remark followed by a manually numbered ACE (using 1 2147483647 ), or the reverse (an unnumbered ACE followed by a manually numbered remark) can result in an "orphan" remark.



Configuring two remarks without including either sequence numbers or an intervening, unnumbered ACE results in the second remark overwriting the first.

Example 76 Overwriting one remark with another HP Switch(config-ipv6-acl)# permit ipv6 host fe80::a1:121 fe80::/104 HP Switch(config-ipv6-acl)# deny tcp any eq ftp 2001:db8:0:a1::/64 HP Switch(config-ipv6-acl)# remark Marketing HP Switch(config-ipv6-acl)# remark Channel_Mktg Port_1_5400(config-ipv6-acl)# show access-list Accounting config ipv6 access-list "Accounting" 10 permit ipv6 fe80::a1:121/128 fe80::/104 20 deny tcp ::/0 eq 21 2001:db8:0:a1::/64 30 remark "Channel_Mktg" exit

NOTE: Where multiple remarks are sequentially entered for automatic inclusion at the end of an ACL, each successive remark replaces the previous one until an ACE is configured for automatic inclusion at the end of the list.

Viewing ACL configuration data The show commands in this section apply to both IPv6 and IPv4 ACLs. For information on IPv4 ACL operation, see chapter "IPv4 Access Control Lists" in the Access Security Guide for your switch. ACL Commands

Function

Page

show access-list

Displays a brief listing of all IPv4 and IPv6 ACLs on the switch.

140

show access-list config

Display the type, identifier, and content of all IPv4 and IPv6 ACLs configured in the switch.

141

show access-list vlan vid

List the name and type for each IPv4 142 and IPv6 ACL application assigned to a particular VLAN on the switch.

show access-list tunnel tunnel-id

List the name and type for each IPv6 143 ACL application assigned to a particular tunnel on the switch.

show access-list ports [ all | port-list ]

Lists the IPv4 and IPv6 ACL static 144 port assignments for either all ports and trunks, or for the specified ports and/or trunks.

show access-list identifier [ config ]

Display detailed content information 145 for a specific IPv4 or IPv6 ACL. Using the config option displays the ACL in a list format similar to that used to display an ACL in the show running-config output.

Viewing ACL configuration data

139

ACL Commands

Function

Page

show access-list resources

Displays the currently available per-slot resource availability. See appendix "Monitoring Resources" in the current Management and Configuration Guide for your switch.

n/a

show access-list radius [ all | port-list ]

Lists the IPv4 and IPv6 RADIUS ACLs currently assigned for either all ports and trunks, or for the specified ports and/or trunks. For more on this topic, see chapter "Configuring RADIUS Server Support for Switch Services" in the Access Security Guide for your switch.

156

show port-access web-based clients port-list detailed show port-access mac-based clients port-list detailed show port-access authenticator clients port-list detailed

For ports in the port-list shows 156 the details of the RADIUS-assigned features, including the ACE matches in RADIUS-assigned ACLs configured with the cnt (counter) option. For more on this topic, see chapter "Configuring RADIUS Server Support for Switch Services" in the Access Security Guide for your switch.

show config show running

show config includes configured ACLs and assignments existing in the startup-config file.

143

show running includes configured ACLs and assignments existing in the running-config file.

Viewing an ACL summary Lists the configured IPv4 and IPv6 ACLs, regardless of whether they are assigned to any interfaces.

Syntax: show access-list Lists a summary table of the name, type, and application status of all ACLs (IPv4 and IPv6) configured on the switch. 140 IPv6 Access Control Lists (ACLs)

HP Switch(config)# show access-list Access Control Lists Type ----ext std ext ipv6 ipv6 ipv6 ipv6

Appl Name ---- -----------------------yes 101 1 yes 55 2 yes Marketing 3 no Accounting 4 no List-01-Inbound 5 yes List-02-Outbound yes Test-1 1

IPv4

2 3

4 5

These ACLs exist in the configuration but are not applied to any interfaces and thus do not affect traffic

Term

Meaning

Type

Shows whether the listed ACL is an IPv6 (ipv6) ACL or one of two IPv4 ACL types: • std (Standard; source-address only) • ext (Extended; protocol, source, and destination data)

Appl

Shows whether the listed ACL has been applied to an interface (yes/no).

Name

Shows the identifier assigned to each ACL configured in the switch.

Viewing the content of all ACLs on the switch Lists the configuration details for every IPv4 and IPv6 ACL in the running-config file, regardless of whether any are actually assigned to filter traffic on specific interfaces.

Syntax: show access-list config Lists the configured syntax for all IPv4 and IPv6 ACLs currently configured on the switch. NOTE: You can use the output from this command for input to an offline text file in which you can edit, add, or delete ACL commands. This information also appears in the show running output. If you execute write memory after configuring an ACL, it appears in the show config output. Example 77 (page 142) shows the ACLs on a switch configured with two IPv6 ACLs named "Accounting" and "List-01-Inbound", and one extended IPv4 ACL named "101":

Viewing the content of all ACLs on the switch

141

Example 77 An ACL configured syntax listing HP Switch(config)# show access-list config ip access-list extended "101" 10 permit tcp 10.30.133.27 0.0.0.0 0.0.0.0 255.255.255.255 20 permit tcp 10.30.155.101 0.0.0.0 0.0.0.0 255.255.255.255 30 deny ip 10.30.133.1 0.0.0.0 0.0.0.0 255.255.255.255 log 40 deny ip 10.30.155.1 0.0.0.255 0.0.0.0 255.255.255.255 exit ipv6 access-list "Accounting" 10 permit tcp 2001:db8:0:1af::10:14/128 ::/0 eq 23 20 permit tcp 2001:db8:0:1af::10:23/128 ::/0 eq 23 30 deny tcp 2001:db8:0:1af::10/116 ::/0 log 40 permit ipv6 2001:db8:0:1af::10/116 ::/0 50 deny ipv6 ::/0 ::/0 log exit ipv6 access-list "List-01-Inbound" 10 permit icmp fe80::10:60/128 ::/0 dscp 38 20 permit icmp fe80::10:77/128 ::/0 dscp 38 30 permit icmp fe80::10:83/128 ::/0 dscp 38 40 deny icmp ::/0 ::/0 dscp 38 50 permit ipv6 fe80::10/112 ::/0 60 deny ipv6 fe80::/64 ::/0 exit

Viewing the IPv4 and IPv6 VACL assignments for a VLAN Lists the identifiers and types of VACLs currently assigned to a particular VLAN in the running-config file. For IPv6 ACLs, the switch supports one VACL assignment per VLAN. For IPv4 ACLs, the switch supports one inbound and one outbound RACL assignment per VLAN, and one VACL assignment per VLAN.

Syntax: show access-list vlan vid Lists the current IPv4 and IPv6 ACL assignments to the specified VLAN (in the running config file). NOTE: This information also appears in the show running output. If you execute write memory after configuring an ACL, it also appears in the show config output.

142

IPv6 Access Control Lists (ACLs)

Example 78 Displaying the IPv4 and IPv6 VACL assignments for a VLAN The following output shows that all inbound IPv6 traffic and the inbound and outbound, routed IPv4 traffic are all filtered on VLAN 20. HP Switch(config)# show access-list vlan 20 Access Lists for VLAN 20 Ipv6 Inbound Access List: Accounting 1 Inbound Access List: None 2 Ipv6 Outbound Access List: None 3 Outbound Access List: 101 4 Type: Extended Ipv6 VACL Access List: None 5 VACL Access List: None 6 Connection Rate Filter Access List: None 7 1 2 3 4 5 6 7

An IPv6 ACL named “Accounting” is assigned to filter routed IPv6 traffic entering the switch on VLAN 20 There is no filtering of routed IPv4 traffic entering the switch on VLAN 20 There is no filtering of routed IPv6 traffic leaving the switch on VLAN 20. An extended ACL named “101” is assigned to filter routed IPv4 traffic exiting from the switch on VLAN 20 There are no per-VLAN IPv6 or IPv4 ACLs assigned to VLAN 20 Applies to IPv4 Connection Rate Filter ACLs. Refer to the chapter titled “Virus Throttling (Connection-Rate Filtering)” in the Access Security Guide for your switch.

Viewing the IPv4 and IPv6 RACL and VACL assignments for a VLAN Lists the identifiers andLists the identifiers and types of RACLs and VACLs currently assigned to a particular VLAN in the running-config file. (The switch allows one inbound and one outbound RACL assignment per VLAN, plus one VACL assignment.)

Syntax: show access-list [ vlan vid | tunnel tunnel-id ] Lists any IPv4 and IPv6 RACL and/or VACL assignments to a VLAN in the running config file. NOTE: This information also appears in the show running output. If you execute write memory after configuring an ACL, it also appears in the show config output.

Viewing the IPv4 and IPv6 RACL and VACL assignments for a VLAN

143

Example 79 Viewing the IPv4 and IPv6 RACL and VACL assignments for a VLAN The following output shows that inbound, routed IPv6 traffic and outbound, routed IPv4 traffic are both filtered on VLAN 20. HP Switch(config)# show access-list vlan 20 Access Lists for VLAN 20 Ipv6 Inbound Access List: Accounting 1 Inbound Access List: None 2 Ipv6 Outbound Access List: None 3 Outbound Access List: 101 4 Type: Extended Ipv6 VACL Access List: None 5 VACL Access List: None 6 Connection Rate Filter Access List: None 7 1 2 3 4 5 6 7

An IPv6 ACL named “Accounting” is assigned to filter routed IPv6 traffic entering the switch on VLAN 20 There is no filtering of routed IPv4 traffic entering the switch on VLAN 20 There is no filtering of routed IPv6 traffic leaving the switch on VLAN 20 An extended ACL named “101” is assigned to filter routed IPv4 traffic exiting from the switch on VLAN 20 There are no per-VLAN IPv6 or IPv4 ACLs assigned to VLAN 20 Applies to IPv4 Connection Rate Filter ACLs. Refer to the chapter titled "Virus Throttling (Connection-Rate Filtering)" in the Access Security Guide for your switch

Viewing static port (and trunk) ACL assignments Lists the identification and types of current static port ACL assignments to individual switch ports and trunks, as configured in the running-config file. (The switch allows one static port ACL assignment per port.)

Syntax: show access-list ports [ all | port-list ] Lists the current static port ACL assignments for ports and trunks in the running config file. NOTE: This information also appears in the show running output. If you execute write memory after configuring an ACL, it also appears in the show config output.

144 IPv6 Access Control Lists (ACLs)

Example 80 Viewing static port (and trunk) ACL assignments The following output shows IPv4 and IPv6 ACLs configured on various ports and trunks on the switch: HP Switch(config)# show access-list ports all 1

Access Lists for Port 1 Inbound Ipv6: List-01-Inbound

2

Access Lists for Port 12 Inbound : 101 Type : Extended Inbound Ipv6: Accounting

3

Access Lists for Port Trk2 Inbound Ipv6: Accounting

4

Access Lists for Port Trk5 Inbound : Marketing Type : Extended 1 2 3 4

An IPv6 ACL is filtering inbound traffic on port 1 Both an IPv4 ACL and an IPv6 ACL are filtering inbound IPv4 and IPv6 traffic, respectively, on port 12 An IPv6 ACL is filtering inbound IPv6 traffic on Trunk 2 (Trk2) An IPv4 ACL is filtering inbound IPv4 traffic on Trunk 5 (Trk5)

Viewing the content of a specific ACL Displays a specific IPv6 or IPv4 ACL configured in the running config file in an easy-to-read tabular format.

Syntax: show access-list identifier [ config ] Displays detailed information on the content of a specific ACL configured in the running-config file. NOTE: This information also appears in the show running display. If you execute write memory after configuring an ACL, it also appears in the show config display. For information on IPv4 ACL operation, see the latest version of the Access Security Guide for your switch.

Viewing the content of a specific ACL

145

Example 81 Viewing the content of a specific ACL Suppose you configured the following two ACLs in the switch: Identifier Accounting

Type

Desired action IPv6

• Permit Telnet traffic from these two IPv6 addresses: • 2001:db8:0:1af::10: 14 • 2001:db8:0:1af::10: 24 • Deny Telnet traffic from all other devices in the same subnet. • Permit all other IPv6 traffic from the subnet. • Deny and log any IPv6 traffic from any other source.

List-120

IPv4 Extended

• Permit any TCP traffic from 10.30.133.27 to any destination. • Deny any other IP traffic from 10.30.133.(1 - 255). • Permit all other IP traffic from any source to any destination.

Use show access-list identifier to inspect a specific IPv6 or IPv4 ACL, as follows:

146

IPv6 Access Control Lists (ACLs)

Example 82 Listing an IPv6 ACL HP Switch(config)# show access-list Accounting Access Control Lists Name: Accounting Type: ipv6 Applied: Yes 1 SEQ Entry -------------------------------------------------------10 Action: permit Remark: Telnet Allowed 2 3 Src IP: 2001:db8:0:1af::10:14 4 Prefix Len: 128 5 Dst IP: :: 6 Prefix Len: 0 7 Src Port(s): 8 Dst Port(s): eq 23 9 Proto : TCP Option(s): 10 Dscp : 20

30

40

Action: permit Src IP: 2001:db8:0:1af::10:23 Dst IP: :: Src Port(s): Dst Port(s): eq 23 Proto : TCP Option(s): Dscp : -

Prefix Len: 128 Prefix Len: 0

Action: deny (log) Src IP: 2001:db8:0:1af::10 Dst IP: :: Src Port(s): Dst Port(s): Proto : TCP Option(s): Dscp : -

Prefix Len: 116 Prefix Len: 0

Action: permit Src IP: 2001:db8:0:1af::10 Dst IP: :: Src Port(s): Dst Port(s): Proto : IPV6 Dscp : 1

2

3 5 7

Indicates whether the ACL is applied to an interface Remark Field (Appears if remark configured.) Source Address Destination Address TCP Source Port

Prefix Len: 116 Prefix Len: 0

9 10 8

Protocol Data DSCP Codepoint or Precedence TCP Destination Port NOTE: An empty TCP field indicates that the TCP port number for that field can be any value

4 6

Source and Destination Prefix Lengths

Viewing the content of a specific ACL

147

Example 83 Listing an IPv4 extended ACL HP Switch(config)# show access-list List-120 Access Control Lists Name: List-120 Type: Extended Applied: No 1 SEQ Entry ---------------------------------------------------------10 Action: permit 2 Remark: Telnet Allowed 3 Src IP: 10.30.133.27 Mask: 0.0.0.0 Port(s): eq 23 4 Dst IP: 0.0.0.0 Mask: 255.255.255.255 Port(s): 5 6 Proto : TCP (Established) 7 TOS : Precedence: routine

1 2 3 4 6 7 5

20 Action: Src IP: Dst IP: Proto : TOS :

deny (log) 10.30.133.1 0.0.0.0 IP -

30 Action: Src IP: Dst IP: Proto : TOS :

permit 0.0.0.0 0.0.0.0 IP -

Mask: 0.0.0.255 Mask: 255.255.255.255

Port(s): Port(s):

Precedence: -

Mask: 255.255.255.255 Mask: 255.255.255.255

Port(s): Port(s):

Precedence: -

Indicates whether the ACL is applied to an interface Remark Field (Appears if remark configured) Source Address TCP Source Port Protocol Data DSCP Codepoint and Precedence Data Empty field indicates that the destination TCP port can be any value

The show access-list identifier config command shows the same ACL data as show access-list identifier but in the format used by the show [run|config] commands to list the switch configuration.

148

IPv6 Access Control Lists (ACLs)

Example 84 An ACL listed with the config option Port-1(config)# show access-list List-120 config ip access-list extended "List-120" 10 remark "Telnet Allowed" 10 permit tcp 10.30.133.27 0.0.0.0 eq 23 0.0.0.0 255.255.255.255 precedence 0 established 20 deny ip 10.30.133.1 0.0.0.255 0.0.0.0 255.255.255.255 log 30 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 exit

Table 18 Descriptions of data types included in show access-list acl-id output Field

Description

Name

The ACL identifier. For IPv6 ACLs, is an alphanumeric name. For IPv4 ACLs, can be a number from 1 to 199 or an alphanumeric name.

Type

IPv6, Standard, or Extended. IPv6 ACLs use a source and a destination address, plus IPv6 protocol specifiers. • Standard ACLs are IPv4 only, and use only a source IP address. • Extended ACLs are available in IPv4 only, and use both source and destination IP addressing, as well as other IP protocol specifiers.

Applied

Yes means the ACL has been applied to an interface. No means the ACL exists in the switch configuration, but has not been applied to any interface, and is therefore not in use.

SEQ

The sequential number of the ACE in the specified ACL.

Entry

Lists the content of the ACEs in the selected ACL.

Action

Permit (forward) or deny (drop) a packet when it is compared to the criteria in the applicable ACE and found to match. Includes the optional log option, if used, in deny or permit actions.

Remark

Displays any optional remark text configured for the selected ACE.

IP

Used for IPv4 standard ACEs: The source IPv4 address to which the configured mask is applied to determine whether there is a match with a packet.

Src IP

Used for IPv6 ACEs and IPv4 extended ACEs: The source IPv6 or IPv4 address to which the configured mask is applied to determine whether there is a match with a packet.

Dst IP

Used for IPv6 ACEs and IPv4 extended ACEs: The source and destination IP addresses to which the corresponding configured masks are applied to determine whether there is a match with a packet.

Mask

Used in IPv4 ACEs, the mask is configured in an ACE and applied to the corresponding IP address in the ACE to determine whether a packet matches the filtering criteria.

Prefix Len (source Used in IPv6 ACEs to specify the number of consecutive high-order (leftmost) bits of the source and destination) and destination addresses configured in an ACE to be used to determine a match with a packet being filtered by the ACE. Proto

Used in IPv6 ACEs and IPv4 extended ACEs to specify the packet protocol type to filter.

Port(s)

Used in IPv4 extended ACEs to show any TCP or UDP operator and port numbers included in the ACE.

Src Ports Dst Ports

Used in IPv6 ACEs to show TCP or UDP source and destination operator and port numbers included in the ACE.

DSCP

Used in IPv6 ACEs to show the DSCP precedence or codepoint setting, if any.

TOS

Used in IPv4 extended ACEs to indicate type-of-service setting, if any.

Precedence

Used in IPv4 extended ACEs to indicate the IP precedence setting, if any.

Viewing the content of a specific ACL

149

Creating or editing an ACL offline The section titled “Editing an existing ACL” (page 162) describes how to use the CLI to edit an ACL and is most applicable in cases where the ACL is short or there is only a minor editing task to perform. The offline method provides a useful alternative to using the CLI for creating or extensively editing a large ACL. For longer ACLs that may be difficult or time-consuming to accurately create or edit in the CLI, you can use the offline method. NOTE: Beginning with software release K_12_XX, copy commands that used either tftp or xmodem also include an option to use usb as a source or destination device for file transfers.

The offline process 1.

Begin by doing one of the following: •

To edit one or more existing ACLs, use copy command-output tftp to copy the current version of the ACL configuration to a file in your TFTP server. For example, to copy the ACL configuration to a file named acl-001.txt in the TFTP directory on a server at FE80::2a1:200: HP Switch# copy command-output 'show access-list config' tftp fe80::2a1:200 acl-001.txt pc

• 2.

To create a new ACL, open a text (.txt) file in the appropriate directory on a TFTP server accessible to the switch.

Use a text editor to create or edit the ACLs in the *.txt ASCII file format. If you are replacing an ACL on the switch with a new ACL that uses the same number or name syntax, begin the command file with a no ip access-list command to remove the earlier version of the ACL from the switch's running-config file. Otherwise, the switch will append the new ACEs in the ACL you download to the existing ACL. For example, if you planned to use the copy command to replace an ACL named "List-120", you would place this command at the beginning of the edited file: no ipv6 access-list List-120 Example 85 An offline ACL file designed to replace an existing ACL no ipv6 access-list List-120 1 ip access-list "List-120" 2 10 remark "THIS ACE ALLOWS TELNET" 10 permit tcp fe80::17/128 ::/0 eq 23 20 deny ipv6 fe80::123/128 fe80::/125 log 30 deny ipv6 fe80::255/128 fe80::/125 log 40 remark "THIS IS THE FINAL ACE IN THE LIST" 40 permit ipv6 ::/0 ::/0 exit 1 2

3.

150

Removes an existing ACL and replaces it with a new version with the same identifier. To append new ACEs to an existing ACL instead of replacing it, you would omit the first line and ensure that the sequence numbering for the new ACEs begin with a number greater than the highest number in the existing list.

Use copy tftp command-file to download the file as a list of commands to the switch.

IPv6 Access Control Lists (ACLs)

Example of using the offline process Suppose that you want to create an IPv6 ACL for a VACL application and download it to a switch from a TFTP server at FE80::1ad:17. Suppose that you want to create an IPv6 ACL for a RACL application and download it to a switch from a TFTP server at FE80::1ad:17. 1. You would create a .txt file with the content shown in Example 84 (page 149). 2. After you copy the above .txt file to the TFTP server at FE80::1ad:17, you would then execute the following command: copy tftp command-file fe80::1ad:17 acl-001.txt pc In this example, the CLI would show output similar to the following to indicate that the ACL was successfully downloaded to the switch: NOTE: If a transport error occurs, the switch does not execute the command and the ACL is not configured. NOTE: Blank lines may appear in the command output when you copy the command file to the switch. However, they are eliminated in the copy of the ACL in switch memory. This is normal operation. See also Example 87 (page 152) for the configuration resulting from this output. Example 86 Using copy tftp command-file to configure an ACL in the switch Switch(config)# copy tftp command-file fe80::1ad:17 acl-001.txt pc Running configuration may change, do you want to continue[y/n]? y 1. ipv6 access-list "acl-001" 6. ; CREATED ON JUNE 10 10. 10 remark "Telnet Denied Here" 13. 10 deny tcp 2001:db8:0:1af::/64 ::/0 eq 23 16. 30 deny tcp ::/0 ::/0 log 19. 40 deny icmp 2001:db8:0:1af::/64 ::/0 134 22. 50 deny icmp 2001:db8:0:1af::/64 ::/0 133 27. ; PERMITS IPV6 ANY ANY 31. 60 permit ipv6 ::/0 ::/0 34. exit 36. vlan 20 ipv6 access-group acl-001 in

3.

In this example, the command to assign the ACL to a VLAN was included in the .txt command file. If this is not done in your applications, the next step is to manually assign the new ACL to the intended VLAN: vlan vid ipv6 access-group identifier vlan vlan vid ipv6 access-group identifier in

4.

You can then use the show run or show access-list config command to inspect the switch configuration to ensure that the ACL was properly downloaded.

Creating or editing an ACL offline

151

Example 87 Verifying the .txt file download to the switch HP Switch(config)# show run . . . ipv6 access-list "acl-001" 10 remark "Telnet Denied Here" 10 deny tcp ::/0 ::/0 eq 23 30 deny tcp ::/0 ::/0 log 40 deny icmp ::/0 ::/0 134 50 deny icmp ::/0 ::/0 133 60 permit ipv6 ::/0 ::/0 exit . . . vlan 20 1 ipv6 access-group "acl-001" vlan ipv6 access-group "acl-001" in exit . . . 1

As a part of the instruction set included in the .txt file, the ACL is assigned to inbound IP traffic on VLAN 20 NOTE: The comment preceded by " ; " in the .txt source file for this configuration do not appear in the ACL configured in the switch

5.

If the configuration appears satisfactory, save it to the startup-config file: HP Switch(config)# write memory

Enabling ACL logging on the switch For more information, see “Testing and troubleshooting ACLs” (page 164). 1. If you are using a syslog server, use the logging ip-addr command to configure the syslog server IP addresses; ensure that the switch can access any syslog servers you specify. 2. Use logging facility syslog to enable the logging for syslog operation. 3. Use the debug destination command to configure one or more log destinations. Destination options include logging and session. For more information on debug, see "Debug and Syslog Messaging Operation" in the appendix, "Troubleshooting", in the latest Management and Configuration Guide for your switch. 4. 5.

152

Use debug acl or debug all to configure the debug operation to include ACL messages. Configure an ACL with the deny or permit action and the log option in one or more ACEs.

IPv6 Access Control Lists (ACLs)

Example 88 Enabling ACL logging on the switch Suppose that you want to configure the following on a switch receiving IPv6 traffic and configured for IPv4 routing: •

For port B1 on VLAN 10, configure an IPv6 ACL with an ACL-ID of "NO-TELNET" and use the PACL in option to deny Telnet traffic entering the switch from IP address FE80::10:3.



Configure the switch to send an ACL log message to the current console session and to a syslog server at 10.10.50.173 on VLAN 50 if the switch detects a packet match denying a Telnet attempt from FE80::10:3.

Figure 8 Example of an ACL log application Syslog

Switch

Consol

Console RS-232 VLAN 50 10.10.50.1

10.10.50.1

VLAN 20 10.10.20.1 VLAN 10

Port

FE80::10:1

FE80::10:3

Apply the ACL "NO TELNET" as a PACL on port B1 to deny Telnet access to inboundTelnet traffic from FE80::10:3.

Example 89 Commands for applying an ACL with logging HP Switch(config)# ipv6 access-list NO-TELNET HP Switch(config-ipv6-acl)# remark "deny fe80::10:3 Telnet traffic." HP Switch(config-ipv6-acl)# deny tcp host fe80::10:3 any eq telnet log HP Switch(config-ipv6-acl)# permit ipv6 any any HP Switch(config-ipv6-acl)# exit HP Switch(config)# vlan 10 ipv6 access-group NO-TELNET vlan 1 HP Switch(config)# logging 10.10.50.173 HP Switch(config)# logging facility syslog HP Switch(config)# debug destination logging HP Switch(config)# debug destination session HP Switch(config)# debug acl HP Switch(config)# write mem HP Switch(config)# show debug Debug Logging Destination: Logging -10.10.50.173 Facility = syslog Severity = debug System Module = all-pass Priority Desc = Session Enabled debug types: event acl log HP Switch(config)# show access-list NO-TELNET config

Enabling ACL logging on the switch

153

ipv6 access-list "NO-TELNET" 10 remark "deny fe80::10:3 TELNET TRAFFIC" 10 deny tcp fe80::10:3/128 ::/0 eq 23 log 20 permit ipv6 ::/0 ::/0 exit 1

Assigns the ACL named “NOTELNET” as a VACL to filter Telnet traffic from FE80::10:3 entering the switch on VLAN 10

154 IPv6 Access Control Lists (ACLs)

Example 90 ACL log application Suppose that you want to configure the following operation: •

For VLAN 10, configure an ACL with an ACL-ID of "NO-TELNET" and use the RACL in option to deny Telnet traffic entering the switch from IP address 2001:db8:0:4b1::10:3 to any routed destination. (This assignment will not filter Telnet traffic from 2001:db8:0:4b1::10:3 to destinations on VLAN 10 itself.)



Configure the switch to send an ACL log message to the current console session and to a syslog server at 2001:db8:0:4b1::20:3 on VLAN 20 if the switch detects a packet match denying a Telnet attempt from 2001:db8:0:4b1::10:3.

(This example assumes that IPv6 routing is already configured on the switch.) Figure 9 Example of an ACL log application Syslog Switch Console RS-232

Consol Subnet

VLAN 20

2002:db8:0:4b1::

2002:db8:0:4b1::20:1 VLAN 10 2002:db8:0:4b1::10:1

2002:db8:0:4b1::

Subnet

Apply the ACL "NO TELNET" as a RACL here to deny Telnet access to inbound, routed Telnet traffic from 2002:db8:0:4b1::10:3.

Block Telnet access to routed destinations from this host.

Example 91 Commands for applying an ACL with logging HP Switch(config)# ipv6 access-list NO-TELNET Switch(config-ipv6-acl)# remark "deny TELNET TRAFFIC IN" Switch(config-ipv6-acl)# deny tcp host 2001:db8:0:4b1::1 any eq telnet log Switch(config-ipv6-acl)# permit ipv6 any any Switch(config-ipv6-acl)# exit Switch(config)# vlan 22 ipv6 access-group NO-TELNET in 1 Switch(config)# logging 2001:db8:0:4b1::20:3 Switch(config)# logging facility syslog Switch(config)# debug destination logging Switch(config)# debug destination session Switch(config)# debug acl Switch(config)# write mem Switch(config)# show debug Debug Logging Source IP Selection: Outgoing Interface Destination: Logging -2001:db8:0:4b1::20:3 Facility = syslog Severity = debug System Module = all-pass Priority Desc = Session Enabled debug types: Enabling ACL logging on the switch

155

event acl log HP Switch(config)# show access-list config ipv6 access-list "NO-TELNET" 10 remark "deny TELNET TRAFFIC IN" 10 deny tcp 2001:db8:0:4b1::10:3/128 ::/0 eq 23 log 20 permit ipv6 ::/0 ::/0 exit 1

Assigns the ACL named “NOTELNET”as an RACL to filter routed Telnet traffic from 2001:db8:0:4b1::10:3 entering the switch on VLAN 10

Monitoring static ACL performance ACL statistics counters provide a means for monitoring ACL performance by using counters to display the current number of matches the switch has detected for each ACE in an ACL assigned to a switch interface. This can help, for example, to determine whether a particular traffic type is being filtered by the intended ACE in an assigned list, or if traffic from a particular device or network is being filtered as intended. NOTE: This section describes the command for monitoring static ACL performance. To monitor RADIUS-assigned ACL performance, use either of the following commands: show access-list radius [ all | port-list ] show access-list radius [ authenticator | mac-based | web-based ] clients port-list detailed See chapter "Configuring RADIUS Server Support for Switch Services" in the latest Access Security Guide for your switch.

Syntax: [ show | clear ] statistics aclv4 acl-name-str port aclv4 acl-name-str vlan aclv6 acl-name-str port aclv6 acl-name-str vlan aclv6 acl-name-str tunnel tunnel-id [ in |

port-# vid [ in | out | vlan ] port-# vid vlan [ in | out | vlan

]

out ]

show Displays the current match (hit) count per ACE for the specified IPv6 or IPv4 static ACL assignment on a specific interface. clear Resets ACE hit counters to zero for the specified IPv6 or IPv4 static ACL assignment on a specific interface. Total This column lists the running total of the matches the switch has detected for the ACEs in an applied ACL since the ACL's counters were last reset to 0 (zero).

156

IPv6 Access Control Lists (ACLs)

Example 92 Both IPv6 and IPv4 ACL activity HP Switch# show statistics aclv6 IPV6-ACL vlan 20 vlan HitCounts for ACL IPV6-ACL Total ( 12) 10 permit icmp ::/0 fe80::20:2/128 128 ( 6) 20 deny tcp ::/0 fe80::20:2/128 eq 23 log ( 41) 30 permit ipv6 ::/0 ::/0 HP Switch# show statistics aclv4 102 vlan 20 vlan HitCounts for ACL 102 Total Delta ( 4) 10 permit icmp 10.10.20.3 0.0.0.0 10.10.20.2 0.0.0.0 8 ( 8) 20 deny icmp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 8 ( 2) 30 permit tcp 10.10.20.3 0.0.0.255 10.10.20.2 0.0.0.255 eq 23 ( 2) 55 deny tcp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 8 (125) 60 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255

ACE counter operation For a given ACE in an assigned ACL, the counter increments by 1 each time the switch detects a packet that matches the criteria in that ACE, and it maintains a running total of the matches since the last counter reset. Example 93 ACE counter operation In ACL line 10 below, there has been a total of 37 matches on the ACE since the last time the ACL's counters were reset. Total 37)

(

10 permit icmp ::/0 fe80::20:2/128 128

NOTE: This ACL monitoring feature does not include hits on the "implicit deny" that is included at the end of all ACLs. Also, if the show statistics command does not show any ACE hit activity at first use, re-execute the command.

Resetting ACE hit counters to zero •

Using the clear statistics command, see “Monitoring static ACL performance” (page 156)



Removing an ACL from an interface zeros the ACL's ACE counters for that interface only.



For a given ACL, either of the following actions clear the ACE counters to zero for all interfaces to which the ACL is assigned: •

Adding or removing a permit or deny ACE in the ACL.



Rebooting the switch.

Example of ACL performance monitoring Example 94 (page 158) shows a sample of performance monitoring output for an IPv6 ACL assigned as a VACL.

Monitoring static ACL performance

157

Example 94 IPv6 ACL performance monitoring output HP Switch# show statistics aclv6 V6-02 vlan 20 vlan HitCounts for ACL V6-02 Total ( 5) ( 4) ( 136) ( 2) ( 10) ( 8) ( 155)

10 20 30 40 50 60 70

permit icmp ::/0 fe80::20:2/128 128 permit icmp ::/0 fe80::20:3/128 128 permit tcp fe80::20:1/128 ::/0 eq 23 deny icmp ::/0 fe80::20:1/128 128 deny tcp ::/0 ::/0 eq 23 deny icmp ::/0 ::/0 133 permit ipv6 ::/0 ::/0

Example 95 (page 158) shows a sample of performance monitoring output for an IPv4 ACL assigned as a VACL. Example 95 IPv4 ACL performance monitoring output HP Switch# show statistics aclv4 102 vlan 20 vlan HitCounts for ACL 102 Total ( 1) ( 2) ( 2) ( 1) (10) log (27)

10 20 30 40 50

permit icmp 10.10.20.3 0.0.0.0 10.10.20.2 0.0.0.0 8 deny icmp 10.10.20.3 0.0.0.0 10.10.20.1 0.0.0.0 8 log deny icmp 10.10.20.2 0.0.0.0 10.10.20.3 0.0.0.0 8 log deny icmp 10.10.20.2 0.0.0.0 10.10.20.1 0.0.0.0 8 log deny tcp 10.10.20.2 0.0.0.255 10.10.20.3 0.0.0.255 eq 23

60 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255

Example of resetting ACE hit counters to zero The following example uses the counter activity in Example 96 (page 159) to demonstrate using clear statistics to reset the counters to zero.

158

IPv6 Access Control Lists (ACLs)

Example 96 IPv6 ACL performance monitoring output HP Switch# show statistics aclv6 V6-02 vlan 20 vlan HitCounts for ACL V6-02 Total ( 5) ( 4) ( 136) ( 2) ( 10) ( 8) ( 155)

10 20 30 40 50 60 70

permit icmp ::/0 fe80::20:2/128 128 permit icmp ::/0 fe80::20:3/128 128 permit tcp fe80::20:1/128 ::/0 eq 23 deny icmp ::/0 fe80::20:1/128 128 deny tcp ::/0 ::/0 eq 23 deny icmp ::/0 ::/0 133 permit ipv6 ::/0 ::/0

HP Switch# clear statistics aclv6 V6-02 vlan 20 vlan HP Switch# show statistics aclv6 V6-02 vlan 20 vlan HitCounts for ACL V6-02

( ( ( ( ( ( (

Total 0) 0) 0) 0) 0) 0) 0)

10 20 30 40 50 60 70

permit icmp ::/0 fe80::20:2/128 128 permit icmp ::/0 fe80::20:3/128 128 permit tcp fe80::20:1/128 ::/0 eq 23 deny icmp ::/0 fe80::20:1/128 128 deny tcp ::/0 ::/0 eq 23 deny icmp ::/0 ::/0 133 permit ipv6 ::/0 ::/0

Options for applying IPv6 ACLs on the switch To apply IPv6 ACL filtering, assign a configured IPv6 ACL to the interface on which you want the traffic filtering to occur. VLAN IPv6 traffic ACLs can be applied statically using the switch configuration. Port traffic ACLs can be applied either statically or dynamically (using a RADIUS server).

Static ACLS Static ACLs are configured on the switch. To apply a static ACL, assign it to an interface (VLAN or port). The switch supports three static ACL types: •

Routed IPv6 traffic ACL (RACL) An ACL configured on a VLAN to filter routed IPv6 traffic entering or leaving the switch on that interface, as well as IPv6 traffic having a destination on the switch itself. (Except for filtering IPv6 traffic to an address on the switch itself, IPv6 RACLs can operate only while IPv6 routing is enabled.



VLAN ACL (VACL) An ACL to a VLAN to filter IPv6 traffic entering the switch on that VLAN interface and having a destination on the same VLAN. The traffic can be either switched or routed.



Static Port ACL An ACL assigned to a port to filter IPv6 traffic entering the switch on that port, regardless of whether the traffic is routed, switched, or addressed to a destination on the switch itself.

Options for applying IPv6 ACLs on the switch

159

RADIUS-assigned ACLs A RADIUS-assigned ACL for filtering traffic from a specific client or group of clients is configured on a RADIUS server. When the server authenticates a client associated with that ACL, the ACL is assigned to filter the inbound IP traffic received from the authenticated client through the port on which the client is connected to the switch. If the RADIUS server supports both IPv4 and IPv6 ACEs, the ACL assigned by the server can be configured to filter both traffic types, or just the IPv4 traffic. When the client session ends, the ACL is removed from the port. The switch allows as many RADIUS-assigned ACLs on a port as it allows authenticated clients. For information on RADIUS-assigned ACLs, see chapter "Configuring RADIUS Server Support for Switch Services" in the latest Access Security Guide for your switch. NOTE: This section describes the IPv6 ACL applications you can statically configure on the switch. For information on static IPv4 ACL applications, see chapter "IPv4 Access Control Lists (ACLs)" in the latest Access Security Guide for your switch.

Using CIDR notation to enter the IPv6 ACL prefix length CIDR (classless inter-domain routing) notation is used to specify ACL prefix lengths. The switch compares the address bits specified by a prefix length for an SA or DA in an ACE with the corresponding address bits in a packet being filtered by the ACE. If the designated bits in the ACE and in the packet have identical settings, the addresses match. Table 19 Examples of CIDR notation for prefix lengths SA or DA used in an ACL with CIDR notation

Resulting prefix length defining an address match

Meaning

2620:0:a03:e102::/64

2620:0:a03:e102

The leftmost 64 bits must match. The remaining 64 bits are wildcards.

2620:0:a03:e102:215::/80

2620:0:a03:e102:215

The leftmost 80 bits must match. The remaining 48 bits are wildcards.

2620:0:a03:e102:215:60ff:fe7a:adc0/128

2620:0:a03:e102:215:60ff:fe7a:adc0

All 128 bits must match. This specifies a single host address.

2001:db8:a03:e102:0:ab4:100::/112

2001:db8:a03:e102:0:ab4:100

The leftmost 112 bits must match. The remaining 16 bits are wildcards.

Overview of IPv6 ACLs IPv6 ACLs enable filtering on the following: •

Source and destination IPv6 addresses (required), in one of the following options: •

Specific host IPv6



Subnet or contiguous set of IPv6 addresses



Any IPv6 address



Choice of any IPv6 protocol



Optional packet-type criteria for ICMP traffic



Optional source and/or destination TCP or UDP port, with a further option for comparison operators



TCP flag (control bit) options

160 IPv6 Access Control Lists (ACLs)



Filtering for TCP traffic based on whether the subject traffic is initiating a connection ("established" option)



Optional DSCP (IP precedence and ToS) criteria

The switch allows up to 2048 ACLs each for IPv4 and IPv6 (with RADIUS-based ACL resources drawn from the IPv4 allocation). The total is determined from the number of unique identifiers in the configuration. For example, configuring two IPv6 ACLs results in an ACL total of two, even if neither is assigned to an interface. If you then assign a nonexistent IPv6 ACL to an interface, the new total is three, because the switch now has three unique IPv6 ACL names in its configuration. For information on determining the current resource availability and usage, see appendix "Monitoring Resources" in the Management and Configuration Guide for your switch. For ACL resource limits, see the appendix covering scalability in the latest Management and Configuration Guide for your switch.

Commands to create, enter, and configure an ACL For a match to occur with an ACE, a packet must have the source and destination IPv6 address criteria specified by the ACE, as well as any IPv6 protocol-specific criteria included in the command. Use the following general steps to create or add to an ACL: 1. Create and/or enter the context of a given ACL. 2. Enter the first ACE in a new ACL, or append an ACE to the end of an ACL. Topic

Page

applying or removing an ACL on an interface

132

deleting an ACL

132

editing an ACL (inserting or removing ACEs from an existing ACL)

162

sequence numbering in ACLs

163

including remarks in an ACL

136

viewing ACL configuration data

139

creating or editing ACLs offline

150

enabling ACL “Deny” logging

164

Example: IPv6 ACL configuration in a routed environment Suppose that you want to implement these policies on a switch configured for IPv6 routing and membership in VLANs 15, 14, and 13: Policy A: 1.

Permit IPv6 Telnet traffic from 2001:db8:0:1af::144 to 2001:db8:0:1ae::178.

2.

Deny all other IPv6 traffic from network 2001:db8:0:1af::/64 (VLAN 15) to 2001:db8:0:1ae::/64 (VLAN 14).

3.

Permit all other IPv6 traffic from 2001:db8:0:1af::/64 (VLAN 15) to any destination. See "A" in Figure 10 (page 162).

Policy B: 1.

Permit FTP traffic from IPv6 address 2001:db8:0:1ae::100 (on VLAN 14) to 2001:db8:0:1ad::55 (on VLAN 13). The TCP port number assigned for FTP traffic is "21".

2.

Deny FTP traffic from other hosts on network 2001:db8:0:1ae::/64 to any destination.

3.

Permit all other IPv6 traffic.

Example: IPv6 ACL configuration in a routed environment

161

Figure 10 Example of an IPv6 ACL application

To implement the policies described above in Figure 10 (page 162), configure ACLs on the switch as shown in Example 97 (page 162). Example 97 Configuration commands for IPv6 ACL Configure ACLs on the switch: Example 98 Switch A shown in Figure 10 (page 162) HP Switch(config-ipv6-acl)# permit tcp host 2001:db8:0:1af::144 host 2001:db8:0:1ae::178 eq telnet HP Switch(config-ipv6-acl)# deny ipv6 2001:db8:0:1af::/64 2001:db8:0:1ae::/64 HP Switch(config-ipv6-acl)# permit ipv6 2001:db8:0:1af::/64 any HP Switch(config-ipv6-acl)# exit HP Switch(config)# vlan 1 ipv6 access-group List-01 in

Example 99 Switch B shown in Figure 10 (page 162) HP Switch(config-ipv6-acl)# 2001:db8:0:1ad::55 eq 21 HP Switch(config-ipv6-acl)# HP Switch(config-ipv6-acl)# HP Switch(config-ipv6-acl)# HP Switch(config-ipv6-acl)#

permit tcp host 2001:db8:0:1ae::100 host deny tcp 2001:db8:0:1ae::/64 any permit ipv6 any any exit vlan 1 ipv6 access-group List-02 in

Editing an existing ACL The CLI provides the capability for editing in the switch by using sequence numbers to insert or delete individual ACEs. An offline method is also available. This section describes using the CLI for editing ACLs. To use the offline method for editing ACLs, see “Creating or editing an ACL offline” (page 150).

General editing rules You can use the CLI to delete individual ACEs from anywhere in an ACL, append new ACEs to the end of an ACL, and insert new ACEs anywhere within an ACL.

162



When you enter a new ACE in an ACL without specifying a sequence number, the switch inserts the ACE as the last entry in the ACL.



When you enter a new ACE in an ACL and include a sequence number, the switch inserts the ACE according to the position of the sequence number in the current list of ACEs.

IPv6 Access Control Lists (ACLs)



You can delete an ACE by using the ipv6 access-list identifier command to enter the ACL's context, and then no seq-# (see page 134).



Deleting the last ACE from an ACL leaves the ACL in the configuration as an "empty" ACL placeholder that cannot perform any filtering tasks. (In any ACL, the implicit deny does not apply unless the ACL includes at least one explicit ACE. See “Deleting an ACL” (page 132)".)

Sequence numbering in ACLs The ACEs in any ACL are sequentially numbered. In the default state, the sequence number of the first ACE in a list is "10," and subsequent ACEs are numbered in increments of 10. The following show run output shows an ACL named "My-list" using the default numbering scheme: Example 100 Default sequential numbering for ACEs ipv6 access-list "My-list" 10 permit ipv6 2001:db8:0:5ad::25/128 ::/0 20 permit ipv6 2001:db8:0:5ad::111/128 ::/0 30 permit icmp 2001:db8:0:5ad::115/128 ::/0 135 40 deny ipv6 2001:db8:0:5ad::/64 ::/0 exit

An ACE can be appended to the end of the ACL by using ipv6 access-list from the global configuration prompt or by entering the ACL context: Example 101 Ways to append a new ACE to the end of an ACL HP Switch(config)# ipv6 access-list My-list permit esp host 2001:db8:0:5ad::19 any 1

2 HP Switch(Config)# ipv6 access-list My-list HP Switch(config-ipv6-acl)# permit ipv6 any host 2001:db8:0:5ad::1

1 2

From the global configuration prompt, appends an ACE to the end of the ACL named My-list Enters the context of the “My-list”ACL and appends an ACE to the end of the list

To continue from Example 101 (page 163) and append a final ACE to the end of the ACL:

Editing an existing ACL 163

Example 102 Appending an ACE to an existing list HP Switch(config-ipv6-acl)# deny ipv6 2001:db8:0:5ad::/64 any 1 HP Switch (config-ipv6-acl)# permit ipv6 any any 2 HP Switch(config-ipv6-acl)# show run . . . ipv6 access-list "My-list" 10 permit ipv6 2001:db8:0:5ad::25/128 ::/0 20 permit ipv6 2001:db8:0:5ad::111/128 ::/0 30 permit icmp 2001:db8:0:5ad::115/128 ::/0 40 permit icmp 2001:db8:0:5ad::/64 ::/0 50 permit 50 2001:db8:0:5ad::19/128 ::/0 60 permit ipv6 ::/0 2001:db8:0:5ad::1/128 70 deny ipv6 2001:db8:0:5ad::/64 ::/0 80 permit ipv6 ::/0 ::/0 exit 1 2

ACE appended as line 70 Appended as line 80

About displaying All ACLs and their assignments in the switch startup-config file and running-config file The show config and show running commands include in their listings any configured ACLs and any ACL assignments to VLANs. See Example 87 (page 152) for an example. Remember that show config lists the startup-config file and show running lists the running-config file.

Testing and troubleshooting ACLs You can monitor ACL performance by using the logging option (which generates log messages when there is a "deny" or “permit” ACE match) and the ACE statistics counters (which maintain running totals of the packet matches on each ACE in an ACL).

Enable IPv6 ACL "Deny" or “Permit” logging ACL logging enables the switch to generate a message when IP traffic meets the criteria for a match with an ACE that results in an explicit "deny" or “permit” action. You can use ACL logging to help: •

Test your network to help ensure that your ACL configuration is detecting and denying the incoming IPv6 traffic you do not want to enter the switch, or permitting the traffic.



Receive notification when the switch denies inbound IPv6 traffic you have designed your ACLs to reject (deny), or permits traffic you have designed your ACLs to allow (permit).

The switch sends ACL messages to syslog and optionally to the current console, Telnet, or SSH session. You can use logging to configure up to six syslog server destinations.

Requirements for using IPv6 ACL logging •

The switch configuration must include an ACL:

164 IPv6 Access Control Lists (ACLs)

1. 2.

Assigned to a port, trunk, or static VLAN interface Containing an ACE configured with the deny or permit action and the log option.



If the RACL application is used, IPv6 routing must be enabled on the switch.



For IPv6 ACL logging to a syslog server: •

The server must be accessible to the switch and identified in the running configuration.



The logging facility must be enabled for syslog.



Debug must be configured to: •

Support ACL messages



Send debug messages to the desired debug destination

These requirements are described in more detail under “Enabling ACL logging on the switch” (page 152).

ACL logging operation When the switch detects a packet match with an ACE and the ACE includes the deny or permit action and the optional log parameter, an ACL log message is sent to the designated debug destination. The first time a packet matches an ACE with deny or permit and log configured, the message is sent immediately to the destination and the switch starts a wait-period of approximately five minutes. (The exact duration of the period depends on how the packets are internally routed.) At the end of the collection period, the switch sends a single-line summary of any additional "deny" matches for that ACE (and any other "deny" ACEs for which the switch detected a match). If no further log messages are generated in the wait-period, the switch suspends the timer and resets itself to send a message as soon as a new "deny" match occurs. The data in the message includes the information illustrated in Example 103 (page 165). Example 103 Content of messages generated by an ACL-deny action Example Syslog report of the first deny event detected by the switch for this ACE. ACL 12/01/08 10:04:45 List NO-TELNET, seq#10 denied tcp 2001:db8:0:1ae::1a:3(1612) ->2001:db8:0:1ad::1a:2(23) on vlan 1, port A7

Example of subsequent deny events detected by the switch for the same ACE. Dec 1 10:04:45 2008:db8:0:1ad::1a:1 ACL: ACL 12/01/08 10:04:45 : ACL NO-TELNET seq#10 denied 6 packets

IPv6 counter operation with multiple interface assignments NOTE: The examples of counters in this section use small values to help illustrate counter operation. The counters in real-time network applications are generally much more active and show higher values. Where the same IPv6 ACL is assigned to multiple interfaces, the switch maintains a separate instance of each ACE counter in the ACL. When there is a match with traffic on one of the ACL's assigned interfaces, only the affected ACE counters for that interface are incremented. Other instances of the same ACL applied to other interfaces are not affected.

IPv6 counter operation with multiple interface assignments

165

Example 104 IPv6 counter operation with multiple interface assignments Suppose that: •

An ACL named "V6-01" is configured as shown in Example 105 (page 166) to block Telnet access to a workstation at FE80::20:2, which is connected to a port belonging to VLAN 20.



The ACL is assigned as a PACL (port ACL) on port 2, which is also a member of VLAN 20:

Example 105 ACL "V6-01" and command for PACL assignment on port 2 HP Switch(config)# show access-list V6-01 config ipv6 access-list "V6-01" 10 permit icmp ::/0 fe80::20:2/128 128 20 deny tcp ::/0 fe80::20:2/128 eq 23 log 30 permit ipv6 ::/0 ::/0 exit HP Switch(config)# int b2 ipv access-group V6-01 in 1 1

Assigns the ACL to port 2

Figure 11 Application to filter traffic inbound on port B2 5400zl Switch VLAN 20 FE80::20:1

FE80::20:2

Port B2

FE80::20:117 ACL "V6-01" assigned as a PACL on port B2.

Using the topology in Figure 11 (page 166), a workstation at FE80::20:117 on port B2 attempting to ping and Telnet to the workstation at FE80::20:2 is filtered through the PACL instance of the "V6-01" ACL assigned to port B2, resulting in the following:

166 IPv6 Access Control Lists (ACLs)

Example 106 Ping and Telnet from FE80::20:117 to FE80::20:2 filtered by the assignment of "V6-01" as a PACL on port B2 HP Switch# ping6 fe80::20:2%vlan20 fe80:0000:0000:0000:0000:0000:0020:0002 is alive, time = 5 ms HP Switch# telnet fe80::20:2%vlan20 Telnet failed: Connection timed out. HP Switch#

Example 107 Resulting ACE hits on ACL “V6-01” HP Switch# show statistics aclv6 IP-01 port 2 Hit Counts for ACL IPV6-ACL Total ( ( (

1) 1 5) 2 4) 3

10 permit icmp fe80::20:3/128 fe80::20:2/128 128 20 deny tcp ::/0 fe80::20:2/128 eq 23 log 30 permit ipv6 ::/0 ::/0 1 2 3

Shows the successful ping permitted by ACE 10 Indicates denied attempts to Telnet to FE80::20:2 via the instance of the "V6-01" PACL assignment on port 2 Indicates permitted attempts to reach any accessible destination via the instance of the “V6-01”PACL assignment on port 2

NOTE: IPv4 ACE counters assigned as RACLs operate differently than described above. For more information, see “IPv4 counter operation with multiple interface assignments” (page 167).

IPv4 counter operation with multiple interface assignments Where the same IPv4 ACL is assigned to multiple interfaces as a VLAN ACL (VACL) or port ACL (PACL), the switch maintains a separate instance of ACE counters for each interface assignment. Thus, when there is a match with traffic on one of the ACL's VACL- or PACL-assigned interfaces, only the ACE counter in the affected instance of the ACL is incremented. However, if an ACL has multiple assignments as an RACL, then a match with an ACE in any RACL instance of the ACL increments that same counter on all RACL-assigned instances of that ACL. (The ACE counters for VACL and PACL instances of an ACL are not affected by counter activity in RACL instances of the same ACL.)

IPv6 counter operation with multiple interface assignments

167

Example 108 IPv4 counter operation with multiple interface assignments Suppose that an IPv4 ACL named "Test-1" is configured as shown in Example 109 (page 168) to block Telnet access to a server at 10.10.20.12 on VLAN 20, and that the Test-1 ACL is assigned to VLANs as follows: •

VLAN 20: VACL



VLAN 50: RACL



VLAN 70: RACL

Example 109 ACL “Test-1” and interface assignment commands HP Switch(config)# show access-list Test1 config ip access-list extended "Test1" 10 deny tcp 0.0.0.0 255.255.255.255 10.10.20.12 0.0.0.0 eq 23 log 20 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 exit HP Switch(config)# vlan 20 ip access-group Test-1 vlan 1 HP Switch(config)# vlan 50 ip access-group Test-1 in 2 HP Switch(config)# vlan 70 ip access-group Test-1 in 3 1 2

Assigns the ACL as a VACL to VLAN 20 Assigns the ACL as an RACL to VLANs 50 and 70

3

Figure 12 Using the same IPv4 ACL for VACL and RACL applications ACL "Test-1" assigned as a VACL to VLAN 20.

5400zl Switch VLAN 20

.0

0 10.10.2

10.10.20.1 VLAN 50 10.10.55.1 10.10.20.12

0.0

10.10.3

VLAN 70 10.10.70.1

.0

0 10.10.7

ACL "Test-1" assigned as an RACL to both VLAN 50 and VLAN 70.

In the above case: •

Matches with ACEs 10 or 20 that originate on VLAN 20 increment only the counters for the instances of these two ACEs in the Test-1 VACL assignment on VLAN 20. The same counters in the instances of ACL Test-1 assigned to VLANs 50 and 70 are not incremented.



Any Telnet requests to 10.10.20.12 that originate on VLANs 50 or 70 are filtered by instances of Test-1 assigned as RACLs and increment the counters for ACE 10 on both RACL instances of the Test-1 ACL.

168 IPv6 Access Control Lists (ACLs)

Using the network in Figure 12 (page 168), a device at 10.10.20.4 on VLAN 20 attempting to ping and Telnet to 10.10.20.12 is filtered through the VACL instance of the "Test-1" ACL on VLAN 20 and results in the following: Example 110 Ping and Telnet from 10.10.20.4 to 10.10.20.2 filtered by the assignment of "Test-1" as an IPv4 VACL on VLAN 20 HP Switch(config)# ping 10.10.20.2 10.10.20.2 is alive, time = 5 ms HP Switch(config)# telnet 10.10.20.2 Telnet failed: Connection timed out. HP Switch(config)#

Example 111 Resulting ACE hits on IPv4 ACL “Test-1” HP Switch(config)# show statistics aclv4 Test-1 vlan 20 vlan Hit Counts for ACL Test-1 Total ( 5) 1 10 deny tcp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 eq 23 log ( 2) 2 20 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 HP Switch# show statistics aclv4 Test-1 vlan 50 in Hit Counts for ACL Test-1 Total ( 0) 3 10 deny tcp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 eq 23 log ( 0)20 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 1 2

3

Indicates denied attempts to Telnet to 10.10.20.12 filtered by the instance of the “Test-1” VACL assignment on VLAN 20 Indicates permitted attempts to reach any accessible destination via the instance of the “Test- 1”VACL assignment on VLAN 20. In this example, shows the successful pings permitted by ACE Shows that the hits on the instance of the “Test-1”VACL assignment on VLAN 20 have no effect on the counters for the RACL assignment of “Test-1” on VLAN 50

However, using a device at 10.10.30.11 on VLAN 50 for attempts to ping and Telnet to 10.10.20.12 requires routing and filters the attempts through the RACL instance of the “Test-1”ACL on VLAN 50. Example 112 Ping and Telnet from 10.10.30.11 to 10.10.20.2 filtered by the assignment of "Test-1" as an IPv4 RACL on VLAN 30 HP Switch# ping 10.10.20.2 10.10.20.2 is alive, time = 25 ms HP Switch# telnet 10.10.20.2 Telnet failed: Connection timed out. HP Switch#

This action has an identical effect on the counters in all RACL instances of the "Test-1" ACL configured and assigned to interfaces on the same switch. In this example, it means that the RACL assignments of "Test-1" on VLANs 50 and 70 are incremented by the above action occurring on VLAN 50.

IPv6 counter operation with multiple interface assignments

169

Example 113 Resulting ACE hits on the VLAN 30 IPv4 RACL assignment of the "Test-1" ACL HP Switch(config)# show statistics aclv4 Test-1 vlan 50 in Hit Counts for ACL Test-1 Total ( 6) 10 deny tcp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 eq 23 log ( 1) 20 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 HP Switch(config)#

NOTE: The Total 6 Indicates the same type of data as shown in Example 111 (page 169)for the VACL assignment of the “Test-1” ACL. That is, the Ping attempt incremented the counters for ACE 20 and the Telnet attempt incremented the counters for ACE 10 in the VLAN 50 RACL instance of the ACL.

Example 114 Resulting ACE hits on the VLAN 70 IPv4 RACL assignment of the "Test-1" ACL HP Switch(config)# show statistics aclv4 Test-1 vlan 70 in HitCounts for ACL Test-1 Total ( 6) 10 deny tcp 0.0.0.0 255.255.255.255 10.10.20.2 0.0.0.0 eq 23 log ( 1) 20 permit ip 0.0.0.0 255.255.255.255 0.0.0.0 255.255.255.255 HP Switch(config)#

NOTE: The Total 6 The ACE counters in the VLAN 70 RACL assignment of “Test-1” are also incremented by the commands executed in Example 112 (page 169). Note that the ACE counters for the VACL assignment of the "Test-1" ACL on VLAN 20 are not affected by ACE hits on the RACL assignments of the same ACL.

General ACL operating notes ACLs do not provide DNS hostname support. ACLs cannot be configured to screen hostname IP traffic between the switch and a DNS. ACLs do not affect serial port access. ACLs do not apply to the switch’s serial port. ACL screening of IPv6 traffic generated by the switch. Outbound IPv6 RACL applications on a switch do not screen IPv6 traffic (such as broadcasts, Telnet, Ping, and ICMP replies) generated by the switch itself. All ACLs applied on the switch do screen this type of traffic when other devices generate it. Similarly, all ACL applications can screen responses from other devices to unscreened IPv6 traffic the switch generates. ACL logging

170



The ACL logging feature generates a message only when packets are explicitly denied as the result of a match, and not when explicitly permitted or implicitly denied. To help test ACL logging, configure the last entry in an ACL as an explicit deny statement with a log statement included and apply the ACL to an appropriate port or VLAN.



The ACL logging feature generates a message only when packets are explicitly denied or permitted as the result of a match, and not when implicitly denied. To help test ACL logging,

IPv6 Access Control Lists (ACLs)

configure the last entry in an ACL as an explicit deny or permit statement with a log statement included, and apply the ACL to an appropriate VLAN. •

A detailed event will be logged for the first packet that matches a “deny” or “permit” ACL logged entry with the appropriate action specified. Subsequent packets matching ACL logged entries will generate a new event that summarizes the number of packets that matched each specific entry (with the time period). See Example 103 (page 165) for an example.



Logging enables you to selectively test specific devices or groups. However, excessive logging can affect switch performance. For this reason, HP recommends that you remove the logging option from ACEs for which you do not have a present need. Also, avoid configuring logging where it does not serve an immediate purpose. (ACL logging is not designed to function as an accounting method.) See also "Apparent Failure To Log All 'Deny' or 'Permit' Matches" in the section "ACL Problems," in appendix "Troubleshooting" of the latest Management and Configuration Guide for your switch.



When configuring logging, you can reduce excessive resource use by configuring the appropriate ACEs to match with specific hosts instead of entire subnets. For more information on resource usage, see page 172.

Minimum number of ACEs in an IPv6 ACL. An IPv6 ACL must include at least one ACE to enable traffic screening. An IPv6 ACL can be created "empty", that is, without any ACEs. However, if an empty ACL is applied to an interface, the Implicit Deny function does not operate, and the ACL has no effect on traffic. Monitoring shared resources. Applied ACLs share internal switch resources with several other features. However, if the internal resources become fully subscribed, additional ACLs cannot be applied until the necessary resources are released from other applications. For information on determining current resource availability and usage, see appendix, "Monitoring Resources" in the latest Management and Configuration Guide for your switch. See also the appendix "Scalability and System Maximums" in the same guide. Protocol support. ACL criteria does not include use of MAC address information or QoS. Replacing or adding to an active IPv6 ACL policy. If you assign an IPv6 ACL to an interface and subsequently add or replace ACEs in that ACL, each new ACE becomes active when you enter it. If the ACL is configured on multiple interfaces when the change occurs, the switch resources must accommodate all applications of the ACL. If there are insufficient resources to accommodate one of several ACL applications affected by the change, the change is not applied to any of the interfaces and the previous version of the ACL remains in effect. See “Monitoring shared resources” (page 171). "Strict" IPv6 TCP and UDP. When the IPv6 ACL configuration includes TCP or UDP options, the switch operates in "strict" TCP and UDP mode for increased control. In this case, the switch compares all IPv6 TCP and UDP packets against the IPv6 ACLs. Connection-rate ACLs. As of software release K.13.01, this ACL connection-rate ACLs? are supported for IPv4 ACLs, but not for IPv6 ACLs.

General ACL operating notes

171

Unable to Delete an ACL in the Running Configuration Attempting to delete an ACL that is currently assigned to an interface removes all configured ACEs from the ACL, but leaves an "empty" ACL in the configuration. To delete an ACL that is currently assigned to an interface, do the following: 1. In the interface context, use the no ipv6 access-group command to remove the ACL from the interface. 2. Use the no ipv6 access-list name-str command to delete the ACL. The no vlan vid ipv6 access-group name-str vlan command does not delete the named ACL if the ACL is currently assigned to an interface.

172

IPv6 Access Control Lists (ACLs)

6 IPv6 Routing Basics Table 20 Summary of commands Command syntax

Description

Default

CLI page reference

show ip ospf (if OSPFv2 is enabled)

Displays router IDs.

-

175

[no] ip router-id n.n.n.n

Configures a router ID for use by any dynamic routing protocol configured on the routing switch.

-

175

[no] ipv6 hop-limit 1 - 255

Sets the maximum number of routers (hops) through which packets originating on the routing switch can pass before being discarded (global hop limit).

64

176

ipv6 route ::/0 ipv6-gateway-addr distance 1 - 255 [no] ipv6 route ::/0 ipv6-addr

Used in the global config context to configure the default route and gateway to use for traffic sent to the default route.

1

177

show ipv6 route [ ipv6-destination [ connected | static | ospf3 ]] [ connected ipv6-destination ] [ static ipv6-destination ] [ ospf3 ipv6-destination ]

Displays the current IPv6 routing table content for the routing switch.

-

177

show ipv6 ospf3 (if OSPFv3 is enabled)

IPv6 routing overview Beginning with software release K.15.01, the switches support these IPv6 routing features: •

“IPv6 Static Routing” (page 187)



“IPv6 Router Advertisements” (page 193)



“DHCPv6-Relay” (page 211)



“OSPFv3 Routing” (page 218)

This chapter covers basic IPv6 routing topics and configuration needed to implement both static and dynamic IPv6 routing. Command

Min. Context

Page

show ipv6 ospf3 (for router ID)

n/a

175

[no] ip router-id n.n.n.n show ipv6 ospf3 general

global config

175

[no] ipv6 hop-limit 1 - 255

global config

176 IPv6 routing overview

173

ipv6 route ::/0 (default route) [ ipv6-destination [ connected | ]] [ connected ipv6-destination ] [ static ipv6-destination ] [ ospf3 ipv6-destination ]

static |

global config

177

n/a

177

ospf3

NOTE: To use the switch in IPv6 host-only mode in an IPv6 routing environment or for information on configuring IPv6 addresses, see: •

“Enabling autoconfiguration of a global unicast address and a default router identity on a VLAN” (page 16)



“Router access and default router selection” (page 35)

In the context of the routing operation supported on the HP switches, the switches are referred to as routing switches. Routing topics and terminology in this chapter refer to IPv6 routing only. For information on IPv4 routing, see the latest Multicast and Routing Guide for your routing switch. For more information on the technical publications for your routing switch, see the “Support and Other Resources” appendix.

Dual stack IPv4/IPv6 operation The switches support IPv4/IPv6 dual-stack operation. This allows full ethernet link support for switching and routing both IPv4 and IPv6 traffic on the same VLAN interfaces configured on the switch without modifying current IPv4 network topologies. This enables you to use IPv6 devices on existing VLANs, manage the switches and other devices from IPv6 management stations, and create dedicated groups of IPv6 devices as needed to accommodate the need for the IPv6 network growth anticipated for the future. For more overview information on IPv6 routing, see the sections beginning with “IPv6 networks and subnets” (page 179). NOTE: Software license requirements: For the 3500/3500yl, 5400zl, and 8200zl switches, OSPFv3 is included with the Premium software license available from HP. In the 6200yl switches, OSPFv3 is included with the base feature set.

IPv6 Routing Features Beginning with software release K.15.01, the routing switches covered by this guide support the following IPv6 routing features:

174

Routing Feature

Application

Further Information

Enabling IPv6 Routing

Configured at the global config level on individual routing switches to enable IPv6 routing.

“Enabling DHCPv6” (page 17)

IPv6 Static Routes

In smaller networks, static and default routes provide the simplest and most reliable configuration for IPv6 routing. Up to 256 static routes are supported.

“IPv6 Static Routing” (page 187)

Router Advertisement (RA) Generation

In IPv6 routed networks this capability enables use of the routing switch to generate the router advertisements IPv6 hosts use to configure themselves with the global unicast addresses and other parameter settings needed to for sending and receiving traffic. Enabled with default settings when “IPv6 Router IPv6 routing is enabled. Reconfigure as needed. Advertisements” (page 193)

IPv6 Routing Basics

DHCPv6 Relay

In routed networks, this feature enables you to use routing to extend the “DHCPv6-Relay” service range of your DHCPv6 server to destinations reached by routing. (page 211)

For larger and more complex routed networks where reliance solely on static routing is not feasible, OSPFv3 enables the routing switch to discover OSPFv3 dynamic routing information from other routers, and to segment an enterprise “OSPFv3 Routing” routing network into routing domains. (page 218)

Viewing the router ID Use one of the following commands: show ip ospf (if OSPFv2 is enabled) show ipv6 ospf3 (if OSPFv3 is enabled) Example 115 Displaying the router ID when OSPFv3 is enabled If a routing switch is using a router ID such as 10.10.10.1: HP Switch(config)# show ipv6 ospf3 OSPFv3 General Status OSPFv3 Protocol Router ID

NOTE:

: Enabled : 10.10.10.1

If one of the following is true, the router ID is set to 0.0.0.0:



No manual router ID, IPv4 network address, or IPv4 loopback interface address is configured on the routing switch.



No dynamic routing protocol is enabled on the routing switch.

Manually configuring a router ID For more information about configuring a router ID, see “Configuring a router ID” (page 179). You can override the current router ID assignment by explicitly configuring the router ID to any valid IPv4 address. (This address must be unique; not configured on another device in the network.)

Syntax: [no] ip router-id n.n.n.n This optional command configures a router ID for use by any dynamic routing protocol configured on the routing switch. The assignment applies across all routable VLANs on the routing switch, and uses the IPv4 32-bit dotted decimal format. The resulting IPv4 address must be unique in the network. The no ip router-id command replaces a manually configured router ID with the first-detected IPv4 network address configured on a VLAN. If an IPv4 address is not detected on a VLAN, then the lowest-numbered IPv4 address of the lowest-numbered IPv4 loopback interface is used.

Changing an existing router ID 1.

Go to the global config context. The CLI prompt will appear similar to the following: HP Switch(config)#_ Viewing the router ID

175

2.

Use ip router-id n.n.n.n to specify a new router ID. This must be in the IPv4 (32-bit) dotted-decimal address format and must be unique in the routing switch configuration and your network. For example: •

If both OSPFv2 and OSPFv3 are enabled, you will see the following: OSPFv3 and OSPFv2 are running. Router-id will be applied immediately. Protocol will be reset. Continue [y/n]? Press Y to continue, or press N to exit from the command without changing the ID.



If a routing protocol is not enabled, the new ID is entered and the global config prompt is displayed.

Viewing a manually configured router ID Use one of the following commands: show run show ip ospf (if OSPFv2 is enabled) show ipv6 ospf3 (if OSPFv3 is enabled)

Configuring the IPv6 hop limit Syntax: [no] ipv6 hop-limit 1 - 255 Global config operation This global config command sets the maximum number of routers (hops) through which packets originating on the routing switch can pass before being discarded (global hop limit). Each router decrements a packet's hop limit by 1 before forwarding the packet. If decrementing the hop limit causes it to go to 0 (zero), the decrementing router drops the packet instead of forwarding it. Effect on the Hop Limit Included in Per Router Advertisements (RAs) If the routing switch is enabled to send RAs on a given IP routing interface, and that interface’s RA configuration does not include a hop limit entry, then the global hop limit configured by this command is inserted in the RAs sent from the routing switch on that interface. But if the interface’s RA configuration does include a local hop limit entry, then the global config hop limit is replaced by the local hop limit entry configured for inclusion in RAs sent by the routing switch on that interface. For information on configuring a hop limit included in RAs sent on an IP routing interface, see “Setting or changing the hop-limit for host-generated packets” (page 196). Default: 64; Range: 1 - 255 The no form of the command resets the global hop-limit to the default 64. If hop-limit is set to a non-default value, you can view the current setting by using the show run command. (When set to the default value, hop-limit does not appear in the show [ run | config ] command output.

176

IPv6 Routing Basics

Configuring the IPv6 default route The IPv6 default route (::/0) is a static route used for all traffic that has a destination network not reachable through any other IPv6 route in the routing table. For more information on static routes, see “IPv6 Static Routing” (page 187).

Syntax: ipv6 route ::/0 ipv6-gateway-addr distance 1 - 255 ipv6 route ::/0 ipv6-addr Used in the global config context to configure the default route and gateway to use for traffic sent to the default route. ::/0 Specifies the default IPv6 route. ipv6-gateway-addr Specifies the next-hop router for traffic sent to the default route. distance 1 - 255 Specifies the administrative distance to associate with a static route. For more on this topic, see “Metric and administrative distance” (page 185). Default: 1; Range: 1 - 255 The no form of the command deletes the default route for the specified next-hop destination from the routing table. Example 116 Configuring the IPv6 default route If 2001:db8:c::9f:35 is the IPv6 address of your ISP router, all non-local traffic could be directed to the ISP by configuring the following default route: HP Switch(config)# ipv6 route ::/0 2001:db8:c::9f:35 To view the default route in the routing table, use show ipv6 route. See Example 117 (page 178).

Viewing the IPv6 routing table The routing table automatically includes directly connected networks (loopback interfaces and destinations on the routing switch itself). Other routes are added when discovered and determined by the routing switch to be the best route to the given destination. For more information about the routing table, see “Different route types in the IPv6 routing table” (page 183).

Syntax: show ipv6 route [ ipv6-destination [ connected | static | ospf3 ]] [ connected ipv6-destination ] [ static ipv6-destination ] [ ospf3 ipv6-destination ] Displays the current IPv6 routing table content for the routing switch. Where there are multiple routes to the same destination, only the route with the lowest administrative distance is entered in the routing table and used to forward traffic to that destination. The complete IPv6 route table is displayed by entering the CLI command show ipv6 route from any context level in the console CLI. Configuring the IPv6 default route 177

Example 117 Viewing the IPv6 routing table HP Switch(config)# show ipv6 route IPv6 Route Entries Destination : ::/0 Gateway : 2001:db8:e::55:2 Type: static Sub-Type: NA Distance: 130 1 Metric: 1 Destination : ::1/128 Gateway : lo0 Type: connected Sub-Type: NA

Distance: 0

Metric: 1

Destination : 2001:db8:1::127/128 Gateway : lo6 Type: connected Sub-Type: NA

Distance: 0

Metric: 1

Destination : 2001:db8:a::/64 Gateway : fe80::22:1%vlan22 Type: ospf3 Sub-Type: InterArea 2 Distance: 110

Metric: 2

Destination : 2001:db8:b::/64 Gateway : VLAN22 Type: connected Sub-Type: NA

Distance: 0 3

Metric: 1

Destination : 2001:db8:c::/64 Gateway : 2001:db8:e::55:2 Type: static Sub-Type: NA

Distance: 120

Metric: 1

1 2 3

The Default Route Configured with a Non-Default Distance OSPFv3 Route, with “InterArea”Sub-Type and Default Distance Static Route with Non-Default Distance

Enabling IPv6 routing 1.

On each VLAN, configure stateless address autoconfiguration and at least one IPv6 global unicast address: vlan n ipv6 address autoconfig vlan n ipv6 address prefix/prefix-length eui-64 These commands result in a link-local address and a global unicast address having an interface ID derived from the routing switch's MAC address. For more information on this topic, or to manually configure link-local and global unicast addressing with a non-EUI interface ID, see “Statically configuring a link-local unicast address” (page 20).

2.

Suppress automatic (default) RAs on VLAN interfaces where you need to make configuration changes or where you do not currently want these advertisements generated: •

To globally suppress RAs on the routing switch, use this command in the global config context: ipv6 nd suppress-ra



To suppress RAs on individual VLANs, use this command in the context of each VLAN where you want the advertisements suppressed: ipv6 nd ra suppress The no form of the above two suppress commands disables RA suppression. For more information on RAs, see “IPv6 Router Advertisements” (page 193).

178

IPv6 Routing Basics

3.

Enable IPv6 routing. (This command enables RA transmission on any VLAN where RAs are not specifically suppressed.) ipv6 unicast-routing

4.

For non-default RA operation, configure RAs per-VLAN, including suppression of RAs on any VLANs where you do not want the routing switch to transmit RAs. See “IPv6 Router Advertisements” (page 193). Configure one or more of the following routing features:

5.



IPv6 static routing. See “IPv6 Static Routing” (page 187).



DHCPv6-relay. See “DHCPv6-Relay” (page 211).



OSPFv3. See “OSPFv3 Routing” (page 218).

Configuring a router ID For more information on configuring routers, see “Configuring global IPv6 routing parameters” (page 183). When router ospf3 enable is used to enable OSPFv3 on a routing switch, the following message appears if an IPv4 network or loopback address is not detected: Either an IPv4 loopback address or a router ID needs to be configured before enabling OSPFv3. Do you want to continue? [y/n]

Typing N (no) keeps OSPFv3 disabled. Typing Y allows OSPFv3 to be enabled without a router ID. However, without the ID, OSPFv3 traffic is not routed. In this case do one of the following: •

Configure an IPv4 address on a loopback interface or VLAN.



Manually configure a router ID on the routing switch.

IPv6 networks and subnets An IPv6 network is a group of hosts and routers that share a common network prefix and exist on the same VLAN interface. Where multiple unique network prefixes exist on the VLAN, each prefix corresponds to a different subnet. For example, if a given network has a prefix of 2001:db8:1ad:27b::/64, any global unicast address for an individual device belonging to this network has: •

The same prefix (2001:db8:1ad:27b::/64).



A unique value (for the interface ID) in the remaining 64 bits of the global unicast addresses.

In the above case, if device "A" has an interface ID of 218:71ff:fedd:cf00, its complete global unicast address is: 2001:db8:1ad:27b:218:71ff:fedd:cf00/64 1

1

2

Prefix

3

2

Interface ID

3

Prefix Length

Traffic between hosts on the same network is switched and employs link-local addresses that include a reserved prefix (FE80::/64) and a unique interface ID generated from the device MAC address. See “IPv6 Addressing Configuration” (page 11). Continuing the example from above, device "A" uses the following link-local address for switching: However, when a packet must be sent from one network to another, where the source and destination have different IPv6 network prefixes, the packet must be routed. For example, routing is required to send traffic between the devices at these two addresses residing in different networks: 2620:0:a03:e102:218:71ff:fedd:cf00/64 1

Configuring a router ID

179

2001:0:db8:17fd:218:71ff:fedd:cf00/64 2

1

Prefix

2

VLANs and routing IP Routing Interfaces and Routing. On the routing switches covered by this guide, IPv6 addresses are associated with individual IP routing interfaces. Link-local addresses are used for switching traffic among devices on the same IP routing interface, and global unicast addresses are used for routing traffic between different IP routing interfaces.

Link-local Only one link-local IPv6 address, such as fe80::215:60ff:fe7a:adc0, can be configured on a given VLAN on a routing switch.

Global unicast Multiple global unicast addresses can be configured on the same VLAN interface, as long as the network prefix for each address occurs only once on the routing switch. For example, you can configure both of the following addresses on either the same VLAN or on different VLANs: 2001:db8:0:1f::1:6/64 (prefix = 2001:db8:0:1f::/64) 2001:db8:0:2f::1:6/64 (prefix = 2001:db8:0:2f::/64) That is, for global unicast addressing: •

The same interface identifier can be used with multiple, unique network prefixes (and the link-local address) on any VLAN.



Different VLANs must be configured with different network prefixes.



Only one instance of a given network prefix can be configured on a routing switch.

To summarize these rules: IPv6 address type

Limit

Application

link-local

one per IP routing interface

Can be either unique or a duplicate of link-local addresses configured on other VLANs on the routing switch.

global unicast

multiple per IP routing interface

The network prefix must be unique for each global unicast address configured on a given routing switch.

For the maximum number of IPv6 addresses configurable on the routing switch and on a given IP routing interface, see the "Scalability" appendix in the Management and Configuration Guide for your routing switch.

IPv6 management interface In the default configuration, there is a single VLAN (Default_VLAN; VID=1) on the routing switch. With only the default VLAN configured, a single link-local IPv6 address serves as the management access address for the entire device. If routing is enabled, the global unicast address on this VLAN also acts as the routing interface.

180 IPv6 Routing Basics

IPv6 routing operation A switch moves packets within the same local network or subnet. A router moves packets between networks or subnets. When a router receives a packet, it matches the packet's destination address to a route in its routing table. This route specifies the gateway, or next-hop, through which the router must forward the packet to enable it to move toward its destination. The gateway is specified as either the next-hop link-local address on a shared VLAN, the ID of that VLAN (vid), or a global unicast address. For example, in Figure 13 (page 181) the gateway for a packet moving through router "D" to the 2001:db8:0:9::/64 network (router "A") is either VLAN 3 or the link-local address FE80::3:2/64 in VLAN 3 on router "C." Figure 13 Example of a routing domain

A routing switch maintains a routing table containing the best routes to the destinations it has acquired. The routing table can be built from statically configured routes and dynamically configured OSPFv3 routes. Static routes must be manually configured and are best suited for small networks having few routes and where topology changes are infrequent. A dynamic routing protocol such as OSPFv3 offers scalable control for discovering and assessing reliable routes. The best route to a given destination may change over time, and dynamic routing protocols can react to such changes by replacing routes in the routing table. Dynamic routing also adapts to changes in network topology, while static routing requires manual configuration changes to support topology changes.

Concurrent static and dynamic routing operation Static and dynamic routing can operate concurrently in a network. Where there are both static and dynamic routes to the same destination, the routing switch selects the route with the lowest administrative distance for inclusion in the routing table. If the selected route goes down, the routing switch can replace it with a previously existing alternative route.

Router Advertisements (RAs) RAs are used in both static and dynamic routing environments and are transmitted per-IP routing interface from the routing switch for host configuration. RAs carry configuration settings for parameters, such as network prefix and neighbor discovery, and are also used to direct hosts to DHCPv6 servers for configuration settings. Enabling IPv6 routing automatically enables RA transmission on all IP routing interface where it has not been suppressed with one of the following commands: Global Config Context: HP Switch(config)# ipv6 nd suppress-ra VLAN Config Context: HP Switch(vlan-1)# ipv6 nd ra suppress Tunnel Config Context: HP Switch(tunnel-3)# ipv6 nd ra suppress IPv6 routing operation

181

For more information on this topic, see “IPv6 Router Advertisements” (page 193).

DHCPv6-relay When a host on a given VLAN is configured to acquire configuration settings from a DHCPv6 server, it transmits a DHCPv6 request on the VLAN. If there is no DHCPv6 server on the VLAN, you can route the host request to a server on another VLAN by enabling a DHCPv6-relay on the routing switch and configuring a helper address on the VLAN. The settings in the managed-config-flag and other-config-flag RA options override an enabled DHCPv6 option in a host on the VLAN. For more information on DHCPv6-relay, see “DHCPv6-Relay” (page 211). For information on enabling DHCPv6 requests on a host, see “Enabling DHCPv6” (page 17).

General Steps for Enabling IPv6 Routing The following steps provide a guide for enabling IPv6 routing on the switches covered by this guide 1. On each VLAN configure stateless address autoconfiguration and at least one IPv6 global unicast address: vlan n ipv6 address autoconfig vlan n ipv6 address prefix/prefix-length eui-64 The above commands result in a link-local address and a global unicast address having a interface ID derived from the routing switch’s MAC address. For more on this topic, or to manually configure link-local and global unicast addressing with a non-EUI interface ID, see “Statically configuring a link-local unicast address” (page 20). 2.

Suppress automatic (default) router advertisements (RAs) on IP routing interfaces where you need to make configuration changes or where you do not currently want these advertisements generated: •

To globally suppress routing advertisements on the routing switch, use this command in the global config context: ipv6 nd suppress-ra



To suppress routing advertisements on individual IP routing interfaces, use this command in the context of each IP routing interface where you want the advertisements suppressed: ipv6 nd ra suppress The no form of the above two suppress commands disables RA suppression. For more on RAs, see “IPv6 Router Advertisements” (page 193).

3.

Enable IPv6 routing. (This command enables RA transmission on any IP routing interface where RAs are not specifically suppressed.) ipv6 unicast-routing

4.

For non-default RA operation, configure RAs per-IP routing interface, including suppression of RAs on any IP routing interfaces where you do not want the routing switch to transmit RAs. (See “IPv6 Router Advertisements” (page 193)) Configure one or more of the following routing features:

5.

182



IPv6 static routing, see “IPv6 Static Routing” (page 187).



DHCPv6-Relay, see “DHCPv6-Relay” (page 211).



OSPFv3, see “OSPFv3 Routing” (page 218).

IPv6 Routing Basics

Configuring global IPv6 routing parameters Feature

Default and range

Page

IPv6 hop-limit

255 (1 - 255)

176

Default network route

None configured

177

Router ID

Lowest-numbered address on the lowest-numbered routing interface

183

The following sections describe how to configure the above global IPv6 routing parameters. NOTE: This section describes how to configure IPv6 parameters for routing switches. For host-based IPv6 configuration information (Ipv6 routing not enabled), see “IPv6 Addressing Configuration” (page 11).

System router ID Each routing switch uses a unique router ID to identify itself when exchanging route information with other devices. This ID is formatted as a 32-bit dotted-decimal (IPv4 format) number. For example: 10.100.215.1 An automatically assigned ID is used unless overridden by a manually configured ID.

Automatic router ID selection The first detected IPv4 address becomes the router ID. Prior to a reboot, this can be any IPv4 address configured on the routing switch. Following a reboot, the router ID is set to the lowest IPv4 loopback address detected. If multiple IPv4 loopback addresses are detected at reboot, the address configured on the lowest-numbered IPv4 loopback interface (lo0 through lo7) becomes the router ID. If the lowest-numbered loopback interface has multiple IPv4 addresses, the lowest of these addresses is selected as the router ID. Once a router ID is selected, it does not automatically change unless a higher-priority address is configured on the routing switch and the routing protocol is restarted with a reboot.

Different route types in the IPv6 routing table Default route (::/0) A static route used by all traffic that has a destination network not reachable through any other IPv6 route in the routing table. See “Configuring the IPv6 default route” (page 177). Directly-connected routes Destinations on the router itself. One route is automatically entered per configured IPv6 interface. Each such route is automatically assigned an administrative distance of "0" and a metric of "1". Directly-connected routes include: IP routing interface Where the routing switch is connected to a next-hop router on the same interface, a route is automatically entered for the network on which the IP routing is configured. This includes destinations for both global unicast and link-local addresses configured on the routing switch for that interface. Manually configured IPv6 loopback interfaces IPv6 loopback interfaces that are manually configured.

Configuring global IPv6 routing parameters 183

Loopback route A static IPv6 route automatically created in the routing table for use if other routes to a destination are not available. The gateway is a loopback interface (lo0) and the destination is ::1/128. Statically configured routes On a given routing switch, one static route can be configured directly into the routing table for each destination. In the default configuration, administrative distance and route metric are both "1". See “About static routing ” (page 189). OSPFv3 If OSPFv3 is enabled, the routing switch learns of routes from the advertisements other OSPFv3 routers transmit. If the OSPFv3 route has a lower administrative distance than any other routes from different sources to the same destination, the routing switch places the route in the IPv6 route table. See “Metric and administrative distance” (page 185).

Routing table content A routing protocol such as OSPFv3 develops its own database of routes. When the protocol has more than one route to a destination, it selects the route with the lowest administrative distance and inserts this route into the routing table. For each such route, the routing table maintains the following data: Parameter

Use

Destination (IPv6 network prefix)

Composed of the contiguous, high-order bits in a packet's destination network prefix that must match the destination network prefix in the routing table entry. For example: Address: Prefix:

2001:db8:1ad:0:218:71ff:fedd:cf00/64 2001:db8:1ad:0/64

Address: Prefix:

2626:17b:1:1: 218:71ff:fedd:cf00/48 2626:17b:1/48

Gateway

The next-hop router in the path to the destination. It can be either the IPv6 address of the next directly connected router or the IP routing interface to use for forwarding the routed traffic toward its destination. If an IPv6 address is used, it can be either the link-local or global unicast address of the interface on the next-hop router.

Type (route type)

Connected: A destination configured on the routing switch itself and can be a loopback interface, a global unicast address, or a link-local address. Static: A manually configured route to a destination on another router. OSPF3: A route discovered by the OSPFv3 protocol running on the routing switch.

Sub-type

Applies to OSPFv3 routes only.

Distance (Administrative Distance)

Used to compare routes to the same destination, learned by different routing methods, to select the best route. The distance for connected routes is always 0. The default distance for static and dynamic routes is configurable; Default: 1. See also “Concurrent static and dynamic routing operation” (page 181).

Metric

Calculated by the routing switch and used to compare different routes, learned by the same routing method, to select the best overall route.

Destination network Destination network prefixes identify the networks known to a routing switch. When the routing switch receives a packet for routing, it matches the packet's destination address to a network prefix in the routing table and forwards the packet to the indicated gateway for that network. The prefix 184 IPv6 Routing Basics

in the routing table defines how many leftmost contiguous bits to use when matching a packet's destination address to a destination network prefix. For example, a route table entry of 2001:db8:0:1ad:0:f1:7a:0/112 applies to all packets with a destination address for which the first 112 bits are 2001:db8:0:1ad::f1:7a If a packet matches more than one routing table entry, the router uses the most specific route (the route with the longest prefix), which is assumed to be the most accurate for that packet. For example, for the packet destination listed below, both route table entries apply, but the route selected will be the 72-bit entry, because it is the more specific route. Packet destination address: 72-bit entry in route table: 64-bit entry in route table:

2001:db8:0:1d5:a15::f:101/64 2001:db8:0:1d5:a00::/72 2001:db8:0:1d5::/64

Gateway for forwarding routed traffic The gateway to a destination network can be either of the following: •

Global unicast or link-local address of the next-hop router in the IP routing providing a path to the packet destination.



The IP routing interconnecting the originating router to the next-hop router.

Metric and administrative distance The routing table contains the single best route to each destination that the router has learned. However, a router may learn more than one route to the same destination. The router compares the metrics and administrative distances of these routes to select the best route to add to its routing table. •

Administrative distance The routing switch uses this parameter to compare routes learned by different routing methods. It indicates how reliable the router considers the method through which it discovered the route: a lower value indicates a more trustworthy route. Administrative distance is not a factor if you are using only static routes. However, if you are using static routing in conjunction with a routing protocol such as OSPFv3 to provide routes to an identical destination, the routing switch selects the route with the lowest administrative distance. Where the default administrative distances are used, a static route normally supersedes a dynamic route to the same destination because the former has the lowest default administrative distance and metric. Routing method

Administrative distance

Metric

Direct connection

0 (not configurable)

1

Static route

Default: 1; range: 1 to 255

1

OSPFv3

Default: 110 (external, inter-area, and intra-area) range (for all three): 1 to 255

Variable

Different route types in the IPv6 routing table 185





To configure administrative distance for a static route, see “Adding static and null routes to IPv6 table” (page 187).



To configure administrative distance for OSPFv3, see “Influencing route choices by changing the administrative distance default” (page 231).

Metric The routing switch uses this parameter to compare routes to identical destinations learned by the same routing protocol. The metric is the cost of sending traffic on a given route and is based on various criteria: •

Link conditions (bandwidth, delay, reliability)



Organizational policies (monetary cost, autonomous systems that a packet must traverse) Each routing protocol has its own method for computing a route's metric. For static routes, the metric defaults to "1" and is not configurable.

186 IPv6 Routing Basics

7 IPv6 Static Routing Table 21 Summary of commands Command syntax

Description

Default

CLI page reference

[no] ipv6 route dest-ipv6-addr/prefix-length [ next-hop-gateway-addr | vlan vid | tunnel tunnel-id | blackhole | reject ] [ distance 1 - 255 ]

Enables you to create static routes (including null routes with or without ICMP notification to the sender) by adding such routes directly to the route table in the routing switch.

-

187

show ipv6 route [ ipv6-addr | connected | static | ospf3 ]

Lists all entries in the IPv6 routing table.

-

188

For conceptual information about static routing, see “About static routing ” (page 189).

Adding static and null routes to IPv6 table This feature enables you to create static routes (including null routes with or without ICMP notification to the sender) by adding such routes directly to the route table in the routing switch.

Syntax: [no] ipv6 route dest-ipv6-addr/prefix-length [ next-hop-gateway-addr | vlan vid | tunnel tunnel-id | blackhole | reject ] [ distance 1 - 255 ] dest-ipv6-addr/prefix-length Network prefix for the destination on IPv6 address. [ next-hop-gateway-addr | vlan vid | tunnel tunnel-id ] The gateway for reaching the destination. The next-hop address option (link-local or global unicast) is not required to be directly reachable on a local subnet. (If it is not directly reachable, the route is added to the routing table when a path to this address is learned.) If the next-hop address is link-local, it must include both the address and the applicable VLAN VID. For example: FE80::127%vlan10,where VLAN 10 is the interface where FE80::127 exists. For a tunnel, it would be FE80::127%tun3. blackhole Specifies a null route where IP traffic for the specified destination is discarded and no ICMP error notification is returned to the sender. reject Specifies a null route where IP traffic for the specified destination is discarded and an ICMP error notification is returned to the sender. distance 1 - 255 Specifies the administrative distance to associate with a static route. For more information on this topic, see “Metric and administrative distance” (page 185). Default: 1; Range: 1 - 255 Adding static and null routes to IPv6 table

187

The no form of the command deletes the specified route from the routing table for the specified destination next-hop pair. Example 118 (page 188) configures two static routes for traffic delivery and identifies two other null routes for which traffic should be discarded instead of forwarded. Example 118 Configuring static routes HP Switch(config)# ipv6 route 2001:db8:0:1::/64 fe80::10.1

Configures static route to a specific destination network . Notice that the next-hop gateway can be either a link-local or a global unicast address. HP Switch(config)# ipv6 route 2001:db8:0:2::/64 reject

Configures a null route to drop traffic for the 2001:db8:0:2::/64 network and return an ICMP notice to the sender. HP Switch(config)# ipv6 route 2001:db8:0:5::/64 blackhole

Configures a null route to drop traffic for the 2001:db8:0:2::/64 network without ICMP notice to the sender. HP Switch(config)# ipv6 route 2001:db8::/48 vlan 66 distance 120

Configures a static route for traffic to destinations in the 2001:db8:0::/48 network. Sets the administrative distance higher than the default distance for any dynamic routes discovered for the same destination, which gives precedence in the routing table to dynamic routes. Switch(config)# ipv6 route 2001:db8:0:3::/64 tunnel 3 distance 130

Configures a static route for traffic to destinations in the 2001:db8:0:3::/64 network.

Viewing static route information Syntax: show ipv6 route [ ipv6-addr | connected | static | ospf3 ] Lists all entries in the IPv6 routing table. ipv6-addr Lists entries for a specific IPv6 address. Can be followed by any of the other options for this command. connected Lists entries for connected routes. Can be followed by the ipv6-addr option to list only the connected routes having a specific link-local or global IPv6 address. static Lists entries for static routes in the routing table. Can be followed by the ipv6-addroption to list only the static routes matching a specific destination. ospf3 Lists the entries for OSPFv3 routes in the routing table. Can be followed by the ipv6-addroption to list only the OSPFv3 routes matching a specific destination. Example 119 (page 189) shows the static routes in the routing table for Router "B" in Figure 14 (page 190). 188 IPv6 Static Routing

Example 119 Displaying static routes in the IPv6 routing table HP Switch(config)# show ipv6 route static IPv6 Route Entries Destination : 2620:a::/64 Gateway : 2620:b::22:1 Type : static Sub-Type : NA

Distance : 1

Metric : 1

Destination : 2620:c::/64 Gateway : 2620:e::55:2 Type : static Sub-Type : NA

Distance : 1

Metric : 1

About static routing Static routes provide tools for restricting and troubleshooting routed traffic flows and in small networks can provide the simplest and most reliable configuration for IPv6 routing. Static routes are manually configured in the routing table. A static route entry comprises the following: •

IPv6 network prefix for the route's destination network



Next-hop gateway, which can be one of the following:





Either the link-local address and VLAN ID or the VLAN link to the next-hop router



Global unicast address on the next-hop router



A "null" interface (the routing switch drops traffic forwarded to the null interface)

Optionally, a nondefault administrative distance

NOTE: To enable routing in both directions on a static route, you must configure reciprocal static routes on the routers at both ends of the route. On a given routing switch you can create one static route or null route to a given destination. Multiple static or null routes to the same destination are not supported. The routing switches can concurrently support a maximum of 256 IPv6 static routes and 256 IPv4 static routes. For example, in Figure 14 (page 190), static routes enabling routed traffic between routers "A," "B," and "C" could be configured as follows: Table 22 Example of static route configuration in a network Router "A"

Router "B"

Router "C"

ipv6 route

ipv6 route

ipv6 route

2620:a::/64

2620:a::/64 2620:b::22:1

2620:c::/64 2620:b::22:2

ipv6 route

ipv6 route

ipv6 route

2620:b::/64

2620:c::/64 2620:e::55:2

2620:e::/64 2620:b::22:2

2620:e::55:1

2620:e::55:1 Note: Next-hop addresses can be either global unicast or link-local.

About static routing 189

Figure 14 Example of a routing domain

Advantages Static routing is relatively reliable and gives you tight control over traffic flow. You determine exactly which connections to use to forward traffic to each destination. In a given VLAN, you can use multiple IPv6 addresses to add multiple static routes in the VLAN. Other advantages include: •

Efficiency in a small network with few paths to manage



Ease of configuration and maintenance



Lower CPU utilization

Disadvantages In a large or expanding network, configuring static routes for all the necessary routes can become increasingly complicated and time-consuming. Ensuring that all routes remain accurate can also add to the administrative burden. Each time you add a connection or change a route, you must configure the change on every routing device in the network. Also, routers do not automatically respond to a failed static connection, so traffic can be lost or misrouted. NOTE: Network management and monitoring applications such as HP PCM/PCM+ can detect failed static routes.

Static route types You can configure these types of static IPv6 routes: Standard The static route consists of •

Destination network prefix



Link-local IPv6 address and VLAN ID of the (next-hop router) gateway IPv6 address

Interface-based The static route consists of: •

Destination network address or host and a corresponding network prefix



VLAN interface through which you want the routing switch to send traffic for the route

Null (discard) Null routes include the following: Default When IPv6 routing is enabled, a route for the ::1/128 network is created and traffic to this network is rejected (dropped). The loopback address (lo0) is entered as the gateway. This route is for all traffic to the "loopback" network, with the single exception of traffic to the host address of the switch's loopback interface.

190 IPv6 Static Routing

Configured Provides a route that is used as a backup route for discarding traffic where the primary route is unavailable. A configured null route consists of: •

Destination network address or host and a corresponding network mask



Either the reject keyword (traffic dropped with ICMP notification to the sender) or blackhole keyword (traffic dropped without any ICMP notification).

Non-default null routes created with the reject or blackhole keywords use a gateway of zero (0). Figure 15 (page 192) illustrates the default and configured null route entries in the switch's routing table.

Static routing default settings The routing switch applies default administrative distance and metric values to ensure that static routes are preferred over dynamic routes to the same destination. Administrative distance In the case of static routes, this is the value the routing switch uses to compare a static route to routes from other route sources to the same destination before placing a route in the routing table. The default administrative distance for static routes is 1, but can be configured to any value in the range of 1 to 255. Metric In the case of static routes, this is the value the routing switch uses when comparing a static route to routes in the routing table from any dynamic routes to the same destination. The metric for static routes is fixed, that is, always set to "1".

Static route states follow VLAN states Static routes remain in the routing table only while the interface link to the next-hop router is up. If the next-hop router interface link goes down, the software removes the static route from the routing table. If the next-hop interface comes up again, the software adds the route back to the routing table. This feature allows the routing switch to adjust to changes in network topology. The routing switch does not continue trying to use routes on unreachable paths, but instead uses routes only when their paths are reachable.

Static routes for ECMP applications Equal-cost multi-path routing (ECMP) is a routing strategy where next-hop packet forwarding to a single destination can occur over multiple "best paths." Each path has the same cost as the other paths, but a different next-hop router. In static routing, load-balancing can be achieved through ECMP. Figure 15 (page 192) illustrates static routes applied to an ECMP topology.

About static routing

191

Figure 15 Example of static routes in an ECMP application

VLAN-1

Routing Switch

Router 2001:db8:1::1

Router 2001:db8:1::2

Static Routes ipv6 route 2001:db8:5::/64 2001:db8:1::1

VLAN-2

The no ip load-sharing 2 - 4 command enables or disables load-sharing for both IPv4 and IPv6 applications and specifies the number of ECMP routes to allow. In the default configuration, load-sharing is enabled with four ECMP routes allowed. For more information, see “About equal-cost multi-path routing” (page 266).

192 IPv6 Static Routing

8 IPv6 Router Advertisements Table 23 Summary of commands Command syntax

Description

Default

CLI page reference

[na] ipv6 nd suppress-ra

Globally suppresses IPv6 RAs on the routing switch.

Disabled

194

[no] ipv6 unicast-routing

Globally enables IPv6 Disabled routing on the routing switch.

194

[no] ipv6 nd ra managed-config-flag [no] ipv6 nd ra other-config-flag

Specifies host Disabled configuration method.

194

[no] ipv6 nd ra max-interval 4 - 1800 [no] ipv6 nd ra min-interval 3 - 1350

Sets the maximum and minimum interval for transmitting RAs on the VLAN.

195

[no] ipv6 nd ra hop-limit 0 - 255

Sets the hop-limit a 64 sending host should include in the packets it transmits.

[no] ipv6 nd ra lifetime 0 - 9000

Sets the period for the routing switch to be used by hosts as a default router.

Max: 600 seconds; Min: 200 seconds

196

3x 196 ra max-interval setting

[no] ipv6 nd ra reachable-time [ 1000 - 3600000 | unspecified ]

Sets the reachable unspecified time duration for a (0) neighbor that has sent a reachability confirmation.

196

ipv6 nd ra ns-interval [ 1000 - 4294967295 | unspecified ] [no] ipv6 nd ra ns-interval

Sets the retransmit time for neighbor solicitations or solicitation requests.

unspecified (0)

197

[no] ipv6 nd ra prefix ipv6-prefix/prefix-len [ valid-lifetime preferred-lifetime | at | valid-date preferred-date | infinite | at | no-advertise ] [ no-autoconfig ] [ off-link ]

Specifies prefixes and lifetimes for global unicast addresses in hosts on the VLAN.

Global 197 prefixes on the VLAN. Valid Lifetime: 30 days Preferred Lifetime: 7 days

[no] ipv6 nd ra prefix default [ valid-lifetime preferred-lifetime | at | valid-date preferred-date | infinite | at | no-advertise ] [ no-autoconfig ] [ off-link ] [no] ipv6 nd ra suppress [no] ipv6 ra-guard ports

[log]

show ipv6 nd ra show ipv6 nd ra prefix vlan vid

Enable restriction of Disabled ports that accept RAs.

202

Without the optional keywords, this

204

-

193

Table 23 Summary of commands (continued) Command syntax

Description

Default

CLI page reference

command displays the global and per-VLAN RA (ND) configuration on a specific routing switch.

Beginning with software release K.15.01, the routing switches support IPv6 RA configuration and transmission based on RFC 4861, "Neighbor Discovery for IP Version 6 (IPv6)" and RFC 4862, "IPv6 Stateless Address Autoconfiguration." IPv6 RAs on a VLAN provide the ND policy the system administrator has configured for devices running in IPv6 host mode with address autoconfiguration enabled. RAs also enable hosts on a VLAN to build a list of default (reachable) routers on that VLAN or tunnel. For general RA operation information, see “General RA operation” (page 207).

Global configuration context commands Enabling or disabling IPv6 Router Advertisement generation Syntax: [na] ipv6 nd suppress-ra Global config command to suppress transmission of IPv6 Router Advertisements on all VLANs or tunnels configured on the routing switch. Overrides RAs enabled per-VLAN. The no form of the command globally disables Router Advertisement suppression. Note that globally enabling Router Advertisements on the routing switch does not override per-VLAN or per-tunnel IPv6 Router Advertisement suppression (using the ipv6 nd ra suppress command in a VLAN or tunnel context). See “Suppressing Router Advertisements on a VLAN” (page 202). Default: RA suppression disabled

Enabling or disabling IPv6 routing Syntax: [no] ipv6 unicast-routing Global config command to enable or disable IPv6 routing. Must be enabled for routing operation. Enabling IPv6 routing activates RA generation on VLANs or tunnels unless RAs are suppressed globally or per-VLAN or per-tunnel. The no form of the command disables IPv6 routing and RAs on the routing switch. Default: Disabled

VLAN or tunnel context ND configuration Configuring DHCPv6 service requirements Syntax: [no] ipv6 nd ra managed-config-flag [no] ipv6 nd ra other-config-flag 194

IPv6 Router Advertisements

managed-config-flag Controls the M-bit setting in RAs the router transmits on the current VLAN. Enabling the M-bit directs clients to acquire their IPv6 addressing and ND host configuration information for the current VLAN or tunnel interface from a DHCPv6 server. •

When the M-bit is enabled, receiving hosts ignore the other-config-flag (O-bit) setting described below.



When the M-bit is disabled (the default), receiving hosts expect to receive their IPv6 addressing and ND configuration settings from the RA unless the O-bit is enabled.

other-config-flag Ignored unless the M-bit (above) is disabled in RAs. Controls the O-bit in RAs the router transmits on the current VLAN or tunnel. Enabling the O-bit while the M-bit is disabled directs hosts on the VLAN to acquire their ND configuration settings from a DHCPv6 server and their global unicast prefixes from the RA. The no form of either command turns off (disables) the setting for that command in RAs. NOTE: In the default configuration, both the M-bit and the O-bit are disabled, and a host receiving the RA must acquire its prefix and ND configuration from the RA itself and not from a DHCPv6 server. Default for both settings: Disabled

Configuring the range for intervals between RA transmissions on a VLAN The interval between RA transmissions on a VLAN is a random value that changes every time an RA is sent. The interval is calculated to be a value between the current max-interval and min-interval settings described below.

Syntax: [no] ipv6 nd ra max-interval 4 - 1800 [no] ipv6 nd ra min-interval 3 - 1350 VLAN or tunnel context commands for changing the maximum and minimum intervals between transmissions of IPv6 RAs on the VLAN. These values have one setting per VLAN or tunnel and do not apply to RAs sent in response to a router solicitation received from another device. max-interval Must be equal to or less than the configured lifetime setting; see “Setting or changing the default router lifetime” (page 196). Attempting to set max-interval to a value greater than the configured lifetime setting results in an error message. The no form of the max-interval command returns the setting to its default, provided the default value is less than or equal to 75% of the new maximum interval you are setting. Attempting to set max-interval to a value that is not sufficiently larger than the current min-interval also results in an error message. Default: 600 seconds; Range: 4 - 1800 seconds min-interval Must be less than or equal to 75% of max-interval. Attempting to set min-interval to a higher value results in an error message. VLAN or tunnel context ND configuration 195

The no form of the min-interval command returns the setting to its default, provided the default value is less than or equal to 75% of the current max-interval setting. Default: 200 seconds; Range: 3 - 1350 seconds

Setting or changing the hop-limit for host-generated packets Syntax: [no] ipv6 nd ra hop-limit 0 - 255 hop-limit VLAN context command to specify the hop-limit a host includes in the packets it transmits. A setting of 0 means the hop-limit is unspecified in the RAs originating on the current VLAN. In this case, the hop-limit is determined by the host. The no form of the command resets the hop-limit to zero (unspecified), which eliminates the hop-limit from the RAs originating on the VLAN or tunnel. Default: 64; Range: unspecified 0 - 255

Setting or changing the default router lifetime Syntax: [no] ipv6 nd ra lifetime 0 - 9000 VLAN or tunnel context command for configuring the lifetime in seconds for the routing switch to be used as a default router by hosts on the current VLAN. This setting must be configured to a value greater than or equal to the max-interval setting. A given host on a VLAN or tunnel refreshes the default router lifetime for a specific router each time the host receives an RA from that router. A specific router ceases to be a default router candidate for a given host if the default router lifetime expires before the host is updated with a new RA from the router. A setting of 0 (unspecified) for default router lifetime in an RA indicates that the routing switch is not a default router on the subject VLAN or tunnel. Default: 3 times the ra max-interval setting. Range: unspecified 0 - 9000 seconds

Changing the reachable time duration for neighbors Syntax: [no] ipv6 nd ra reachable-time [ 1000 - 3600000 | unspecified ] VLAN or tunnel context command for all hosts on the VLAN or tunnel to configure as the reachable time duration for a given neighbor after receiving a reachability confirmation from the neighbor. This value is used to ensure a uniform reachable time among hosts on the VLAN or tunnel by replacing the individually configured settings on various hosts on the VLAN. 1000 - 3600000 Reachable time in milliseconds. unspecified Configures the reachable time to zero, which disables the reachable-time setting in RAs on the current VLAN.

196

IPv6 Router Advertisements

The no form also disables the reachable-time setting in RAs on the current VLAN or tunnel. Default: unspecified (0); Range: 1000 - 3600000 ms NOTE: If multiple routers on the same VLAN or tunnel are configured to advertise a reachable time, all such routers should use the same reachable-time setting.

Setting or changing the Neighbor Discovery retransmit timer Syntax: ipv6 nd ra ns-interval [ 1000 - 4294967295 | unspecified ] [no] ipv6 nd ra ns-interval Used on VLAN or tunnel interfaces to advertise the period (retransmit timer) in milliseconds between ND solicitations sent by a host for an unresolved destination, or between DAD neighbor solicitation requests. Increasing this setting is indicated where neighbor solicitation retries or failures are occurring, or in a "slow" (WAN) network. 1000 - 4294967295 An advertised setting in this range replaces the corresponding, locally configured setting in hosts on the VLAN. unspecified Sets the retransmit timer value in RAs to zero, which causes the hosts on the VLAN or tunnel to use their own locally configured NS-interval settings instead of using the value received in the RAs. The no form returns the setting to its default. Default: unspecified (0) ; Range: 1000 - 4294967295 ms NOTE: This is the retransmit timer advertised as a host-specific variable. It is separate from the retransmit timer used by the routing switch for its own ND solicitations (ipv6 nd ns-interval). If multiple routers on the same VLAN or tunnel are configured to advertise an ns-interval (retransmit time), all such routers should use the same NS-interval setting. The default NS-interval setting for IPv6 host operation on HP devices is 1000 ms. When the above command is used with the unspecified option to configure RAs, host devices configured by using the RA maintain their preconfigured NS-interval settings.

Configuring the global unicast prefix and lifetime for hosts on a VLAN These commands define the content of RAs transmitted on a VLAN or tunnel.

Syntax: [no] ipv6 nd ra prefix ipv6-prefix/prefix-len [ valid-lifetime preferred-lifetime | at | valid-date preferred-date | infinite | at | no-advertise ] [ no-autoconfig ] [ off-link ]

Syntax: [no] ipv6 nd ra prefix default

VLAN or tunnel context ND configuration

197

[ valid-lifetime preferred-lifetime | at | valid-date preferred-date | infinite | at | no-advertise ] [ no-autoconfig ] [ off-link ] Options for valid-lifetime preferred-lifetime: Time in seconds: [ 0 - 4294967295 | 0 - 4294967295 ] Specific date and time [ valid-lifetime preferred-lifetime ] valid-lifetime-MM/DD/YY valid-lifetime-HH:MM:SS preferred-lifetime-MM/DD/YY preferred-lifetime-HH:MM:SS at valid-date preferred-date valid-date - MM/DD/[YY]YY] valid-date - HH:MM[:SS} preferred-date - MM/DD/[YY]YY preferred-date - HH:MM[:SS} VLAN or tunnel context command for specifying prefixes for the routing switch to include in RAs transmitted on the VLAN or tunnel. IPv6 hosts use the prefixes in RAs to autoconfigure themselves with global unicast addresses. A host’s autoconfigured address is composed of the advertised prefix and the interface identifier in the host’s current link-local address. valid-lifetime The total time the prefix remains available before becoming unusable. After preferred-lifetime expiration, any autoconfigured address is deprecated and used only for transactions that began before the preferred-lifetime expired. If the valid lifetime also expires, the address becomes unusable. Default: 2,592,000 seconds - 30 days; Range: 0 - 4294967295 seconds preferred-lifetime The span of time during which the address can be freely used as a source and destination for traffic. This setting must be less than or equal to the corresponding valid-lifetime setting. Default: 604,000 second – 7 days; Range: 0 - 4294967295 seconds NOTE: The valid and preferred lifetimes designated in this command are fixed values. Each successive transmission of the same RA contains the same valid and preferred lifetimes. For more information on valid and preferred lifetimes, see “Address lifetimes” (page 30). default Applied to all on-link prefixes that are not individually set by theipv6 ra prefix ipv6-prefix/prefix-len command. It applies the same valid and preferred lifetimes, link state, autoconfiguration state, and advertise options to the advertisements sent for all on-link prefixes that are not individually

198 IPv6 Router Advertisements

configured with a unique lifetime. This also applies to the prefixes for any global unicast addresses configured later on the same VLAN or tunnel. Using default once, and then using it again with any new values results in the new values replacing the former values in advertisements. If default is used without the no-advertise, no-autoconfig, or the off-link keyword, the advertisement setting for the absent keyword is returned to its default setting. NOTE: To configure a prefix as off-link or no-autoconfig , you must enter unique valid and preferred lifetimes with the prefix command (instead of the default command). ipv6-prefix / prefix-len Specifies the prefixes to advertise on the subject VLAN or tunnel. A separate instance of the command must be used for each prefix to advertise. infinite Specifies that the prefix lifetime will not expire. This option sets the valid and preferred lifetimes to infinity. (All bits set to 1; ffffffff.) no-advertise Specifies no advertisement for the prefix. For example, if the routing-switches VLAN or tunnel interface is configured with any prefixes that you do not want advertised on the VLAN, use this command to specify the prefixes to withhold from advertisements on the subject VLAN or tunnel. Default: Advertising enabled. no-autoconfig Disables host autoconfiguration by turning off the A-bit in RAs. This requires hosts to acquire prefixes through manual or DHCPv6 assignments. Depending on the host implementation, a host that was previously configured by an RA to use autoconfiguration will not be affected by a later RA that includes no-autoconfig (unless the host disconnects and reconnects to the network). To re-enable host autoconfiguration (turn on the A-bit in RAs) for a given RA, use ipv6 nd ra prefix again, without invoking no-autoconfig. Default: A-bit turned on— host autoconfig turned on. off-link Sets the (L-bit) prefix information in an RA to indicate that the advertised prefix is not on the subject VLAN. A host that was previously configured using an RA without off-link will not be affected by a later RA that includes off-link (unless the host disconnects and reconnects to the network). Can be used in instances where the prefix is being deprecated, and you do not want any newly brought up hosts to use the prefix. Default: L-bit turned off. The no form of the command deletes the specified prefix from RAs.

VLAN or tunnel context ND configuration 199

Example 120 Using the default command to configure prefix advertisement content Table 24 (page 200) lists the global unicast addresses configured on a VLAN, with original and updated settings configured using the default command. Table 24 Example using the default command to configure prefix advertisement content Address or prefix

Interface

Original lifetime & autoconfig

Updated lifetime & autoconfig

2001:db8:0:f::f1/64

VLAN 100

15 days

30 days

2001:db8:0:b::b1/64

VLAN 100

14 days

25 days

2001:db8:0:c::c1/64

VLAN 100

2001:db8:0:d::d1/64

VLAN 100

Auto: Yes Set in Example 121 (page 201).

Auto: No (Changed in Example 121 (page 201).

2001:db8:0:a::/64

Off-Link

12/31/2010

not updated

at 00:00:01 12/20/2010 at 00:00:01 Auto: Yes

200 IPv6 Router Advertisements

Advertise on VLAN 100? Yes

Example 121 Using the default command to configure and update prefix advertisements HP HP HP HP HP HP

Switch(config)# vlan 100 Switch(vlan-100)# ipv6 address 2001:db8:0:f::f1/64 Switch(vlan-100)# ipv6 address 2001:db8:0:b::b1/64 1 Switch(vlan-100)# ipv6 address 2001:db8:0:c::c1/64 Switch(vlan-100)# ipv6 nd ra prefix default 1296000 1209600 Switch(vlan-100)# show ipv6 nd ra prefix vlan 100

2

IPv6 Neighbor Discovery Prefix Information 3

VLAN Name : VLAN100 IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

Default 15 days 14 days On On On

ipv6 address 2001:db8:0:d::d1/64 4 ipv6 nd ra prefix 2001:db8:0:d::/64 infinite no-autoconfig ipv6 nd ra prefix 2001:db8:0:a::/64 at 12/31/2010 00:00:01 12/20/2010

HP Switch(vlan-100)# HP Switch(vlan-100)# HP Switch(vlan-100)# 00:00:01 off-link 5 HP Switch(vlan-100)#

show ipv6 nd ra prefix vlan 100

IPv6 Neighbor Discovery Prefix Information VLAN Name : VLAN100 6

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

Default 15 days 14 days On On On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:0:a::/64 7 12/31/2010 00:00:01 12/20/2010 00:00:01 Off On On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:0:d::/64 Infinite Infinite On Off On

8

HP Switch(vlan-100)# ipv6 nd ra prefix default 2592000 2160000 no-autoconfig HP Switch(vlan-100)# show ipv6 nd ra prefix vlan 100

9

IPv6 Neighbor Discovery Prefix Information VLAN Name : VLAN100 10

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

Default 30 days 25 days On Off On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:0:a::/64 11 12/31/2010 00:00:01 12/20/2010 00:00:01 Off On On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:0:d::/64 Infinite Infinite On Off On

12

VLAN or tunnel context ND configuration 201

1 2

Global unicast addresses configured on VLAN 100 To enable advertising prefixes of global unicast addresses configured on the VLAN, the default command sets default lifetime, prefix link status (on or off-link), autoconfiguration (Autonomous Flag) status (on or off), and advertisement setting (on or off). NOTE: Applies only to prefixes in global unicast addresses configured on the VLAN and not uniquely configured by the prefix command.

3 4

5

6

7 8 9

10

11 12

Show command displays default prefix mode settings for global unicast addresses configured on VLAN 100 New global unicast address configured on the VLAN. Followed by command to assign unique lifetime and autoconfig setting in the advertisements for this prefix. Link flag and Advertise flag omitted from the command and therefore set to “On” by default. Off-link prefix designated with unique lifetime. Autoconfig (Autonomous) flag and Advertise flag omitted from the command and therefore set to “On” default Show command displays default advertisement settings for prefixes of global unicast addresses configured on VLAN 100 Show command displays unique advertisement settings for 2001:db8:0:a::/64 also configured on VLAN 100 Show command displays unique advertisement settings for 2001:db8:0:d::/64 identified as an off-link prefix For prefixes configured on the VLAN and not specifically addressed by a prefix command, default changes the default lifetime and the autoconfig setting in advertisements for these prefixes. On-Link flag and Advertise flag omitted from the command and therefore set to “On” by default Show command displays changes in default prefix mode settings for global unicast addresses configured on VLAN 100 No change for the on-link prefix specifically configured by a prefix command, and the off-link prefix that is also configured for advertisement on the VLAN

Suppressing Router Advertisements on a VLAN Syntax: [no] ipv6 nd ra suppress VLAN or tunnel context command to turn off (disable) transmission of RAs from the routing switch on the VLAN. The no form of the command turns on (enables) router advertisement transmission from the routing switch on the current VLAN or tunnel. Default: Suppression disable, that is, router advertisement enabled on the VLAN or tunnel.

202 IPv6 Router Advertisements

Restricting IPv6 Router Advertisements The RA Guard feature restricts the ports (or trunks) that can accept IPv6 Router Advertisements (RAs). Additionally, ICMPv6 router redirects are blocked on the configured ports. Only physical ports and trunk ports are supported. Dynamic ports, dynamic trunks, and mesh ports are not supported. NOTE: IPv6 RAs are ICMPv6 type 134 messages and may be sent to either the “all nodes” multicast address (FF02:0:0:0:0:0:0:1) or to the address of the device itself as a result of an IPv6 router solicitation. IPv6 router redirect messages are ICMPv6 type 137 messages. They are sent to the source address of the packet that triggered the redirect.

Configuring RA Guard Syntax: [no] ipv6 ra-guard ports port-list [log] Enables or disable RA Guard on the specified ports, which blocks IPv6 router advertisements and router redirects. The no form of the command disables RA Guard. [log]: Enables debug logging of RA and redirects packets to debug output. Figure 16 Enabling RA Guard

Operating Notes •

When a logical trunk port is enabled, all members of the trunk are enabled for RA Guard. Likewise, when a logical trunk port is disabled, (no ipv6 raguardports ), all members of the trunk are disabled for RA.



When ports are configured for RA Guard, hardware resources are allocated. If there are not enough hardware resources, this message displays: Commit failed



When debug logging is enabled (ipv6 ra-guard ports log), the RA and redirect packets are sent to the CPU, which can be CPU-intensive. This message displays: The log option uses a lot of CPU and should be used only for short periods of time.



The debug security ra-guard command is used to filter and display RA Guard debug log messages.

To display configuration and statistical information aboutRAGuard, enter the show ipv6 ra-guard command. Figure 17 Output Showing Configuration and Statistics for RA Guard

VLAN or tunnel context ND configuration 203

When RA Guard is enabled, there will be one or two lines displayed in the running config file. Figure 18 Running Config File Showing Line for RA-Guard

Displaying the Router Advertisement configuration Syntax: show ipv6 nd ra show ipv6 nd ra prefix vlan vid Without the optional keywords, this command displays the global and per-VLAN, and per tunnel router advertisement neighbor discovery configuration on a specific routing switch. This indicates the per-VLAN or per-tunnel content of RAs transmitted from the routing switch. For a description of each value, see “Global configuration context commands” (page 194) and “VLAN or tunnel context ND configuration” (page 194). prefix Displays the prefixes, valid lifetime, and onlink/auto values advertised by the routing switch on all VLANs or tunnels configured for RA operation. prefix vlan vid Displays values for each prefix configured using ipv6 nd ra prefix on the specified VLAN or tunnel; see page 197. IPv6 Prefix Displays values for specific prefixes configured for RAs on a VLAN or tunnel by the ipv6 nd ra prefix command, plus Default (to apply to any global unicast prefixes on the same VLAN(s) or tunnel(s) that have not been specifically configured by ipv6 nd ra prefix). Valid Lifetime The valid lifetime configured for the indicated prefix. Preferred Lifetime The preferred lifetime configured for the indicated prefix. On-link Flag Indicates whether the prefix is advertised as on-link. Default: On; On-link enabled. 204 IPv6 Router Advertisements

Autonomous Flag Indicates whether address autoconfiguration is turned on. Default: On; Autoconfiguration enabled. Advertise Flag Indicates whether advertisement for the subject prefix is turned on. Default: On.

Displaying the Router Advertisement configuration 205

Example 122 General Output Listing the RA Configuration on a Routing Switch HP Switch(config)# show ipv6 nd ra IPv6 Router Advertisement Configuration Global RA Suppress : No Global Hop Limit : 10 IPv6 Unicast Routing : Enabled Interface ID --------vlan-1 vlan-22 tunnel-3

Supp RA ---Yes No Yes

Interval Min/Max -------200/600 200/600 200/600

Lifetime (sec) -------1800 1800 1000

Mngd Flag ---No No No

Other Flag ----No No No

RCH Time NS Intrvl Hop (ms) (ms) Limit -------- ------- ----0 0 10 0 0 10 0 0 4

Example 123 Output Where Specific Prefixes Have Been Configured for RAs HP Switch(tunnel-3)# ipv6 nd ra prefix default 1296000 1209600 HP Switch(config)# show ipv6 nd ra prefix IPv6 Neighbor Discovery Prefix Information VLAN Name : VLAN22 IPv6 Prefix ---------------------------Default 2001:db8:0:a::/64

Valid Lifetime ------------------Infinite 1h:20m:44s

Onlink/Auto ----------On/On Off/On

Tunnel Name : Tunnel3 IPv6 Prefix Valid Lifetime Onlink/Auto ---------------------------- ------------------- ----------Default 15 days On/On

206 IPv6 Router Advertisements

Example 124 Detailed prefix configuration data for a specific VLAN HP Switch(config)# show ipv6 nd ra prefix vlan 30 IPv6 Neighbor Discovery Prefix Information VLAN Name : VLAN30 IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

Default Infinite Infinite On On On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:f:1b::/64 11/31/2010 00:00:01 11/01/2010 00:00:01 Off On On

IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : :

2001:db8:f:1d::/64 11/31/2010 00:00:01 11/01/2010 00:00:01 On On On

General RA operation An IPv6 routing switch configured as a member of a given VLAN transmits RAs for use by hosts on the VLAN or tunnel. It also transmits unscheduled RAs in response to router solicitations received from IPv6 hosts on the VLAN. The values a host receives in an RA are applied to settings that have not already been configured on the host by the system operator. (Values in an RA can also replace host settings that were learned from a previous RA.) When IPv6 unicast routing is enabled, RAs are transmitted by default on VLANs or tunnels enabled for IPv6 and configured with an IPv6 link-local address, unless RA transmission has been explicitly suppressed. RA configuration includes: Advertisement Value

Default

Page

managed flag (M-bit)

Not set

194

other-config-flag (O-bit)

Not set

194

prefix

The prefix of any global unicast IPv6 address configured on the VLAN interface1

197

length

N/A; based on existing configuration



valid lifetime

2,592,000 seconds (30 days)



preferred lifetime

604,800 seconds (7 days)



autoconfig (A-bit)

Set (host autoconfig enabled)



on-link (L-bit)

Set (use prefix on subject VLAN)



router advertisement transmission interval maximum

— 600 seconds

195

General RA operation 207

Advertisement Value

Default

Page

minimum

200 seconds

195

current hop limit

64

196

default lifetime

1800 seconds (3 x max. transmission interval)

196

reachable time

Unspecified (0)

196

retransmission timer

Unspecified (0)

197

1

Default operation excludes prefixes of stateless autoconfigured addresses.

RA basics •

Enabling IPv6 unicast routing on a routing switch initiates transmission of RAs on active, IPv6-enabled VLANs unless RA transmission has been suppressed.



RAs are not routed.



A host response to an RA depends on how the host implements IPv6. Generally, settings in an RA received by a host replaces settings received from an earlier RA. Settings configured directly on a host by an operator may override values received in an RA for the same settings.



When a host receives a default "unspecified" value in an RA, the host applies either its own current setting for that value, or the defaults specified in RFC 4861 or other applicable RFCs, depending on how IPv6 is implemented in the host.



The M-bit and O-bit flags enable RAs to be configured either to act as the sole source of host addressing and related settings, or to direct the host to use a DHCPv6 server for some or all such settings.

About setting up your IPv6 RA policy •

Is there a role for a DHCPv6 server in host configuration on a given VLAN, and what host services and policy will be configured? Affects M-bit and O-bit options—see page 194.



What is the ND policy that should be advertised? Includes hop-limit for host-generated traffic, the default router period, neighbor reachable time, and retransmit time for neighbor solicitations.



What prefixes should be advertised, and what prefixes should be suppressed? Prefixes configured on the routing switch VLAN interface will be included in RAs on that VLAN unless specifically denied.



What should be the maximum and minimum intervals (in seconds) for transmitting RAs?



Are there any VLANs or tunnels on the routing switch where RAs should be suppressed?



Will multiple routing devices be used to send RAs on a VLAN? •

The first RA received by a host determines the default router for that host. Other routers included in subsequent RAs received by the host become backup default routers for that host.



What, if any, differences are acceptable in RAs from different routing devices?

Steps for configuring IPv6 RAs When IPv6 unicast routing is enabled on the routing switch, RAs are transmitted on all IPv6-enabled VLANs or tunnels unless explicitly suppressed globally or per-VLAN.

208 IPv6 Router Advertisements

The following steps provide a general outline of the steps for configuring the routing switch for non-default RA operation on all IPv6-enabled VLANs or tunnels: 1. Enable IPv6 routing on your network. 2. Enable IPv6 unicast routing. (This must be enabled to allow configuration of other routing protocols). HP Switch(config)# ipv6 unicast-routing (This command enables RA transmission on any VLAN where RAs are not specifically suppressed.) 3.

Configure the desired per-VLAN or per-tunnel RA operation: a. Use the M-bit and O-bit settings to specify the source for IPv6 host configuration; see page 194: i. M-bit setting:

ii.



Get configuration from RAs (default).



Get configuration from DHCPv6.

O-bit setting (applies only if M-bit setting is left in default state): •

Use RA source for global unicast prefixes (default).



Do not use the RA for non-prefix configuration.

b.

Configure global unicast prefix assignments; see page 197: i. Specify any prefixes not configured on the routing switch VLAN or tunnel interface that should be transmitted in RAs to IPv6 hosts on the VLAN. ii. Deny any prefixes configured on the routing switch VLAN or tunnel interface that should not be transmitted in RAs to IPv6 hosts on the VLAN. Default: Global unicast prefixes configured on the routing switch VLAN interface are included in RAs.

c.

Configure the maximum and minimum interval for transmitting router advertisements on the VLAN; see page 195. NOTE: The routing switch also transmits RAs when it receives router solicitations from a host. Autoconfiguration must be enabled on the host before it will generate router solicitations on the VLAN or tunnel.

d.

Configure the ND policy for hosts on the VLAN or tunnel to use: i. Hop-limit; Default; 64, see page 196. ii. Default router lifetime; Default: 1800 seconds, see page 196. iii. Reachable time duration to advertise for confirmed neighbors; Default: unspecified (0); see page 197. iv. Retransmit time to advertise for neighbor solicitations; Default: unspecified (0); see page 197.

e.

Configure per-VLAN RA suppression for any VLAN or tunnel on which you do not want the routing switch to transmit RAs. See page 204. HP Switch(vlan-1)# ipv6 nd ra suppress HP Switch(tunnel-3)# ipv6 nd ra suppress

Guidelines for configuring RAs on multiple routing switches for the same VLAN Multiple routing switches transmitting RAs on the same VLAN or tunnel can provide redundancy. Typically, a host identifies the first router from which it receives an RA as the default router. The host uses any RAs received later from other routers to identify backup default routers.

Guidelines for configuring RAs on multiple routing switches for the same VLAN 209

While advertised prefixes can be different, the per-VLAN or per-tunnel RA policy should be the same for all routers transmitting RAs on a given VLAN. This includes the following parameters:

210



Managed-config-flag (M-bit)



Other-config-flag (O-bit)



Default router lifetime



Hop-limit



Reachable-time for neighbors



Retransmit time for neighbor solicitations

IPv6 Router Advertisements

9 DHCPv6-Relay Table 25 Summary of Commands Command syntax

Description

Default

CLI page reference

[no] dhcpv6-relay

Used in the global config context to enable DHCPv6-relay globally in the routing switch.

Disabled

211

[no] ipv6 helper-address unicast ipv6-unicast-helper-addr [no] ipv6 helper-address multicast [ all-dhcp-servers | ipv6-multicast-helper-addr ] [ egress vlan vid ]

Used in the VLAN context to enable DHCPv6-relay operation on the VLAN, and to specify either a unicast or multicast DHCPv6-relay helper address for forwarding DHCPv6 service requests from hosts on the subject VLAN.

-

211

show ipv6 helper-address [ vlan vid ]

Displays the DHCPv6-relay configuration on all VLANs configured on the routing switch or on the VLAN you specify.

-

212

For introductory information on DHCPv6-relay, see “About configuring DHCPv6 relay” (page 214).

Configuring DHCPv6-relay DHCPv6-relay is disabled by default. To enable and configure it, use the commands in this section:

Syntax: [no] dhcpv6-relay Used in the global config context to enable DHCPv6-relay globally on the routing switch. The no form disables DHCPv6-relay operation on the routing switch. Default: Disabled NOTE: To use DHCPv6-relay on a given VLAN, at least one IPv6 helper-address must be configured on the VLAN, and IPv6 routing (ipv6 unicast-routing) must be enabled.

Syntax: [no] ipv6 helper-address unicast ipv6-unicast-helper-addr [no] ipv6 helper-address multicast [ all-dhcp-servers | ipv6-multicast-helper-addr ] [ egress vlan vid ]

Configuring DHCPv6-relay

211

Used in the VLAN context to enable DHCPv6-relay operation on the VLAN, and to specify either a unicast or multicast DHCPv6-relay helper address for forwarding DHCPv6 service requests from hosts on the subject VLAN. ipv6-unicast-helper-addr Specifies the global unicast address of a remote DHCPv6 server configured to support hosts on the indicated VLAN. all-dhcp-servers Specifies that the routing switch forward host requests for DHCPv6 service to multicast address FF05::1:3 via the VLAN specified by egress vlan vid egress vlan vid Specifies the VLAN on which DHCPv6 service requests forwarded to a multicast destination will be relayed. The egress VLAN must be a different VLAN than the one on which the multicast helper address is configured. A service request relayed on the egress VLAN to a downstream router remains in that VLAN unless the downstream router is configured on that VLAN with a unicast helper address for a server on another VLAN. See the example in Figure 20 (page 217). Using the no form of the command removes the specified helper address. Removing all helper addresses from a given VLAN disables DHCPv6-relay on that VLAN. NOTE: DHCPv6-relay operation must be enabled with dhcpv6-relay at the global config level.

Viewing the DHCPv6-relay configuration Syntax: show ipv6 helper-address [ vlan vid ] Displays the DHCPv6-relay configuration on all VLANs configured on the routing switch or on the VLAN you specify.

212

DHCPv6-Relay

Example 125 Display of Unicast Helper Address Configured on VLAN 10 in Figure 19 (page 216) HP Switch(config)# show ipv6 helper-address VLAN: 10 IPv6 Helper Address Egress Vlan --------------------------------------- ----------1 2001:db8:0:12::11 1

Egress VLAN not used with unicast Helper addresses

Example 126 Display of Multicast Helper Address Configured on VLAN 10 in Example 126 (page 213) HP Switch(config)# show ipv6 helper-address VLAN: 14 IPv6 Helper Address Egress Vlan --------------------------------------- ----------1 FF05::1:3 1

Egress VLAN required for multicast Helper addresses

NOTE: Configuring a unicast IPv6 helper address does not require the specification of an egress VLAN. However, an egress VLAN must be included when configuring a multicast helper address. Using the show ipv6 help-address vlan vid command displays the helper address information for the specified VLAN in the same format as that shown above.

Syntax: show run Use this option to verify whether DHCPv6-relay is enabled on the routing switch. The output includes per-VLAN listings of any configured helper addresses. HP Switch(config)# show run Running configuration: . . . ipv6 hop-limit 25 1 ipv6 unicast-routing 2 interface loopback 1 ip address 1.1.1.1 exit snmp-server community "public" unrestricted vlan 10 untagged 20-22 ipv6 address fe80::1 link-local ipv6 address 2001:db8:0:10::1 ipv6 helper-address unicast 2001:db8:0:12::11 exit vlan 14 ipv6 address fe80::1 link-local ipv6 address 2001:db8:0:14::1 exit dhcpv6-relay

3

4

Configuring DHCPv6-relay

213

1 2 3 4

Non-Default Hop-Limit Configured IPv6 Unicast Routing Enabled DHCPv6 Helper Address Configured Per-VLAN DHCPv6-Relay Globally Enabled

Use the show dhcpv6-relay command to display statistical information about DHCPv6 relay. Example 127 DHCPv6 relay information HP Switch(config)# show dhcpv6-relay DHCPV6 Relay Agent : Enabled Client Requests Server Responses VLAN Name --------DEFAULT.. VLAN20

Received -------120 30

Dropped ------0 1

Failed -----0 0

Received -------120 29

Dropped ------0 2

Failed -----0 1

DHCPv6-relay operating notes •

A DHCPv6-relay message sent to any multicast address carries a hop limit of 32. The hop limit for DHCPv6 requests sent to a unicast address is determined by the ipv6 hop-limit 1 - 255 command at the global configuration level; Default: 64, see “Configuring global IPv6 routing parameters” (page 183).



Multicast addresses reserved for DHCPv6 include the following:





FF02::1:2—All_DHCP_Relay_Agents_and_Servers



FF05::1:3—All_DHCP_Servers

DHCPv6 client and relay functions are mutually exclusive on a VLAN. Attempting to configure one of these functions on a VLAN while the other is already configured results in one of the following messages: •

Cannot configure an IPv6 helper-address as DHCPv6 client is enabled on the VLAN



Cannot enable DHCPv6 client as an IPv6 helper-address is configured on the VLAN.



The routing switch supports concurrent, independent operation of DHCPv4 and DHCPv6.



Operating limits: DHCPv6-relay feature

Maximum

Unique helper addresses supported on the routing switch

321

Unique helper addresses per VLAN interface

321

1

If the same helper address is used on multiple VLANs, it is counted as one address toward these maximums.

About configuring DHCPv6 relay Beginning with software release K.15.01, the routing switches include operation as DHCPv6 relay agents between DHCPv6 servers and clients. Dynamic Host Configuration Protocol (DHCPv6) is used for configuring clients with IPv6 address and other configuration parameters without user intervention on the client. Where a DHCPv6 server 214

DHCPv6-Relay

and a client exist on the same local network, the client's requests for service are received directly by the DHCPv6 server, and a relay agent is not needed. However, if a client and the DHCPv6 server available to support it are in different networks, or subnets, a DHCPv6 relay agent is needed to forward client service requests to the server and to relay server responses back to the client. (The DHCPv6 relay agent is transparent to the client.) Three main elements comprise DHCPv6-relay operation: •

DHCPv6 clients per-VLAN or subnet



A routing switch configured with the following:





Either IPv6 static routing or the OSPFv3 routing protocol (or both)



DHCPv6 relay agent enabled to forward DHCPv6 client/server traffic between a host or another relay agent and a remote DHCPv6 server

One or more remote DHCPv6 servers reachable from the routing switch

DHCPv6 request forwarding With DHCPv6-relay enabled and a reachable unicast helper address configured on a given VLAN, a client request for DHCPv6 service will be routed to the designated server. If a multicast helper address is configured on the VLAN, the client request will be sent from the routing switch on the VLAN designated in the configuration for that helper address.

DHCPv6-relay helper addresses A unicast helper address enables routing of a client service request to the IPv6 address of a specific, remote DHCPv6 server. The multicast forwarding options route DHCPv6 requests on a VLAN interface to either •

The "All_DHCP_Servers" (FF05::1:3) multicast address



A user-selected multicast address

The routing switch supports up to 32 unique helper addresses and counts multiple instances of the same helper address on different VLANs as one address. Where multiple helper addresses are configured on the same VLAN, the routing switch forwards client service requests to all such addresses, and selects the server from which it receives the first response.

General steps for enabling DHCP relay operation For the DHCPv6 relay agent to function on the routing switch, you must complete the following steps: 1. Ensure that there is a route configured between a DHCPv6 server and the routing switch and that the server is configured to support host requests forwarded from the routing switch. 2. For each VLAN on which you want the routing switch to provide DHCPv6-relay services, determine the helper addresses the relay agent should have for forwarding client DHCPv6 requests to reachable DHCPv6 servers. You can configure one or more helper addresses and can use either or both of the following types: Unicast Specifies the global unicast address of a specific DHCPv6 server. Multicast Specifies a group of DHCPv6 servers (well-known or user-defined) in a defined network scope and group identification. (For more information on this topic, see section 2.7, "Multicast Addresses" in RFC 4291.) This option includes specifying the VLAN interface on which requests to a given multicast address will exit from the routing switch. 3.

In each VLAN context where DHCPv6-relay service is needed, use the ipv6 helper-address unicast / multicast command to configure one or more DHCPv6 helper addresses. About configuring DHCPv6 relay

215

4. 5. 6.

In the global config context, use the dhcpv6-relay command to globally enable DHCPv6 relay on the routing switch. If IPv6 routing is not already enabled on the routing switch, use the ipv6 unicast-routing command in the global config context to enable IPv6 routing. On each VLAN where you have configured a helper address, ensure that the target DHCPv6 server is reachable.

Multiple-hop forwarding of DHCPv6 service requests If a routing switch receives a unicast DHCPv6 service request forwarded by a relay agent, the request is routed to the specified server. For example, in Figure 19 (page 216), router "X" is a relay agent configured to forward DHCPv6 requests received from VLAN 10 to unicast helper address 2001:db8:0:12::11. Router "Y" receives the request from router "X" on VLAN 10 and routes it to the DHCPv6 server at 2001:db8:0:12::11 on VLAN 12. In this case, router "Y" acts as an IPv6 router and not as a DHCPv6 relay. Figure 19 Routing a unicast DHCPv6 request across a multiple-hop topology Router "Y" DHCPv6 Server

VLAN-12

VLAN-12: 2001:db8:0:12::10

2001:db8:0:12::11

VLAN-10: 2001:db8:0:10::2

DHCPv6 Server

Client

2001:db8:0:15::33 VLAN-10 VLAN-15

Router "X": Relay Agent VLAN-10: 2001:db8:0:10::1 Unicast helper address = 2001:db8:0:12::11 VLAN-14:2001:db8:0:14::1

Router "Z" VLAN-15: 2001:db8:0:15::1 VLAN-14

VLAN-14:2001:db8:0:14::2

In Figure 20 (page 217) , router "X" is a relay agent configured to forward DHCPv6 service requests received from VLAN 10 to the "all-DHCPv6-servers" multicast helper address (FF05::1:3) through VLAN 14. Router "Z" receives the request from router "X" on VLAN 14. Because router "Z" is configured with unicast helper address 2001:db8:0:15::33 on VLAN 14, the service request is relayed to the DHCPv6 server at 2001:db8:0:15::33 on VLAN 15. (In this example, both router "X" and router "Z" act as DHCPv6 relay agents.)

216

DHCPv6-Relay

Figure 20 Routing a multicast DHCPv6 request across a multiple-hop topology Router "Y" VLAN-B: 2001:db8:0:12::10

VLAN-12

VLAN-A: 2001:db8:0:10::2

Client

2001:db8:0:15::33

VLAN-10

VLAN-15

VLAN-10: 2001:db8:0:10::1

VLAN-14:2001:db8:0:14::1

2001:db8:0:2::11

DHCPv6 Server

Router "X": Relay Agent

Multicast helper address = all-dhcp-servers egress vlan 14

DHCPv6 Server

Router "Z"

VLAN-14

VLAN-E: 2001:db8:0:15::1 VLAN-D: 2001:db8:0:14::2

A multi-hop relay scenario such as is shown in Figure 20 (page 217) requires the following: •

All relays in the path except the relay closest to the server have a multicast helper address.



The last relay in the path to the server has a unicast helper address.

Thus, if router "Z" was not configured with a helper address as shown above, the relayed service request would be restricted to VLAN 14 and would not reach the server at 2001:db8:0:15::33 on VLAN 15.

Multiple-hop forwarding of DHCPv6 service requests

217

10 OSPFv3 Routing Table 26 Summary of Commands Command syntax

Description

Default

CLI page reference

ip router-id ip-addr interface loopback 0 - 7 ip address n.n.n.n

Configuring a router ID or an IPv4 loopback address.

disabled

220

[no] ipv6 unicast-routing

Enable IPv6 Unicast Routing.

disabled

220

router ospf3 [ enable | disable ]

Enable Global OSPFv3 Routing.

area [ ospf3-area-id | backbone ] [ normal ] no area [ ospf3-area-id | backbone ]

Assign the Routing Switch to OSPFv3 Areas.

none

222

area ospf3-area-id stub [ metric-cost 0 - 16777215 ] [ no-summary ] area ospf3-area-id nssa [ metric-cost 0 - 16777215] [ metric-type [ type1 | type2 ]]

Configure a stub or NSSA area.

no areas

222

none

224

221

[ no-summary ] no area ospf3-area-id

218

vlan vid ipv6 ospf3 [ area ospf3-area-id ] [no] vlan vid ipv6 ospf3 interface tunnel tunnel-id ipv6 ospf3 [ area | ospf3-area-id ] [no] interface tunnel tunnel-id ipv6 ospf3

Assign VLANs, tunnels, and/or Subnets to Each Area.

interface loopback 0 - 7 ipv6 ospf3 area [ ospf3-area-id | backbone ] [no] interface loopback 0 - 7 ipv6 ospf3

Assign user-defined IPv6 loopback addresses to an area (optionl).

[no] route-map name [ permit | deny 4294967295 ]

External Route Redistribution.

n/a

226

[no] router ospf3 redistribute [ connected | static ] route-map map-name

Enable ASBR operation on a routing switch.

n/a

226

router ospf3 default-metric 0 - 16777215 [no] router ospf3 default-metric

Modify the default metric for redistribution.

10

227

router ospf3 default-metric-type [ type1 | type2 ]

Modify the redistribution metric type.

type2

227

router ospf3 area [ ospf3-area-id | backbone ] range ipv6-addr/prefix [ type | [summary [cost 1 16777215]] | inter-area | nssa ] [ no-advertise ] [no] router ospf3 area [ ospf3-area-id | backbone ] range ipv6-addr/prefix [ type | [summary [cost 1 16777215]] | inter-area | nssa ]

Configure Ranges on an ABR To Reduce Advertising.

none

229

distance [ external | inter-area | intra-area ] [ 1 255 ]

Use Administrative Distance To Influence Route Choices.

110

231

OSPFv3 Routing

] [ seq 1 -

225

Table 26 Summary of Commands (continued) Command syntax

Description

Default

CLI page reference

[no] restart strict-lsa

Strict LSA Operation for Graceful Restart Helper Mode.

disabled

232

ipv6 ospf3 cost 1 - 65535

IP Routing Interface Settings.

1

232

ipv6 ospf3 dead-interval 1 - 65535

Dead interval per interface.

40 seconds 232

ipv6 ospf3 hello-interval 1 - 65535

Hello interval per interface.

10 seconds 233

ipv6 ospf3 priority 1 - 255

Priority per interface.

1

233

ipv6 ospf3 retransmit-interval 1 - 3600

Retransmit interval per interface.

5 seconds

233

ipv6 ospf3 transit-delay 1 - 3600

Transit delay per interface.

1 second

233

[no] area area-id virtual-link router-id

Virtual Links Between ABRs.

none

233

[no] area area-id virtual-link router-id dead-interval Adjusting a dead interval on a virtual 1 - 65535 link.

40 seconds 234

area area-id virtual link router-id hello-interval 1 Adjusting a hello interval on a virtual - 65535 link.

10 seconds 235

area area-id virtual link router-id retransmit-interval 1 - 3600

Adjusting the retransmit interval on a virtual link.

5 seconds

236

area area-id virtual-link router-id transit-delay 0 - 3600

Adjusting transit-delay on a virtual link.

1 second

236

[no] ipv6 ospf3 passive

OSPFv3 Passive.

OSPFv3 active

237

Multiple commands

Displaying OSPFv3 Information.

n/a

239

debug ipv6 ospf3 [ adj | event | flood | lsa-generation | packet | retransmission | spf ]

Debugging OSFP routing messages.

off

256

[no] ip load-sharing 2 - 4

Enable load-sharing among next-hop routes.

Enabled

256

OSPFv3 is the IPv6 implementation of open shortest path first protocol. (OSPFv2 is the IPv4 implementation of this protocol.) Beginning with software version K.15.01, the switches can be configured to run OSPFv3 either alone or simultaneously with OSPFv2. (OSPFv3 and OSPFv2 run as independent protocols on the routing switch and do not have any interaction when run simultaneously.) NOTE: License Requirements — In the 3500, 3500yl, 5400zl, 6600, and 8200zl switches, OSPFv3 is included with the optional Premium License. In the 6200yl switches, this feature is included with the base feature set.

219

This section describes OSPFv3 terms, basic features, and general operation. Both VLANS and tunnels can be assigned to areas and may be collectively referred to as an IP routing interface. For specific configuration information, turn to the topics referenced in the following command index. For information on configuring tunnels, see “IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels” (page 273). NOTE:

In this chapter, "OSPF" refers to OSPFv3 for IPv6 operation unless otherwise stated.

OSPFv3 RFC compliance The OSPFv3 features covered comply with the following: •

RFC OSPFv3 for IPv6



RFC 3101 option

Activating OSPFv3 After either an IPv4 loopback address or a router ID has been configured on the routing switch, OSPFv3 activates when enabled with the following two commands: HP Switch(config): ipv6 unicast-routing HP Switch(config): router ospf3 enable NOTE:

The router ospf3 enable command enables OSPFv3 without a system reset.

For information on dynamically configuring OSPFv3, see “OSPFv3 Activation and Dynamic Configuration” (page 268).

Configuring OSPFv3 on the routing switch Router ID or IPv4 loopback address requirement OSPFv3 routing requires either a router ID or an IPv4 loopback address configured on the routing switch. If this requirement is not already satisfied, use one of the following commands to configure a router ID or IPv4 loopback address:

Syntax: ip router-id ip-addr interface loopback 0 - 7 ip address n.n.n.n ip router-id ip-addr Executed at the global configuration level to assign a router ID to the routing switch. For more information, see “System router ID” (page 183) Default: Disabled interface loopback Executed at the global or deeper configuration level to assign an IPv4 address to a loopback interface on the routing switch. For more information, see "Loopback Interfaces" in the latest Basic Operation Guide.

Enabling IPv6 Routing Syntax: [no] ipv6 unicast-routing 220 OSPFv3 Routing

Executed at the global configuration level to enable IPv6 routing on the routing switch. Default: Disabled The no form disables IPv6 routing. (Global OSPFv3 routing must be disabled before you disable IPv6 routing.) Example 128 Enabling IPv6 Routing HP Switch(config)# ipv6 unicast-routing

Enabling global OSPFv3 routing Syntax: router ospf3 [ enable | disable ] The router ospf3 command executed alone puts the routing switch into ospf3 context. The keyword options enable or disable OSPFv3 on the routing switch. This command allows you to configure OSPFv3 before activating it on the routing switch. Global IPv6 unicast-routing must be enabled before executing this command. Default: Disabled NOTE: If you disable OSPFv3, the switch retains all the configuration information for the disabled protocol in flash memory. If you subsequently restart OSPF, the existing configuration will be applied. Example 129 enable global OSPFv3 routing HP Switch(config)# router ospf3 enable HP Switch(ospf3)#

Assigning the routing switch to OSPFv3 areas After you globally enable OSPFv3 on the routing switch, use this command to create one or more OSPFv3 areas within your autonomous system (AS). A routing switch can belong to one area or to multiple areas. Participation in a given area requires configuring one or more VLANs and assigning each to the desired area. This is covered in “Enabling global OSPFv3 routing” (page 221). •

If you want the VLANs configured on the routing switch to all reside in the same area, you need to configure only that one area. (In this case, the routing switch would operate as an internal router for the area.)



If you want to put different VLANs on the routing switch into different areas, you need to re-execute this command for each area. (In this case, the routing switch operates as an ABR for each of the configured areas.)

NOTE: Each ABR must be either directly connected to the backbone area (0) or be configured with a virtual link to the backbone area through another ABR that is directly connected to the backbone area. This is covered in “About adjusting performance by changing the VLAN interface settings (optional)” (page 271).

Configuring OSPFv3 on the routing switch 221

Configuring an OSPFv3 backbone or normal area Syntax: area [ ospf3-area-id | backbone ] [ normal ] no area [ ospf3-area-id | backbone ] After using router ospf3 to globally enable OSPFv3 and enter the global OSPF3 context, execute this command to assign the routing switch to a backbone or other normal area. The no form of the command removes the routing switch from the specified area. Default: No areas. Range: 1 - 16 areas (of all types) ospf-area-id Specifies a normal area to which you are assigning the routing switch. You can assign the routing switch to one or more areas, depending on the area in which you want each configured VLAN or subnet to reside. You can enter area IDs in either whole number or dotted decimal format. (The routing switch automatically converts whole numbers to the dotted decimal format.) For example, if you enter an area-ID of 1, it appears in the switch's configuration as 0.0.0.1, and an area-ID of 256 appears in the switch configuration as 0.0.1.0. Entering an area ID of 0 or 0.0.0.0 automatically joins the routing switch to the backbone area. The maximum area ID value is 255.255.255.254 (4,294,967,294). backbone Assigns the routing switch to the backbone area and automatically assigns an area ID of 0.0.0.0 and an area type of normal. Using 0 or 0.0.0.0 with the above ospf3-area-id option achieves the same result. The backbone area is automatically configured as a "normal" area type. normal Applied by default if not specified in an area command. Required to convert an existing NSSA or stub area to a normal area. Example 130 Configuring an OSPFv3 backbone or normal area To configure a backbone and a normal area with an ID of "1" (0.0.0.1) on a routing switch: HP Switch(ospf3)# area backbone HP Switch(ospf3)# area 1 To convert an existing NSSA or stub area to a normal area, you would include the normal keyword. For example, if area 10 was configured as an NSSA area you wanted to convert to a normal area, you would use the following command: HP Switch(ospf3)# area 10 normal

Configuring a stub or NSSA area Syntax: area ospf3-area-id stub [ metric-cost 0 - 16777215 ] [ no-summary ] area ospf3-area-id nssa [ metric-cost 0 - 16777215 ] [ metric-type [ type1 | type2 ]]

222 OSPFv3 Routing

[ no-summary ] no area ospf3-area-id After using router ospf3 to globally enable OSPFv3 and enter the global OSPF3 context, execute this command to assign the routing switch to a stub area or NSSA (does not apply to backbone and normal OSPFv3 area ABRs). The no form of the command removes the routing switch from the specified area. Default: No areas. Range: 1 - 16 areas (of all types) ospf3-area-id Same area ID as in “Configuring an OSPFv3 backbone or normal area” (page 222), except you cannot assign a backbone area number (0 or 0.0.0.0) to a stub or NSSA area. stubnssa Designates the area identified by ospf3-area-id as a stub area or NSSA. metric-cost 0 - 16777215 If the routing switch is used as an ABR for the designated area, assigns the cost of the default route (to the backbone) that is injected into the area. NOTE: If the routing switch is not an ABR for a stub area or NSSA, the above cost setting is still allowed, but is not used. In the default configuration, a routing switch acting as an ABR for a stub area or NSSA injects type-7-LSA default routes into the area. If no-summary is configured on the ABR, it injects inter-area-prefix-LSA routes into the area. For more information on no-summary, see “About replacing inter-area-prefix-LSAs and type-7-external-LSA default routes with an AS-external-LSA default route” (page 265). metric-type [ type1 | type2 ] Used in NSSA ABRs only. Specifies the type of external cost metric to include in type-7-LSAs advertised for redistribution of external routes in the NSSA. The metric-type command specifies whether to include the redistribution cost in the cost metric calculation for a type-7-LSA default route injected into the area. type1 : Calculate external route cost for a type-7-LSA default route as the sum of (1) the external route cost assigned by the ASBR plus (2) the internal cost from the router with traffic for the external route to the ASBR advertising the route. type2 : Use the external route cost assigned by the ASBR advertising the route. Default: Enabled with metric-type type2. NOTE: Different routers in the NSSA can be configured with different metric-type values. no-summary Where the routing switch is an ABR for a stub area or an NSSA, this option reduces the amount of LSA traffic entering the area from the backbone by replacing the injection of inter-area-prefix-LSA routes and type-7-LSA default external routes with injection of an inter-area-prefix-LSA default route. Default: Disabled For more information on this topic, see “Not-so-stubby-area (NSSA)” (page 264), “Stub area” (page 264), and “About replacing inter-area-prefix-LSAs and type-7-external-LSA default routes with an AS-external-LSA default route” (page 265). Configuring OSPFv3 on the routing switch 223

Using no area ospf3-area-id nssa no-summary resets the routing switch to the state where injection of inter-area-prefix-LSA routes and the type-7-LSA default external routes is enabled with metric-type set to type2. Example 131 Creating stub area and NSSA assignments The following examples of configuring a stub area and an NSSA on a routing switch use an (arbitrary) cost of "15." HP Switch(ospf3)# area 2 stub metric-cost 15

Assigns a stub area with a cost of 15. HP Switch(ospf3)# area 3 nssa metric-cost 15

Assigns an NSSA with a cost of 15 and, by default, uses a Network-LSA default cost metric for Type-7-LSA (external) routes received from the backbone. HP Switch(ospf3)# area 4 nssa metric-cost 15 no-summary

Assigns an NSSA with a cost of 15, blocks injection of Inter-Area-Prefix- LSA routes, and starts injection of Inter-Area-Prefix-LSA default routes from the backbone. HP Switch(ospf3)# area 5 nssa metric-cost 15 metric-type type1

Sets the cost metric type for Type-7- LSA default routes injected into the NSSA.

Enabling OSPFv3 on an interface and assigning one or more VLANs to each area After you define an OSPFv3 area, you can assign one or more VLANs to it. When a VLAN is assigned to an area, all currently configured IPv6 addresses in the VLAN are automatically included in the assignment. NOTE: All static VLANs configured on a routing switch configured for OSPFv3 must be assigned to one of the defined areas in the AS.

Syntax: vlan vid ipv6 ospf3 [ area ospf3-area-id ] [no] vlan vid ipv6 ospf3 interface tunnel tunnel-id ipv6 ospf3 [ area | ospf3-area-id ] [no] interface tunnel tunnel-id ipv6 ospf3 Executed in a specific VLAN context to assign the VLAN to the specified area. If area is not specified, the command defaults to the backbone area. Requires that the area is already configured on the routing switch. This command assigns all configured networks in the VLAN to the specified OSPFv3 area. vlan vid Defines the VLAN context for executing the area assignment. interface tunnel tunnel-id Defines the tunnel context for executing the area assignment. area ospf3-area-id Identifies the OSPFv3 area to which the VLAN should be assigned. NOTE: If you add a new IPv6 address to a VLAN after assigning the VLAN to an OSPFv3 area, the new network automatically joins the area. 224 OSPFv3 Routing

Before adding a new VLAN to an area, you must enable IPv6 on the VLAN. Otherwise the following CLI message appears: IPV6 should be enabled before configuring OSPFv3.

The no form deletes the OSPFv3 configuration from the specified VLAN. Example 132 Assign VLAN 8 on a routing switch To assign VLAN 8 on a routing switch to area 3 and include all IP addresses configured in the VLAN, enter the following commands: HP Switch(ospf3)# vlan 8 HP Switch(vlan-8)# ipv6 ospf3 area 3

Assigning IPv6 loopback addresses to an area (optional) After you define the OSPFv3 areas to which the switch belongs, you can assign user-defined IPv6 loopback addresses to an area. An IPv6 loopback interface is configured with an IPv6 address that is unique in an AS and is always reachable as long as at least one of the IPv6 interfaces on the routing switch is operational. NOTE: For information on configuring IPv6 loopback interfaces, see “Assigning an IPv6 address to a loopback interface” (page 22).

Syntax: interface loopback 0 - 7 ipv6 ospf3 area [ ospf3-area-id | backbone ] [no] interface loopback 0 - 7 ipv6 ospf3 Executed in a specific loopback context to assign an IPv6 loopback interface to the specified area. Requires that the loopback interface is already configured with an IP v6 address on the routing switch. loopback interface 0 - 7 Defines the loopback context for executing the area assignment ipv6 ospf3 area ospf3-area-id Identifies the OSPFv3 area to which the loopback interface is assigned. NOTE: The area must already exist, and the loopback interface must already be configured with a minimum of one IPv6 address. An IPv6 loopback interface can be assigned to only one area at any time. When an IPv6 loopback interface is assigned to a given area, the no form removes the interface from that area. Example 133 Assigning IPv6 loopback addresses to an area To assign loopback interface 3 on the routing switch to area 0.0.0.12, enter the following commands: HP Switch(config)# interface loopback 3 HP Switch(lo-3)# ipv6 ospf3 area 12

OSPFv3 redistribution of loopback addresses For information on this topic, see “About configuring ranges on an ABR to reduce advertising to the backbone” (page 270). Configuring OSPFv3 on the routing switch 225

Configuring route-maps Use the route-map command to enter the route-map context and configure one or more route-maps.

Syntax: [no] route-map name [ permit | deny ] [ seq 1 - 4294967295 ] Used in the OSPFv3 context (router-ospf3) of a routing switch operating as an ASBR. This command enters the route-map context and enables configuration of one or more route-maps for permitting or denying external connected or static routes. The no form of the command removes the named route-map from the switch configuration. For details on configuring route-maps, including several commands used in the route-map context, see "Route Maps" in the "IP Routing Features" chapter of the latest Multicast and Routing Guide for your routing switch. Example 134 Configuring route-maps To permit the content of a route-map named "mymap" with a sequence number of 100 on a routing switch operating as an ASBR, enter the following command in the global config context: HP Switch(config)# route-map mymap permit seq 100 HP Switch(route-map-mymap-10)_ After entering the route-map context, configure the route-map using the commands described in the chapter referenced in the above Syntax description.

Enabling route redistribution This step enables ASBR operation on a routing switch and must be executed on each routing switch connected to external routes you want to redistribute in your OSPFv3 domain. NOTE: This step assumes you have configured and implemented the route redistribution policies in “Enabling IPv6 routing” (page 178) that are needed to restrict unwanted routes from being redistributed in the OSPFv3 area and domain to which the ASBR belongs.

Syntax: [no] router ospf3 redistribute [ connected | static ] route-map map-name Executed on an ASBR to permit or deny redistribution of static and/or connected routes to the ASBR’s domain, as specified in the named route-map. static Redistribute from manually configured routes. connected Redistribute from locally connected networks. The no form removes the redistribution configuration for the specified route-map.

226 OSPFv3 Routing

Example 135 Enabling route redistribution To implement redistribution for the connected and static routes configured in the route-map named "mymap," you would execute the following commands on the applicable ASBR: HP Switch(config)# router ospf3 redistribute connected route-map mymap HP Switch(config)# router ospf3 redistribute static route-map mymap

Modifying the default metric for redistribution The default metric is a global parameter that specifies the cost applied to all external OSPFv3 routes by default.

Syntax: router ospf3 default-metric 0 - 16777215 [no] router ospf3 default-metric Globally assigns the cost metric to apply to all external routes redistributed by the ASBR. By using different cost metrics for different ASBRs, you can prioritize the ASBRs in your AS. Default: 10 Example 136 Modifying the default metric for redistribution To assign a default metric of 4 to all routes imported into an OSPFv3 domain through an ASBR, enter the following command in the ASBR: HP Switch(config)# router ospf3 default-metric 4

Modifying the redistribution metric type The redistribution metric type is used by default for all routes imported into OSPFv3. Type 1 metrics are the same "units" as internal OSPFv3 metrics and can be compared directly. Type 2 metrics are not directly comparable and are treated as larger than the largest internal OSPFv3 metric.

Syntax: router ospf3 default-metric-type [ type1 | type2 ] Globally reconfigures the redistribution metric type on an ASBR. type1 Specifies the OSPFv3 metric plus the external metric for an external route. type2 Specifies the external metric for an external route. Default: type2

Modifying the default metric for redistribution 227

Example 137 Modifying the redistribution metric type To change from the default setting on an ASBR to type 1, enter the following command: HP Switch(config)# router ospf3 default-metric-type type1

Enabling redistribution of loopback IPv6 addresses in OSPFv3 when the addresses are not assigned to an OSPFv3 area Enter the redistribute connected command as described in “Enabling route redistribution” (page 226). Example 138 Enabling redistribution of loopback IPv6 addresses In the following configuration, loopback interface 6 is configured with IPv6 address 2001:db8:1::127 and is assigned to OSPFv3 area 0.0.0.1, and thus is advertised as an OSPFv3 IntraArea route, regardless of whether route redistribution is enabled. In the same configuration, loopback interface 2 is configured with IPv6 address 2001:db8:2:133 but is not assigned to an OSPFv3 area. As a result, it will be advertised to neighbors as an External route, and only if it is a "permitted" route in a route-map invoked by the redistribute command. Example 139 Assigning loopback IPv6 addresses to OSPFv3 areas HP HP HP HP HP HP

Switch(config)# interface loopback 6 Switch(lo-6)# ipv6 address 2001:db8:1:127 Switch(lo-6)# ipv6 ospf3 area 1 Switch(lo-6)# interface loopback 2 Switch(lo-2)# ipv6 address 2001:db8:2:133 Switch(lo-2)# exit 1

1

Assigns an IPv6 address to loopback interface 2, but does not assign the interface to an OSPFv3 area

Verifying the OSPFv3 redistribution of loopback interfaces Enter the show ipv6 route ospf3 command from a neighboring router to display the IPv6 route table entries for detected OSPFv3 routes.

228 OSPFv3 Routing

Example 140 Verifying OSPFv3 redistribution of loopback interfaces on a neighboring router HP Switch(config)# show ipv6 route ospf3 IPv6 Route Entries Destination : 2001:db8::333/128 Gateway : fe80::55:1%vlan55 Type: ospf3 Sub-Type: External2 1

Distance: 110

Metric: 1

Destination : 2001:db8:1::127/128 Gateway : fe80::55:1%vlan55 Type: ospf3 Sub-Type: IntraArea 2

Distance: 110

Metric: 1

1

2

Indicates a loopback interface configured on a neighbor router with redistribution enabled, but not assigned to an OSPFv3 area Indicates a loopback interface configured on a neighbor router and assigned to an OSPFv3 area

For information on OSPFv3 redistribution of loopback addresses, see “OSPFv3 redistribution of loopback addresses” (page 225).

Configuring ranges on an ABR to reduce advertising to the backbone For more information, see “About configuring ranges on an ABR to reduce advertising to the backbone” (page 270).

Syntax: router ospf3 area [ ospf3-area-id | backbone ] range ipv6-addr/prefix [ type | [ summary [ cost 1 - 16777215 ]] | inter-area | nssa ] [ no-advertise ]

Syntax: [no] router ospf3 area [ ospf3-area-id | backbone ] range ipv6-addr/prefix [ type | [ summary [ cost 1 - 16777215 ]] | inter-area | nssa ] Use this command on a routing switch intended to operate as an ABR for the specified area to do either of the following: •

Simultaneously create the area and corresponding range setting for routes to summarize or block.



For an existing area, specify a range setting for routes to summarize or block (prevent).

ospf3-area-id Same area ID as in “Configuring an OSPFv3 backbone or normal area” (page 222), except you cannot use a backbone area number (0 or 0.0.0.0) for a stub area or NSSA. range ipv6-addr/prefix Defines the range of RAs to either summarize for injection into the backbone area or to prevent from being injected into the backbone area. The ipv6-addr value specifies the IPv6 address portion of the range, and prefix specifies the leftmost significant bits in the address.

Configuring ranges on an ABR to reduce advertising to the backbone 229

The ABR for the specified area compares the IPv6 address of each outbound RA with the address and significant bits in the mask to determine which routes to select for either summarizing or blocking. For example, 2001:db8:0:f::/64 defines a range including any address that has 2001:db8:0:f in the leftmost 64 bits. [ type | [ summary [ cost 1 - 16777215 ]] | inter-area | nssa ] [ no-advertise ] Configures the type of route summaries to advertise or block. [summary [ cost 1 - 16777215 ]] Specifies internal routes in the configured range of route advertisements. If no-advertise (above) is used in the command, then the ABR prevents the selected internal routes from being summarized in a type-3 LSA and advertised to the backbone. If no-advertise is not used in the command, then the selected routes are summarized to the backbone in a type-3 LSA. cost 1 - 16777215 User configured cost for an area summary range. If cost is specified, then the range will advertise the specified cost instead of the calculated cost. inter-area Specifies internal routes in the configured range of RAs. If no-advertise (below) is used in the command, the ABR prevents the selected internal routes from being summarized in an inter-area-prefix-LSA and advertised to the backbone. If no-advertise is not used in the command, the selected routes are summarized to the backbone in an inter-area-prefix-LSA. (The inter-area keyword in OSPFv3 is the equivalent of the summary keyword in OSPFv2.) nssa Specifies external routes (type-7-LSAs) in the configured range of RAs. If no-advertise (below) is used in the command, the ABR prevents the selected external routes from being summarized in an AS-external-LSA and advertised to the backbone. (Configure this option where an ABR for an NSSA advertises external routes that you do not want propagated to the backbone.) If no-advertise is not used in the command, the selected routes learned from type-7-LSAs in the area are summarized to the backbone in an AS-external-LSA. The no form of the command removes the specified range from the configuration. no-advertise Use this keyword only if you want to configure the ABR to prevent advertisement to the backbone of a specified range of routes. (This has the effect of "hiding" the specified range from the backbone area.) If you do not use this option, the ABR advertises the specified range of routes according to the type [ inter-area | nssa ] selection described above.

230 OSPFv3 Routing

Example 141 Assigning a Cost. The cost parameter provides a way to define a fixed, user-assigned cost of an LSA type 3 summarized prefix. To set the summary cost to 100 for area 10 with and address range of 10.10.0.0/ 16, enter the command as shown: HP Switch(ospf3)# area 10 range 10.10.0.0/16 type summary cost 100

To use the standard method for determining the summarized cost, enter the command as shown: HP Switch(ospf3)# area 10 range 10.10.0.0/16 type summary

You must execute write mem in order to preserve these settings across reboots. The show ip ospf3 command displays information about summary costs. An entry of “auto” indicates that the cost is calculated by the OSPF standard for summarized networks. The no form removes the specified range from the configuration. Example 142 ABR allowing or blocking advertisement of a range of internal routes available in an area Example of defining a range of internal routes to advertise to the backbone The following command defines a range of internal routes in area 30 to summarize for injection into the backbone area. (In this example, area 30 can be a normal or stub area, or an NSSA.) HP Switch(ospf3)# area 30 range 2001:db8:1a/48 type inter-area

For the same range of routes, you can use either of the following commands to block injection of a range of inter-area routes (inter-area-prefix-LSAs) from area 30 into the backbone. HP Switch(ospf3)# area 30 range 2001:db8:1a/48 type inter-area no-advertise

Examples of allowing or blocking a range of external routes available through an ASBR in an NSSA follow. This example applies only to external routes that can be advertised from an NSSA to the backbone. HP Switch(ospf3)# area 7 range 2001:db8:5f:1::/64 type nssa

Defines the range of external routes in the Area 7 NSSA to advertise to the backbone. HP Switch(ospf3)# area 7 range 2001:db8:7a:15::/64 type nssa no-advertise

Defines the range of external routes in the Area 7 NSSA to block from advertising to the backbone.

Influencing route choices by changing the administrative distance default For more information, see “About influencing route choices by changing the administrative distance default (optional)” (page 270).

Syntax: distance [ external | inter-area | intra-area ] [ 1 - 255 ] Used in the OSPFv3 configuration context (router ospf3) to globally reconfigure the administrative distance priority for the specified route type. Influencing route choices by changing the administrative distance default

231

1 is the highest priority; 255 is the lowest priority. external 1 - 255 Changes the administrative distance for routes between the OSPFv3 domain and other EGP domains. inter-area 1 - 255 Changes the administrative distance for routes between areas within the same OSPFv3 domain. intra-area 1 - 255 Changes the administrative distance for routes within OSPFv3 areas. Default: 110; Range: 1 - 255

Enforcing strict LSA operation for graceful restart helper mode For more information, see “About enforcing strict LSA operation for graceful restart helper mode (optional)” (page 270).

Syntax: [no] restart strict-lsa Used in the OSPFv3 context to enable or disable strict LSA operation in a network segment for a neighboring router that is attempting a graceful restart. When enabled, this operation halts helper mode support if a change in LSAs (topology change) is detected during the restart period of the neighbor. Default: Strict LSA operation enabled The no form disables strict LSA operation.

Adjusting performance by changing the VLAN interface settings Cost per interface Syntax: ipv6 ospf3 cost 1 - 65535 Used in the VLAN context to indicate the overhead required to send a packet across an interface. You can modify the cost to differentiate between 100 Mbps and 1000 Mbps (1 Gbps) links. This command assigns the specified cost to all networks configured on the VLAN. Default: 1

Dead interval per interface Syntax: ipv6 ospf3 dead-interval 1 - 65535 Used in the VLAN context to indicate the number of seconds that a neighbor router waits for a hello packet from the specified interface before declaring the interface "down". This command assigns the specified dead interval to all networks configured on the VLAN. Default: 40 seconds

232 OSPFv3 Routing

Hello interval per interface Syntax: ipv6 ospf3 hello-interval 1 - 65535 Used in the VLAN context to indicate the length of time between the transmission of hello packets from the routing switch to adjacent neighbors on that VLAN. This command assigns the specified Hello interval to all networks configured on the VLAN. Default: 10 seconds

Priority per interface Syntax: ipv6 ospf3 priority 1 - 255 Used in the VLAN context to enable changing the priority of an OSPFv3 router. The priority is used when selecting the designated router (DR) and backup designated routers (BDRs). The value can be from 0 to 255 (with 255 as the highest priority). If you set the priority to 0, the routing switch does not participate in DR and BDR election. This command assigns the specified priority to all networks configured on the VLAN. Default: 1

Retransmit interval per interface Syntax: ipv6 ospf3 retransmit-interval 1 - 3600 Used in the VLAN context to enable changing the retransmission interval for link-state advertisements (LSAs) on an interface. Default: 5 seconds

Transit delay per interface Syntax: ipv6 ospf3 transit-delay 1 - 3600 Used in the VLAN context to enable changing the time it takes to transmit link state update packets on this interface. This command reconfigures the estimated number of seconds it takes to transmit a link state update packet to all networks configured on the VLAN. Default: 1 second

Configuring a virtual link For more information, see “About configuring an ABR to use a virtual link to the backbone” (page 271).

Syntax: [no] area area-id virtual-link router-id In the ospf3 context, used on a pair of ABRs at opposite ends of a virtual link in the same area to configure the virtual link connection.

Configuring a virtual link 233

area-id This must be the same for both ABRs in the link and is the area number of the virtual link transit area in either decimal or 32-bit dotted decimal format. If area-id is not already configured on the routing switch, this command creates it. router-id On an ABR directly connected to the backbone area, this value must be the router ID of an ABR (in the same area) needing a virtual link to the backbone area as a substitute for a direct physical connection. On the ABR that needs the virtual link to the backbone area, this value must be the router ID of the ABR (in the same area) having a direct physical connection to the backbone area. The no form removes the virtual link. Example 143 Configuring a virtual link Figure 21 (page 234) shows an OSPFv3 ABR, routing switch "A" that lacks a direct connection to the backbone area (area 0). To provide backbone access to routing switch "A," you can add a virtual link between routing switch "A" and routing switch "C," using area 1 as a transit area. To configure the virtual link, define it on the routers that are at each end of the link. No configuration for the virtual link is required on the other routers on the path through the transit area (such as routing switch "B" in this example). Figure 21 Defining OSPFv3 virtual links within a network

To configure the virtual link on routing switch "A," enter the following command specifying the area 1 interface on routing switch "C": HP Switch(ospf3)# area 1 virtual-link 209.157.22.1 To configure the virtual link on routing switch "C," enter the following command specifying the area 1 interface on routing switch "A." HP Switch(ospf3)# area 1 virtual-link 10.0.0.1 For descriptions of virtual link interface parameters you can either use in their default settings or reconfigure as needed, see “About adjusting virtual link performance by changing the interface settings” (page 272).

Adjusting a dead interval on a virtual link For more information, see “About adjusting virtual link performance by changing the interface settings” (page 272). 234 OSPFv3 Routing

Syntax: [no] area area-id virtual-link router-id dead-interval 1 - 65535 In the ospf3 context, this command is used on both ABRs in a virtual link to change the number of seconds that a neighbor router waits for a hello packet from the specified interface before declaring the interface "down." This should be some multiple of the Hello interval. The dead-interval setting must be the same on both ABRs on a given virtual link. area-id Specifies the OSPFv3 area in which both ABRs in a given virtual link operate. In this use, the area ID is sometimes termed "transit area ID." This value must be the same for both ABRs in the virtual link. If the area does not exist, this command creates it. router-id For an ABR in a given virtual link, this is the router ID (in decimal or 32-bit dotted decimal format) used to create the link on that ABR. This value is the router ID of the opposite router in the virtual link. See the description of router-id in the syntax description in the section “Configuring a virtual link” (page 233). The no version restores the default value. Use show ipv6 ospf3 virtual-link router-id to view the current setting. See the example at “Viewing OSPFv3 virtual link information” (page 254). Default: 40 seconds

Adjusting a hello interval on a virtual link For more information, see “About adjusting virtual link performance by changing the interface settings” (page 272).

Syntax: area area-id virtual link router-id hello-interval 1 - 65535 In the ospf3 context, used on both ABRs in a virtual link to indicate the length of time between the transmission of hello packets between the ABRs on opposite ends of the virtual link. The hello-interval setting must be the same on both ABRs on a given virtual link. area-id Specifies the OSPFv3 area in which both ABRs in a given virtual link operate. In this use, the area ID is sometimes termed "transit area ID". This value must be the same for both ABRs in the virtual link. If the area does not exist, this command creates it. router-id For an ABR in a given virtual link, this is the router ID (in decimal or 32-bit dotted decimal format) used to create the link on that ABR. This value is the router ID of the opposite router in the virtual link. See the description of router-id in the section “Configuring a virtual link” (page 233). The no version restores the default value. Use show ipv6 ospf3 virtual-link router-id to view the current setting. See the example at “Viewing OSPFv3 virtual link information” (page 254). Default: 10 seconds Configuring a virtual link 235

Adjusting the retransmit interval on a virtual link For more information, see “About adjusting virtual link performance by changing the interface settings” (page 272).

Syntax: area area-id virtual link router-id retransmit-interval 1 - 3600 In the ospf3 context, used on both ABRs in a virtual link to change the number of seconds between LSA retransmissions on the virtual link. The retransmit-interval setting must be the same on both ABRs on a given virtual link. This value is also used when retransmitting database description and link-state request packets. area-id Specifies the OSPFv3 area in which both ABRs in a given virtual link operate. In this use, the area ID is sometimes termed "transit area ID." This value must be the same for both ABRs in the virtual link. If the area does not exist, this command creates it. router-id For an ABR in a given virtual link, this is the router ID (in decimal or 32-bit dotted decimal format) used to create the link on that ABR. This value is the router ID of the opposite router in the virtual link. See the description of router-id in the section “Configuring a virtual link” (page 233). The no version of the command restores the default value. Use show ipv6 ospf3 virtual-link router-id to view the current setting. See the example at “Viewing OSPFv3 virtual link information” (page 254). Default: 5 seconds

Adjusting transit-delay on a virtual link For more information, see “About adjusting virtual link performance by changing the interface settings” (page 272).

Syntax: area area-id virtual-link router-id transit-delay 0 - 3600 In the ospf3 context, used on both ABRs in a virtual link to change the estimated number of seconds it takes to transmit a link state update packet over a virtual link. The transit-delay setting must be the same on both ABRs on a given virtual link. area-id Specifies the OSPFv3 area in which both ABRs in a given virtual link operate. In this use, the area ID is sometimes termed "transit area ID." This value must be the same for both ABRs in the virtual link. If the area does not exist, this command creates it. router-id For an ABR in a given virtual link, this is the router ID (in decimal or 32-bit dotted decimal format) used to create the link on that ABR. This value is the router ID of the opposite router in the virtual link. See the description of router-id in the section “Configuring a virtual link” (page 233). The no version of the command restores the default value.

236 OSPFv3 Routing

Use show ipv6 ospf3 virtual-link ip-address to view the current setting. See the example at “Viewing OSPFv3 virtual link information” (page 254). Default: 1 second Example 144 Adjusting transit-delay on a virtual link To change the hello-interval on the virtual link configured for the network in Figure 21 (page 234) to 60 seconds: •

On routing switch “A” (router ID 10.0.0.1) you would use the following command to reconfigure the current hello-interval to 60 seconds: HP Switch(ospf3)# area 1 virtual-link 209.157.22.1 hellointerval 60



On routing switch “C” (router ID 209.157.22.1) you would use the following command to reconfigure the current hello-interval to 60 seconds HP Switch(ospf3)# area 1 virtual-link 10.0.0.1 hello-interval 60

Configuring OSPFv3 passive For more information, see “OSPFv3 passive” (page 272).

Syntax: [no] ipv6 ospf3 passive VLAN context command for enabling or disabling passive OSPFv3 operation on the VLAN. The no option returns the VLAN interface to active OSPFv3 operation. Default: OSPFv3 active Example 145 Configuring OSPFv3 passive To configure an OSPFv3 interface as passive, enter this command in the VLAN context: HP Switch(vlan-1)# ipv6 ospf3 passive To display the OSPFv3 passive information, enter the command shown in Example 146 (page 237): Example 146 show ipv6 ospf3 interface command with passive configured on an interface HP Switch(config)# show ipv6 ospf3 interface OSPFv3 configuration and statistics for interfaces Interface ---------vlan-55 vlan-75

Status -------Enabled Enabled

Area ID --------0.0.0.1 0.0.0.3

State ------BDR BDR

Cost ----1 1

Pri ----1 1

Passive ------No Yes

You can display the OSPFv3 passive information for a particular VLAN: suppose that a routing switch has OSPFv3 configured on VLAN 75. Example 147 (page 238) shows an example of detailed output for VLAN 75 alone.

Configuring a virtual link 237

Example 147 show ipv6 ospf3 interface command for a specific VLAN with passive configured on an interface HP Switch(config)# show ipv6 ospf3 interface 75 detail OSPFv3 configuration and statistics for VLAN 75 Interface Area ID Priority Type Hello Interval Transit Delay Events Neighbors

: : : : : : : :

vlan-75 0.0.0.3 1 BCAST 10 1 0 1

Status State Cost Passive Dead Interval Retransmit Interval Designated Router Backup Designated Router

: : : : : : : :

Enabled WAIT 1 Yes 40 5 15.1.1.2 15.1.4.4

Troubleshooting: Logging neighbor adjacency change events In the default configuration, the routing switch generates event log messages to indicate neighbor adjacency changes during initialization and normal operation. This enables OSPFv3 misconfiguration troubleshooting, while producing a lower volume of event log messages than is seen with the debug troubleshooting option. Both a standard (default) mode and an optional detail mode are provided. Using the optional debug destination command, the logging output can be directed to a syslog server or a terminal. For more information on debug, see the "Troubleshooting" appendix in the latest Management and Configuration Guide for your routing switch.

Syntax: logging neighbor-adjacency [ detail ] Used in the ospf3 context to enable logging of standard or detailed adjacency changes. In the default configuration, logs OSPFv3 neighbor changes into or out of the full adjacency state. detail Generates event log messages for all OSPFv3 neighbor adjacency state changes. no logging neighbor-adjacency Disables logging neighbor adjacency on the routing switch. no logging neighbor-adjacency detail Cancels detailed neighbor adjacency change logging and returns the routing switch to logging only neighbor changes into our out of full adjacency. Default: Standard full adjacency change logging enabled. The neighbor-adjacency event log messages are described in the latest Event Log Reference Guide for your routing switch. In the event log output, neighbors are identified by router ID.

238 OSPFv3 Routing

Example 148 Neighbor-adjacency change logging HP Switch(ospf3)# show log -r OSPF3: Keys: W=Warning I=Information M=Major D=Debug E=Error ---- Reverse event Log listing: Events Since Boot ---e 05/01/10 15:21:09 02809 OSPF3: ADJCHG: Neighbor 15.255.155.1 on interface vlan-22 moved to Down state, Inactivity Timer e 04/27/10 14:36:48 02809 OSPF3: ADJCHG: Neighbor 10.10.10.45 on interface vlan-11 moved to Full state, Loading Done HP Switch(ospf3)# show log -r OSPF3: Keys: W=Warning I=Information M=Major D=Debug E=Error ---- Reverse event Log listing: Events Since Boot ---e 05/01/10 15:21:09 02809 OSPF3: ADJCHG: Neighbor 15.255.155.1 on interface vlan-22 moved to Down state, Inactivity Timer e 04/27/10 14:36:48 02809 OSPF3: ADJCHG: Neighbor 10.10.10.45 on interface vlan-11 moved to Full state, Loading Done

Displaying OSPv3F Information You can use CLI commands to display the following OSPFv3 information: Command syntax

Description

CLI page reference

show ipv6 ospf3 [ general ]

General Information

240

show ipv6 route ospf3 [ dest-ipv6-addr ]

OSPFv3 Route information

242

show ipv6 ospf3 area [ area-id | backbone ] [ detail ] Area Information

243

show ipv6 ospf3 interface [ vlan vid | tunnel tunnel-id | loopback lo-id ] [ detail ]

Interface Information

243

show ipv6 ospf3 neighbor [ router-id ] [ detail ]

Neighbor router Information

246

show ipv6 ospf3 statistics [ vlan vid | tunnel tunnel-id ] clear ipv6 ospf3 statistics [ vlan vid | tunnel tunnel-id ]

Routing Interface Packet Statistics

247

show ipv6 ospf3 link-state as-scope [ lsid lsid-# ] [ router-id rtr-id-# ] [ detail | advertise ]

Link State Autonomous System Information

248

show ipv6 ospf3 link-state area-scope Link-State Area-Scope 250 area area-id lsid lsid-# router-id rtr-# type lsa-type Information [ detail | advertise ] show ipv6 ospf3 link-state link-scope interface vlan-id lsid lsid-# router-id rtr-id-# [ detail | advertise ]

Link-State Link-Scope Information

252

show ipv6 ospf3 redistribute

OSPFv3 redistribution 253 information

Displaying OSPv3F Information 239

Command syntax

Description

CLI page reference

show ipv6 ospf3 virtual-link [ rtr-id ] [ area area-id ]

Virtual Link information

254

show ipv6 ospf3 virtual-neighbor [ rtr-id ] [ area area-id ]

Virtual Neighbor information

254

show ipv6 ospf3 spf-log

OSPFv3 SPF Statistics 255

Viewing a summary of OSPFv3 configuration information Syntax: show ipv6 ospf3 [ general ] Displays the summary of OSPFv3 information, such as the areas configured, address ranges defined, interface information, timers, and virtual links. general Displays the OSPFv3 general status information and other generic information.

240 OSPFv3 Routing

Example 149 Output for show ipv6 ospf3 HP Switch# show ipv6 ospf3 OSPFv3 Configuration Information OSPFv3 Protocol : Enabled Router ID : 10.0.8.35 Currently defined areas:

Area ID --------backbone 10.3.16.0 10.3.32.0

Type -----Normal Normal Normal

Stub Default Cost -----------1 1 1

Stub Summary LSA ----------don’t send don’t send don’t send

Stub Metric Type SPF Runs ------------- -------Ospfv3 Metric Ospfv3 Metric Ospfv3 Metric

Currently defined address ranges: Prefix AreaID LSAType Advt IPv6Addr --------------- ---------- ------ -------------OSPFv3 interface configuration:

Interface ---------vlan-55 vlan-75 tunnel-3

Area ID --------------0.0.0.1 0.0.0.1 0.0.0.0

Admin Status -------Enabled Enabled Enabled

Type ----BCAST BCAST P2P

Cost ----1 1 1

Pri --1 1 1

OSPFv3 configured interface timers:

Interface ---------vlan-55 VLAN-75 tunnel-3

Transit Delay ------1 1 1

Retransmit Interval ---------5 5 5

Hello Interval -------10 10 10

Dead Interval ---------40 40 40

OSPFv3 configured virtual interfaces: Xmit Rxmt Hello Transit AreaID Neighbor Router Delay Intvl Intvl --------------- --------------- ----- ----- ----0.0.0.1 15.255.155.1 1 5 10

Dead Interval ---------40

Viewing general OSPFv3 configuration information

Viewing a summary of OSPFv3 configuration information

241

Example 150 show ipv6 ospf3 general output HP Switch# show ipv6 ospf3 general OSPFv3 General Status OSPFv3 protocol Router ID

: enabled : 10.0.8.36

Intra-area distance Inter-area distance AS-external distance

: 110 : 110 : 110

Default import metric : 111 Default import metric type : external type 2 Area Border : yes AS Border : yes External LSA Count : 9 Originate New LSA Count Receive New LSA Count

: 24814 : 14889

Graceful Restart Strict-Lsa Checking Neighbor Adjacency Logging

: Disabled : Enabled

Viewing OSPFv3 route information Syntax: show ipv6 route ospf3 [ dest-ipv6-addr ] Displays OSPFv3 route entries in the routing table. dest-ipv6-addr Displays the OSPFv3 routing table entry for a specific destination.

242 OSPFv3 Routing

Example 151 Output for all OSPFv3 routes in the routing table HP Switch# show ipv6 route ospf3 IPv6 Route Entries Destination : 2001:db8::333/128 Gateway : fe80::55:1%vlan55 Type : ospf3 Sub-Type : External2

Distance : 110

Metric : 1

Destination : 2001:db8:1::12/128 Gateway : fe80::55:1%vlan55 Type : ospf3 Sub-Type : IntraArea

Distance : 110

Metric : 1

Example 152 Output for a specific OSPFv3 route in the routing table HP Switch# show ipv route ospf3 2001:db8:1::127 IPv6 Route Entries to 2001:db8:1::127 Destination : 2001:db8:1::12/128 Gateway : fe80::55:1%vlan55 Type : ospf3 Sub-Type : IntraArea

Distance : 110

Metric : 1

Viewing OSPFv3 area information Syntax: show ipv6 ospf3 area [ area-id | backbone ] [ detail ] Displays summary information on all configured areas. [ area-id | backbone ] Displays summary information for the specified area. detail Displays area summary information in a modified format. ospf-area-id Shows information for the specified area. If no area is specified, information for all the OSPFv3 areas configured is displayed. Example 153 show ipv6 ospf3 area output HP Switch# show ipv6 ospf3 area OSPFv3 Area Information Area ID -------0.0.0.0 0.0.0.1 0.0.0.3

Type ------Normal Normal Normal

Cost ----10 10 10

SPF Runs -------780 780 780

ABR ---1 2 1

ASBR ----0 1 0

LSA ---6 9 11

Checksum ---------0x0005050f 0x0004d3f1 0x0007f854

Viewing OSPFv3 interface information Syntax: show ipv6 ospf3 interface [ vlan vid | tunnel tunnel-id | loopback lo-id ] [ detail ] Viewing a summary of OSPFv3 configuration information 243

Displays basic OSPFv3 information related to the VLANs configured on the routing switch. vlan-id Displays information for a specific VLAN. tunnel tunnel-id Displays information for a specific tunnel. loopback lo-id Displays information for a loopback interface. detail Displays additional, VLAN-specific OSPFv3 information.

244 OSPFv3 Routing

Example 154 show ipv6 ospf3 interface output HP Switch# show ipv6 ospf3 interface OSPFv3 configuration and statistics for interfaces Interface -----------vlan-55 vlan-75 tunnel-3

Status -------Enabled Enabled Enabled

Area ID -----------0.0.0.1 0.0.0.3 0.0.0.0

State ------BDR BDR Down

Cost -----1 1 1

Pri ----1 1 1

Passive ------No No No

Example 155 show ipv6 ospf3 interface tunnel Output HP Switch(config)# show ipv6 ospf3 interface tunnel 3 OSPFv3 configuration and statistics for Tunnel 3 Interface Status Area ID State Cost Pri Passive ------------ -------- ------------ ------ ------ ----- ------tunnel-3 Enabled 1.2.3.4 DOWN 1 1 No

Viewing a summary of OSPFv3 configuration information 245

Example 156 show ipv6 ospf3 interface detail Output HP Switch(config)# show ipv6 ospf3 interface detail OSPFv3 configuration and statistics for VLAN 22 Interface

: vlan-22

Status

: Enabled

Area ID Priority Type Hello Interval Transit Delay Events

: : : : : :

State Cost Passive Dead Interval Retransmit Interval Designated Router

: : : : : :

Neighbors

: 0

Backup Designated Router

: 0.0.0.0

1.2.3.4 1 BCAST 10 1 0

DOWN 1 No 890 5 0.0.0.0

OSPFv3 configuration and statistics for Tunnel 3 Interface

: tunnel-3

Status

: Enabled

Area ID Priority Type Hello Interval Transit Delay Events

: : : : : :

State Cost Passive Dead Interval Retransmit Interval Designated Router

: : : : : :

Neighbors

: 0

Backup Designated Router

: 0.0.0.0

0.0.0.0 1 P2P 10 1 0

DOWN 1 No 50 5 0.0.0.0

OSPFv3 configuration and statistics for Loopback 1 Interface

: lo-1

Status

: Enabled

Area ID Priority Type Hello Interval Transit Delay Events Neighbors

: : : : : : :

State Cost Passive Dead Interval Retransmit Interval Designated Router Backup Designated Router

: : : : : : :

1.2.3.4 n/a BCAST 10 n/a 0 n/a

DOWN 1 n/a 50 n/a n/a n/a

Example 157 Detail Option for a VLAN HP Switch# show ipv6 ospf3 interface vlan 55 detail OSPFv3 configuration and statistics for VLAN 55 Interface

: vlan-55

Status

Area ID : 0.0.0.1 Priority : 1 Type : BCAST Hello Interval : 10 Transit Delay : 1 Events : 0

State Cost Passive Dead Interval Retransmit Interval Designated Router

: BDR : 1 : No : 40 : 5 : 10.0.0.1

Neighbors

Backup Designated Router

: 10.0.1.4

: 1

Viewing OSPFv3 interface information for neighbor routers 246 OSPFv3 Routing

: Enabled

Syntax: show ipv6 ospf3 neighbor [ router-id ] [ detail ] Displays OSPFv3 information learned for neighbor routers. router-id Displays information for a specific neighbor router. detail Displays additional, neighbor-specific OSPFv3 information. Example 158 show ipv6 ospf3 neighbor output HP Switch(ospf3)# show ipv6 ospf3 neighbor OSPFv3 Neighbor Information Interface --------Vlan-55 Vlan-75 tunnel-3

Router ID --------------15.1.0.1 15.1.1.2 4.3.2.1

Pri ----1 1 1

State ---------FULL FULL Full

Rxmt QLen ---------0 0 0

Events -----0 0 0

Example 159 show ipv6 ospf3 neighbor detail output HP Switch(ospf3)# show ipv6 ospf3 neighbor 15.1.0.1 detail OSPFv3 Neighbor Information for neighbor 15.1.0.1 IPv6 Address Router ID : Interface : Area : Priority : Options : Events :

: fe80::55:1 15.1.0.1 State vlan-55 Designated Router 0.0.0.1 Backup Designated Router 1 Retransmit Queue Length 19 Neighbor Uptime 0 Dead Timer Expires

: : : : : :

FULL 15.1.0.1 15.1.0.4 0 1days 33 sec

: : : : : :

FULL 0.0.0.0 0.0.0.0 0 1days 33 sec

OSPFv3 Neighbor Information for neighbor 15.1.0.1 IPv6 Address Router ID : Interface : Area : Priority : Options : Events :

: fe80::55:1 15.1.0.1 State tunnel-3 Designated Router 0.0.0.1 Backup Designated Router 1 Retransmit Queue Length 19 Neighbor Uptime 0 Dead Timer Expires

Viewing or clearing OSPFv3 packet statistics counters Syntax: show ipv6 ospf3 statistics [ vlan vid | tunnel tunnel-id ] clear ipv6 ospf3 statistics [ vlan vid | tunnel tunnel-id ] Displays the statistics on OSPFv3 packets sent and received on the VLAN interfaces on an OSPFv3-enabled routing switch, including the number of errors that occurred during packet transmission. vlan vid Displays the statistics for the specified VLAN. clear Resets the OSPFv3 traffic counters to zero. Viewing a summary of OSPFv3 configuration information 247

vlan vid Resets only those counters in the specific VLAN. tunnel tunnel-id Using the tunnel option resets only those counters in the specific tunnel. Example 160 Displaying OSPFv3 traffic statistics for all VLANs configured for OSPFv3 operation HP Switch# show ipv6 ospf3 statistics OSPFv3 statistics for Interfaces Interface -----------Vlan-22 Vlan-55 tunnel-3

Total Tx --------------9022 28041 5432

Total Rx --------------9018 28480 5430

Total Errors -----------3 15 2

Example 161 Displaying OSPFv3 statistics for a single VLAN HP Switch# show ipv6 ospf3 statistics vlan 55 OSPFv3 statistics for VLAN 55 Tx Tx Tx Tx Tx

Hello Packets DD Packets LSR Packets LSU Packets LSA Packets

OSPFv3 Errors

: : : : :

26005 3 1 1436 615

Rx Rx Rx Rx Rx

Hello Packets DD Packets LSR Packets LSU Packets LSA Packets

: : : : :

26005 3 1 1046 1444

: : : : :

26000 3 1 1046 1444

: 0

Example 162 Displaying OSPFv3 Statistics for a Tunnel HP Switch# show ipv6 ospf3 statistics tunnel 3 OSPFv3 statistics for Tunnel 3 Tx Tx Tx Tx Tx

Hello Packets DD Packets LSR Packets LSU Packets LSA Packets

: : : : :

26000 3 1 1436 615

Rx Rx Rx Rx Rx

Hello Packets DD Packets LSR Packets LSU Packets LSA Packets

OSPFv3 Errors : 0

Viewing OSPFv3 link-state AS-scope information The commands in this section enable display of the routing-switch's OSPFv3 link-state AS-scope database for the entire AS, with options for narrowing the scope of the output and increasing the range of settings included in the output.

Syntax: show ipv6 ospf3 link-state as-scope [ lsid lsid-# ] [ router-id rtr-id-# ] [ detail | advertise ] Displays link-state AS-scope database for all links configured on the routing switch. The range of displayed data can be reduced by using one or more subset options:

248 OSPFv3 Routing

lsid lsid-# Subset option to filter displayed LSA database or advertisements to show only the AS-scope data having the specified (32-bit) IP address as a link-state ID. Can also be filtered with the router-id option to further define the source of displayed information. router-id rtr-id-# Subset option to filter displayed LSA database or advertisements to show only the AS-scope data having the specified router-ID. Can also be filtered with thelsid option to further define the source of displayed information. detail Displays additional details for each LSA included in the range of displayed LSAs for any of the above options. advertise Displays the hexadecimal data in LSA packets (advertisements) within the OSPFv3 AS- scope on the routing switch. The output can also be filtered by lsid and router-id. To display link-state AS-scope link-state information, enter show ipv6 ospf3 link-state as-scope at any CLI level. When you enter this command, an output similar to the following is displayed: Example 163 show ipv6 ospf3 link-state as-scope output HP Switch# show ipv6 ospf3 link-state as-scope OSPFv3 AS Scope Link State Database

LSA Type ----------As-External As-External As-External As-External

Advertising Router ID -----------15.1.1.2 15.1.1.2 15.1.1.2 15.255.155.1

Link State ID ------------0 2 3 1

Age --57 945 987 250

Sequence # ---------0x80000037 0x80000062 0x80000062 0x8000006f

Checksum ---------0x0000b76b 0x0000b3d1 0x0000dfa0 0x0000fae0

To display link-state AS-Scope LSA advertisements in hexadecimal format, use the advertise keyword. Example 164 (page 250) displays an example of this output for router ID 15.1.1.2 in an AS.

Viewing a summary of OSPFv3 configuration information 249

Example 164 show ipv6 ospf3 link-state as-scope advertise output HP Switch# show ipv6 ospf3 link-state as-scope router-id 15.1.1.2 advertise OSPFv3 AS Scope Link State Database Raw Advertisements -----------------------------------------------------------------------1d6e4005000000000f01010280000037b76b00280500000a400000002620000f00000000 00000000 19f64005000000020f01010280000062b3d100380700000a400000002620000b00000000 2620000a00000000000000000011000100000000 19cc4005000000030f01010280000062dfa000380700000a400000002621000e00000000 2620000a00000000000000000011000100000000

Viewing OSPFv3 link-state Area-Scope information The commands in this section enable display of the routing-switch's OSPFv3 link-state area-scope database, with options for narrowing the scope of the output and increasing the range of settings included in the output.

Syntax: show ipv6 ospf3 link-state area-scope area area-id lsid lsid-# router-id rtr-# type lsa-type [ detail | advertise ] Displays link-state database for all areas configured on the routing switch. The range of displayed data can be reduced by using one or more subset options: area area-id Subset option to filter displayed LSA database or advertisements to show only the data from a specific OSPFv3 area. Can also be filtered with other subset options (lsid, router-id , and type) to further define the source of displayed information. lsid lsid-# Subset option to filter displayed LSA database or advertisements to show only the data from sources having the specified IP address as a link-state ID. Can also be filtered with other subset options (area, lsid, router-id, and type) to further define the source of displayed information. router-id rtr-# Subset option to filter displayed LSA database or advertisements to show only the data from sources having the specified router ID. Can also be filtered with other subset options (area, lsid, and type) to further define the source of displayed information. type lsa-type Subset option to filter displayed LSA database or advertisements to show only the data from sources having the specified type. Can also be filtered with other subset options (area, lsid, and router-id) to further define the source of displayed information. LSA type options include: router | network | inter-area-prefix | inter-area-router | nssa | intra-area-prefix

250 OSPFv3 Routing

detail Displays additional details for each LSA included in the range of displayed LSAs for any of the above options. advertise Displays the hexadecimal data in LSA packets (advertisements) for the OSPFv3 areas configured on the routing switch. The output can also be filtered by area (area-id), lsid, router-id, and/or type. Default: All OSPFv3 areas on the routing switch. To display OSPFv3 link-state information, enter show ipv6 ospf3 link-state area-scope at any CLI level. When you enter this command, the switch displays an output similar to the following for all configured areas: Example 165 show ipv6 ospf3 link-state area-scope output HP Switch# show ipv ospf3 link-state area-scope OSPFv3 Area Scope Link State Database for area 0.0.0.0

LSA Type ----------------Router Router Router Network Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix

Advertising Router ID -----------1.0.0.4 1.1.1.1 15.255.155.1 15.255.155.1 1.0.0.4 1.0.0.4 1.0.0.4 1.0.0.4 15.255.155.1 15.255.155.1

Link State ID -------0 0 0 599 1 3 5 7 2 3

Age --2 20 1 21 22 22 22 22 61 22

Sequence # ---------0x80000037 0x8000038a 0x80000373 0x80000069 0x80000002 0x80000002 0x80000002 0x80000002 0x80000002 0x80000002

Checksum ---------0x00000a25 0x00004be4 0x00006c10 0x0000cd3e 0x00003cbf 0x00002870 0x00002273 0x00003c54 0x0000ba35 0x000080d1

Inter-Area-Prefix Inter-Area-Router Inter-Area-Router Intra-Area-Prefix

15.255.155.1 1.0.0.4 1.0.0.4 15.255.155.1

5 10 12 599

62 23 23 22

0x80000002 0x80000002 0x80000002 0x80000068

0x00006689 0x00003139 0x00005da3 0x00002bd7

OSPFv3 Area Scope Link State Database for area 0.0.0.1

LSA Type ----------------Router Router Network Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Prefix Inter-Area-Router Inter-Area-Router Intra-Area-Prefix Intra-Area-Prefix Intra-Area-Prefix

Advertising Router ID -----------1.0.0.4 15.255.155.1 15.255.155.1 1.0.0.4 1.0.0.4 15.255.155.1 15.255.155.1 1.0.0.4 15.255.155.1 1.0.0.4 15.255.155.1 15.255.155.1

Link State ID -------0 0 632 6 14 4 6 13 7 0 0 632

Age --14 13 23 25 25 64 19 25 19 15 14 24

Sequence # ---------0x80000004 0x80000004 0x80000002 0x80000002 0x80000002 0x80000002 0x80000002 0x80000002 0x80000002 0x80000004 0x80000004 0x80000002

Checksum ---------0x000085ef 0x0000ddee 0x00005ef2 0x0000187c 0x000095f9 0x000046ad 0x0000707c 0x000053ac 0x0000eb72 0x00006d56 0x000070a1 0x00003ae9

To display area-scope LSA advertisements in hexadecimal format, use the advertise keyword. Example 166 (page 252) shows an example of this output for router ID 1.0.0.4 in area 1 of an AS. Viewing a summary of OSPFv3 configuration information

251

Example 166 Output for show ipv6 ospf3 link-state area-scope advertise HP Switch# show ipv ospf3 link-state area-scope advertise area 1 router-id 1.0.0.4 OSPFv3 Area Scope Link State Database for area 0.0.0.1 Raw Advertisements -----------------------------------------------------------------------0e60200100000000010000048000000583f0002801000013020000010000027800000278 0fff9b01 0e382003000000060100000480000003167d002400000001400000002620000f00000000 0e3820030000000e010000048000000393fa002400000002400000002620000b00000000 0e3820040000000d010000048000000351ad002000000013000000010f010102 0e6a20090000000001000004800000056b57003400012001000000000100000480020000

Viewing OSPFv3 link-state link-scope information The commands in this section enable display of the routing-switch's OSPFv3 link-state link-scope database, with options for narrowing the scope of the output and increasing the range of settings included in the output.

Syntax: show ipv6 ospf3 link-state link-scope interface vlan-id lsid lsid-# router-id rtr-id-# [ detail | advertise ] Displays link-state link-scope database for all links configured on the routing switch. The range of displayed data can be reduced by using one or more subset options: interface vlan-id Subset option to filter displayed LSA database or advertisements to show only the link-scope data from a specific VLAN. Can also be filtered with other subset options (lsid and router-id) to further define the source of displayed information. lsid lsid-# Subset option to filter displayed LSA database or advertisements to show only the link-scope data having the specified (32-bit) IP address as a link-state ID. Can also be filtered with other subset options (router-id and interface) to further define the source of displayed information. router-id rtr-id-# Subset option to filter displayed LSA database or advertisements to show only the link-scope data having the specified router-ID. Can also be filtered with other subset options (lsid and interface) to further define the source of displayed information. detail Displays additional details for each LSA included in the range of displayed LSAs for any of the above options. advertise Displays the hexadecimal data in LSA packets (advertisements) within the OSPFv3 link scope on the routing switch. The output can also be filtered by lsid and router-id. To display link-scope link-state information, enter show ipv6 ospf3 link-state link-scope at any CLI level. When you enter this command, an output similar to the following is displayed: 252 OSPFv3 Routing

Example 167 show ipv6 ospf3 link-state link-scope output HP Switch# show ipv6 ospf3 link-state link-scope OSPFv3 Link Scope Link State Database for LS index 599

LSA Type -------Link Link Link Link

Advertising Router ID -----------1.1.1.1 15.255.155.1 1.0.0.4 15.255.155.1

Link State ID ------------599 599 632 632

Age --842 903 845 874

Sequence # ---------0x80000009 0x80000009 0x80000009 0x80000009

Checksum ---------0x0000fb93 0x00006185 0x00002a0c 0x0000662a

Example 168 Link-State Link-Scope Output for a Specific Tunnel HP Switch(config)# show ipv6 ospf3 link-state link-scope interface tunnel 3 OSPFv3 Link Scope Link State Database for LS index for Tunnel 3

LSA Type -------Link Link

Advertising Router ID ----------1.1.1.1 2.2.2.2

Link State ID ------------578 578

Age --106 114

Sequence # ---------0x80000004 0x80000004

Checksum ---------0x0000583b 0x000065d5

To display link-scope LSA advertisements in hexadecimal format, use the advertise keyword. Example 169 (page 253) shows an example of this output. Example 169 show ipv6 ospf3 link-state link-scope advertise output HP Switch# show ipv6 ospf3 link-state link-scope advertise OSPFv3 Link Scope Link State Database for LS index 599 Raw Advertisements ----------------------------------------------------------------------11a00008000002570101010180000009fb93003801000013fe800000000000000000000 0022000100000001400000002620000b00000000 11630008000002570fff9b01800000096185003801000013fe800000000000000000000 0022000200000001400000002620000b00000000 119e00080000027801000004800000092a0c003801000013fe800000000000000000000 0055000200000001400000002620000e00000000 11810008000002780fff9b0180000009662a003801000013fe800000000000000000000 0055000100000001400000002620000e00000000

Viewing OSPFv3 redistribution information As described in “About configuring for external route redistribution in an OSPFv3 domain (optional)” (page 269), you can configure the routing switch to redistribute connected and static routes into OSPFv3. When you redistribute a route, the routing switch can use OSPFv3 to advertise the route to its OSPFv3 neighbors.

Syntax: show ipv6 ospf3 redistribute Viewing a summary of OSPFv3 configuration information 253

Displays the route types currently enabled for route redistribution on the routing switch. Example 170 Output for show ipv6 ospf3 redistribute HP Switch# show ipv6 ospf3 redistribute OSPFv3 redistributing Route Type ---------Connected Static

RouteMap --------------------Net-01 Net-02

Viewing OSPFv3 virtual link information Syntax: show ipv6 ospf3 virtual-link [ rtr-id ] [ area area-id ] Displays OSPFv3 information learned about all virtual links detected by the routing switch. rtr-id Displays virtual link information for a specific virtual-neighbor router detected by the routing switch. area area-id Displays information learned from a virtual neighbor detected in a specific area. Example 171 Display output for all virtual links detected on the routing switch HP Switch# show ipv6 ospf3 virtual-link OSPFv3 Virtual Interface Status Transit AreaID Neighbor Router Interface State --------------- --------------- --------------0.0.0.1 1.0.0.4 P2P

Example 172 Display output for a specific virtual link HP Switch# show ipv6 ospf3 virtual-link 1.0.0.4 Transit AreaID : 0.0.0.1 Neighbor Router : 1.0.0.4 Interface State : P2P Events : 1

Transit Delay Rtr Interval Hello Interval Dead Interval

: : : :

1 5 10 40

Viewing OSPFv3 virtual neighbor information Syntax: show ipv6 ospf3 virtual-neighbor [ rtr-id ] [ area area-id ] Displays OSPFv3 information learned about all virtual neighbor routers detected by the routing switch.

254 OSPFv3 Routing

rtr-id Displays information for a specific virtual-neighbor router detected by the routing switch. area area-id Displays information learned from a virtual neighbor detected in a specific area. Example 173 Display output for all virtual neighbors detected on the routing switch HP Switch# show ipv6 ospf3 virtual-neighbor OSPFv3 Virtual Interface Neighbor Information Router ID State IPv6 Addr Events ------------- -------- ------------------------- --------1.0.0.4 FULL 2620:e::55:2 7

Example 174 Display output for a specific virtual neighbor HP Switch# show ipv6 ospf3 virtual-neighbor 1.0.0.4 Router ID : 1.0.0.4 State : FULL IPv6 Addr : 2620:e::55:2 RtQLen : 5 Events : 7

Viewing OSPFv3 SPF statistics Enter this command to display the log used to record SPF calculations on an OSPFv3-enabled routing switch. The SPF algorithm recalculates the routes in an OSPFv3 domain when a change in the area topology is received.

Syntax: show ipv6 ospf3 spf-log Displays the event that resulted in the last 100 executions of the SPF algorithm on the routing switch. Possible events (reasons) are: Re-Init OSPFv3 was enabled or disabled on the routing switch. Router LS Update A Router-LSA was received. Network LS Update A Network-LSA was received. Generated RTR LSA A Router-LSA was generated on the routing switch. Generated NTW LSA A Network-LSA was generated on the routing switch.

Viewing a summary of OSPFv3 configuration information 255

Example 175 Displaying the OSPFv3 SPF log HP Switch(ospf3)# show ipv6 ospf3 spf-log OSPFv3 SPF (SHORTEST PATH FIRST) LOG spf instance --------------1 2 3 4 5 6 7 8 9 10 11 ...

Reason --------------------------Router LS Update Router LS Update Generated RTR LSA Generated NTW LSA Network LS Update Network LS Update Generated RTR LSA Router LS Update Generated RTR LSA Re-Init Incremental LS Update ...

Debugging OSFP routing messages Syntax: debug ipv6 ospf3 [ adj | event | flood | lsa-generation | packet | retransmission | spf ] Turns on the tracing of OSPFv3 packets. For more information, see the "Debug Command" section in the "Troubleshooting" appendix of the Management and Configuration Guide for your routing switch.

Graceful Shutdown of OSPF Routing It is now possible to gracefully shut down OSPF routing on HP switches without losing packets that are in transit. OSPF neighbors are informed that the router should not be used for forwarding traffic, which allows for maintenance on the switch without interrupting traffic in the network. There is no effect on the saved switch configuration. Prior to a switch shutdown, the CLI/SNMP reload command or the CLI boot command is executed to initiate the sending of OSPF “empty Hello list” messages on the interfaces that are part of the OSPF routing configuration. After a small delay (approximately 2 seconds) that allows the messages to be transmitted on all applicable interfaces, the boot or reload command continues.

Modules Operating in NonStop Mode When a switch is in standalone mode and OSPF routing is enabled, the “empty Hello list” is transmitted whenever the boot or reload commands are executed. When the switch is operating in nonstop switching mode (redundant) and a single module is being reloaded or booted, the standby module will notify neighboring switches of the management module failover. If the failover fails, the “empty Hello list” is transmitted before the switch is rebooted. When a switch is operating with multiple management modules in warm standby mode, the “empty Hello list” is sent when a reload or boot command is executed. The standby management module sends out OSPF Hello packets after becoming the active management module.

Enabling load-sharing among next-hop routes For more information about the equal-cost multi-path (ECMP) feature, see “About equal-cost multi-path routing” (page 266).

256 OSPFv3 Routing

Syntax: [no] ip load-sharing 2 - 4 When OSPF is enabled and multiple, equal-cost, next-hop routes are available for traffic destinations on different subnets, this feature, by default, enables load-sharing among up to four next-hop routes. 2 - 4 Specifies the maximum number of equal-cost next-hop paths the router allows. Default: Enabled with four equal-cost, next-hop routes allowed The no form of the command disables this load-sharing so that only one route in a group of multiple, equal-cost, next-hop routes is used for traffic that could otherwise be load-shared across multiple routes. NOTE: Disabling load-sharing means that router "1" selects only one next-hop router for traffic that is actually eligible for load-sharing through different next-hop routers. In the default configuration, load-sharing is enabled by default for both IPv4 and IPv6. However, it has no effect unless routing and OSPF are enabled. For example, in Figure 22 (page 257), the next-hop routers "B," "C," and "D" are available for equal-cost load-sharing of eligible traffic. Disabling IP load-sharing means that router "A" selects only one next-hop router for traffic that is actually eligible for load-sharing through different next-hop routers. In Figure 22 (page 257), the ECMP inter-area routes to destination network 2001:db8:0:e::/64 consist of the following next-hop gateway addresses: •

2001:db8:0:b::b:101



2001:db8:0:c::c:101



2001:db8:0:d::d:101

Figure 22 OSPFv3 ECMP multiple next-hop routing (inter-area)

However, the forwarding software distributes traffic across the three possible next-hop routes in such a way that all traffic for a specific host is sent to the same next-hop router. Enabling load-sharing among next-hop routes 257

As shown in Figure 22 (page 257), one possible distribution of traffic to host devices is: •

Traffic to host "A" passes through next-hop router "3"



Traffic to host "B" passes through next-hop router "2"



Traffic to host "C" passes through next-hop router "3"



Traffic to host "D" passes through next-hop router "4" IP packet destination

Next hop used

2001:db8:0:e::100

2001:db8:0:b::b:10

2001:db8:0:e::110

2001:db8:0:c::c:20

2001:db8:0:e::120

2001:db8:0:b::b:10

2001:db8:0:e::130

2001:db8:0:d::d:30

Displaying the current IP load-sharing configuration Use show running to view the currently active load-sharing configuration and show config to view the load-sharing configuration in the startup-config file. (While in its default configuration—ip load-sharing 4—load-sharing does not appear in the command output.) If load sharing is configured with non-default settings (disabled or configured for either two or three equal-cost next-hop paths), the current settings are displayed in the command output. Example 176 Displaying a non-default IP load-sharing configuration HP Switch(config)# show running Running configuration: ; J8697A Configuration Editor; Created on release #K.15.xx hostname "HP Switch" module 1 type J8702A snmp-server community "public" Unrestricted vlan 1 name "DEFAULT_VLAN" untagged 1-24 ip address dhcp-bootp exit ip load-sharing 3

NOTE: ip load-sharing 3 indicates a non-default load-sharing configuration allowing three equal-cost next-hop paths for routed traffic with different subnet destinations. If the routing switch is configured with the default load-sharing configuration, load-sharing does not appear in the show config or show running command output.

License requirements In the 3500, 3500yl, 5400zl, 6600, and 8200zl switches, OSPFv3 is included with the optional Premium License. In the 6200yl switches, this feature is included with the base feature set.

258 OSPFv3 Routing

Overview of OSPFv3 Factor

Detail

Minimum software version

K.15.01

Application

OSPFv3 applications only; runs independent of the OSPFv2 protocol used for IPv4 OSPFv2 applications.

Concurrent IPv4/IPv6 operation

Concurrent OSPFv2 and OSPFv3 operation supported on all VLAN interfaces configured on the routing switch.

Beginning with software version K.15.01, the routing switches support concurrent operation of both OSPFv2 (for IPv4) and OSPFv3 (for IPv6). The two versions of the OSPFv3 protocol operate independently of each other. (For information on OSPFv2 for IPv4 operation, see the latest Multicast and Routing Guide for your routing switch.) OSPFv3 is a link-state routing protocol applied to IPv6 routers grouped into OSPFv3 areas identified by the IPv6 routing configuration on each routing switch. Each OSPFv3 area includes one or more networks. OSPFv3 routers use hello packets and LSAs to maintain OSPFv3 operation across networks within an area and between areas within an OSPFv3 domain.

Hello packets OSPFv3 uses hello packets to initiate and preserve relationships between neighboring routers on the same VLAN interface. OSPFv3 automatically transmits hello packets that are configurable for transmission interval (to indicate a link that has become unavailable).

Link-state advertisements (LSAs) OSPFv3 uses LSAs transmitted by each router to update neighboring routers regarding its interfaces and the routes available through those interfaces. Each routing switch in an area also maintains an LSDB that describes the area topology. (All routers in a given OSPFv3 area have identical LSDBs, and each router uses the LSDB to build its own shortest-path tree.) The routing switches used to connect areas to each other flood inter-area-prefix-LSAs, inter-area-router-LSAs, and AS-external-LSAs to neighboring OSPFv3 areas to update them regarding available routes. Through this means, each OSPFv3 router determines the shortest path between itself and a desired destination router in the same OSPFv3 domain (AS). Routed traffic in an OSPFv3 AS is classified as one of the following: •

Intra-area traffic



Inter-area traffic



External traffic

The routing switches support the LSAs listed in the following table. For more information, see RFCs and 3101 (for Type-7-LSA). Table 27 OSPFv3 LSA types LSA type

Description

Use

Flood scope

0x2001

Router-LSA

Describes the state of each active interface on a router for a given area. (Excludes loopback interfaces and interfaces that have not achieved full adjacency.)

Area

0x2002

Network-LSA

Describes the OSPFv3 routers in a given network.

Area

0x2003

Inter-area-prefix-LSA

Describes the route to a prefix in another OSPFv3 area of the same AS. Propagated through backbone area to other areas.

Area

Overview of OSPFv3 259

Table 27 OSPFv3 LSA types (continued) LSA type

Description

Use

Flood scope

0x2004

Inter-area-router-LSA

Describes the route to an ASBR in another OSPFv3 normal area Area (including the backbone area) of the same AS. (Excludes prefixes for link-local addresses.) Propagated through backbone area to other areas. (Excludes any ASBR in the same area as the router sending the LSA.)

0x2005

AS-external-LSA

Describes the route to a destination prefix in another AS (external AS route). (Excludes prefixes for link-local addresses.) Originated by ASBR in normal or backbone areas of an AS and propagates through backbone area to other normal areas. Does not flood over virtual links and is not summarized in virtual links. For injection into an NSSA, an NSSA ABR generates a type-7-default-LSA advertising the default route (::/0).

0x2007

Type-7-LSA

Describes the route to a destination in another AS (external NSSA route). Originated by ASBR in NSSA. ABR translates type-7 LSAs to AS-external-LSAs for injection into the backbone area.

0x2008

Link-LSA

For other routers on the same VLAN interface, describes the Link-local router's link-local address and any other IPv6 prefixes reachable on the VLAN. Link LSAs are not flooded over virtual links.

0x2009

Intra-area-prefix-LSA

Generated on transit links within an area by the DR operating on those links. Also, every OSPFv3 router generates this LSA to refer to stub and loopback prefixes on the router.

Area

OSPFv3 router types Interior routers This type of OSPFv3 router belongs to only one area. Interior routers flood router-LSAs to all routers in the same area and maintain identical LSDBs. In Figure 23 (page 260), routers R1, R3, R4, and R6 are all interior routers because they link to other routers in the same area. Figure 23 Example of interior routers Area 0 (Backbone)

Interior Router

R1 R5

R2

R3

Area 1

R4

2 ea Ar

R6

Interior Routers

Area border routers (ABRs) This type of OSPFv3 router has membership in multiple areas. ABRs are used to connect the various areas in an AS to the backbone area for that AS. Multiple ABRs can be used to connect a given area to the backbone, and a given ABR can belong to multiple areas other than the backbone. 260 OSPFv3 Routing

An ABR maintains a separate LSDB for each area to which it belongs. (All routers within the same area have identical LSDBs.) The ABR is responsible for flooding inter-area-prefix-LSAs and inter-area router LSAs between its border areas. You can reduce this LSA flooding by configuring area ranges. An area range enables you to assign an aggregate address to a range of IPv6 addresses. This aggregate address is advertised instead of all the individual addresses it represents. You can assign up to eight ranges in an OSPFv3 area. In Figure 24 (page 261), routers R2 and R5 are ABRs because they both have membership in more than one area. Figure 24 Example of deploying ABRs to connect areas to the backbone Area Border Router (ABR)

Area 0 (Backbone) Area Border Router (ABR)

R3

R1

R5

R2

Area 1

R4

2 ea Ar

R6

Autonomous system boundary router (ASBR) This type of OSPFv3 router runs one or more Interior Gateway protocol and serves as a gateway to other autonomous systems operating with interior gateway protocols. The ASBR imports and translates different protocol routes into OSPFv3 through redistribution. ASBRs can be used in backbone areas, normal areas, and NSSAs, but not in stub areas. For more details on redistribution and configuration examples, see “Enabling route redistribution” (page 226).

Designated routers In an OSPFv3 network having two or more routers, one router is elected to serve as the designated router (DR) and another router to act as the backup designated router (BDR). All other routers in the area forward their routing information to the DR and BDR, and the DR forwards this information to all routers in the network. This minimizes the amount of repetitive information that is forwarded on the network by eliminating the need for each individual router in the area to forward its routing information to all other routers in the network. If the area includes multiple networks, each network elects its own DR and BDR. In an OSPFv3 network with no DR and no BDR, the neighboring router with the highest priority is elected as the DR, and the router with the next highest priority is elected as the BDR. If the DR goes off-line, the BDR automatically becomes the DR, and the router with the next highest priority then becomes the new BDR. If multiple HP routing switches on the same OSPFv3 network are declaring themselves as DRs, both priority and router ID are used to select the designated router and backup designated routers. Priority is configurable by using the vlan vid ipv6 ospfv3 priority 0 - 255 command at the interface level. You can use this parameter to help bias one router as the DR. For more information on this command, see “Priority per interface” (page 233). If two neighbors share the same priority, the router with the highest router ID is designated as the DR. The router with the next highest router ID is designated as the BDR.

OSPFv3 router types

261

For example, in Figure 25 (page 262), the DR and BDR for the 2001:db8:0:5::/64 network in area 5 are determined as follows: •

Router A Priority: 0 Cannot become a DR or BDR.



Router B Priority: 3 DR for the 2001:db8:0:5::/64 network.



Router C Priority: 2 BDR for the 2001:db8:0:5::/64 network.



Router D Priority: 0 Cannot become a DR or BDR.



Router E Priority: 1 Becomes the new BDR if router B becomes unavailable and router C becomes the new DR.

Figure 25 Example of designated routers in an OSPFv3 area

Area 0 (Backbone)

Router "X" 2001:db8:0::1:2/64 ID: 9.1.1.2

Router "A" 2001:db8:0::1:1/64 ID: 9.1.1.1 Router "E" 2001:db8:0:5::10:4/64 ID:10.1.1.4 Priority: 1

Router "D" 2001:db8:0:5::10:5/64 ID:10.1.1.5 Priority: 0

2001:db8:0:5::10:1/64 ID:10.1.1.1 Priority 0

Router "B" 2001:db8:0:5::10:2/64 ID:10.1.1.2 Priority: 3

Area 5

Router "C" 2001:db8:0:5::10:3/64 ID:10.1.1.3 Priority: 2

To verify the router priority on an interface, use the show ipv6 ospf3 interface command and check the Pri field. NOTE: Once a DR is elected, the DR and BDR status do not change if a higher-priority router joins the network, unless the DR or BDR goes down. By default, the router ID is typically the lowest-numbered IPv4 loopback address or the lowest-numbered (user-configured) loopback interface configured on the device. For more information, or to change the router ID, see “System router ID” (page 183). If multiple networks exist in the same OSPFv3 area, the recommended approach is to ensure that each network uses a different router as its DR. Otherwise, if a router is a DR for more than one network, latency in the router could increase because of the increased traffic load resulting from multiple DR assignments. When only one router on an OSPFv3 network claims the DR role despite neighboring routers with higher priorities or router IDs, this router remains the DR. This is also true for BDRs. 262 OSPFv3 Routing

The DR and BDR election process is performed when one of the following events occurs: •

An interface is in a waiting state and the wait time expires



An interface is in a waiting state and a hello packet is received that addresses the BDR



A change in the neighbor state occurs, such as:



A neighbor state transitions from 2 or higher



Communication to a neighbor is lost



A neighbor declares itself to be the DR or BDR for the first time

OSPFv3 area types OSPFv3 is built upon a hierarchy of network areas. All areas for a given OSPFv3 domain reside in the same AS. An AS is defined as a number of contiguous networks, all of which share the same interior gateway routing protocol. An AS can be divided into multiple areas, including the backbone (area 0). Because each area represents a collection of contiguous networks and hosts, the topology of a given area is not known by the internal routers in any other area. Areas define the boundaries to which router-LSAs and network-LSAs are broadcast. This limits the amount of LSA flooding that occurs within the AS and also helps to control the size of the link-state databases (LSDBs) maintained in OSPFv3 routers. An area is represented in OSPFv3 by either a 32-bit dotted-decimal address or a number. Area types include: •

Backbone



Normal



Not-so-stubby (NSSA)



Stub

All areas in an AS must connect with the backbone through one or more ABRs. If a normal area is not directly connected to the backbone area, it must be configured with a virtual link to an ABR that is directly connected to the backbone. The stub and NSSA area types do not allow virtual link connections to the backbone area. Figure 26 Example of an AS with multiple areas and external routes

External (IGP) Domain

Backbone Area

AB R

NSSA

ABR

ABR

Normal Area ASBR

ASBR

External (IGP) Domain

Stub Area

OSPFv3 Domain

ABR

Normal Area Virtual Link

Normal area This area type allows inter-area-prefix-LSAs and AS-external-LSAs to and from the backbone area. (As noted earlier, the backbone area is a special type of normal area.) A normal area connects OSPFv3 area types 263

to the AS backbone area through one or more ABRs (physically or through a virtual link). ASBRs are allowed in normal areas.

Backbone area Every AS must have one (and only one) backbone area (identified as area 0 or 0.0.0.0). The ABRs of all other areas in the same AS connect to the backbone area, either physically through an ABR or through a configured, virtual link. The backbone is a special type of normal area and serves as a transit area for carrying the inter-area-prefix-LSAs, AS-external-LSAs, and routed traffic between non-backbone areas, as well as the router-LSAs and network-LSAs and routed traffic internal to the area. ASBRs are allowed in backbone areas.

Stub area This area connects to the AS backbone through one or more ABRs. It does not allow an internal ASBR and does not allow AS-external-LSAs. A stub area supports these actions: •

Advertise the area's inter-area routes to the backbone area.



Advertise inter-area routes from other areas.



Use the inter-area-prefix-LSA default route to advertise routes to an ASBR and to other areas.

You can configure the stub area ABR to do the following: •

Suppress advertising some or all of the area's summarized internal routes into the backbone area.



Suppress LSA traffic from other areas in the AS by replacing inter-area-prefix-LSAs and the default external route from the backbone area with the default route (::/0).

Virtual links are not allowed for stub areas.

Not-so-stubby-area (NSSA) This area type connects to the backbone area through one or more ABRs. NSSAs are used where an ASBR exists in an area where you want to: •

Block injection of external routes from other areas of the AS.



Advertise type-7-LSA external routes (learned from the ASBR) to the backbone area as AS-external-LSAs.

NSSAs also support the following: •

Advertise inter-area-prefix-LSAs from the backbone area into the NSSA. (If no-summary is enabled, the NSSA ABR suppresses these LSAs from the backbone and, instead, injects the inter-area-prefix-LSA default route into the NSSA.)



Advertise NSSA inter-area-prefix-LSAs to the backbone area.

In the above operation, the ASBR in the NSSA injects external routes as type-7-LSAs. (AS-external-LSAs are not allowed in an NSSA.) The ABR connecting the NSSA to the backbone converts the type-7-LSAs to AS-external-LSAs and injects them into the backbone area for propagation to networks in the backbone and to any normal areas configured in the AS. The ABR also injects inter-area-prefix-LSAs from the backbone area into the NSSA. The default route (::/0) is always injected into the NSSA as either a type-7-LSA or an inter-area-LSA, depending on the no-summary configuration (default: disabled). That is, if inter-area-prefix-LSAs are allowed in the NSSA (the default operation), a type-7-LSA default route (::/0) is injected into the NSSA. But if inter-area-prefix-LSAs are blocked (by enabling no-summary), the inter-area-prefix-LSA default route is injected into the NSSA instead of the type-7-LSA default route. You can also configure the NSSA ABR to suppress advertising some or all of the area's summarized internal or external routes into the backbone area. See “Configuring ranges on an ABR to reduce advertising to the backbone” (page 229). 264 OSPFv3 Routing

Virtual links are not allowed for NSSAs.

Reducing AS-external-LSAs and inter-area-prefix-LSAs An OSPFv3 ASBR uses AS-external-LSAs to originate advertisements of a route to another routing domain. These advertisements are: •

Flooded in the area in which the ASBR operates.



Injected into the backbone area and then propagated to any other OSPFv3 areas (except stub and NSSA areas) within the local OSPFv3 AS. If the AS includes an NSSA, there are two additional options: •

If the NSSA includes an ASBR, you can suppress advertising some or all of its summarized external routes into the backbone area. See “Configuring ranges on an ABR to reduce advertising to the backbone” (page 229).



Replace all inter-area-prefix-LSAs and all external routes from the backbone area with the default route (::/0).

Algorithm for AS-external-LSA reduction The AS-external-LSA reduction feature behavior changes under the following conditions: •

There is one ASBR advertising (originating) a route to the external destination, but one of the following happens: •

A second ASBR comes online.



A second ASBR that is already online begins advertising an equivalent route to the same destination.

In either case above, the routing switch with the higher router ID floods the AS-external-LSAs and the other routing-switch flushes its equivalent AS-external-LSAs. •

One of the ASBRs starts advertising a route that is no longer equivalent to the route the other ASBR is advertising. In this case, the ASBRs each flood AS-external-LSAs. Since the LSAs either no longer have the same cost or no longer have the same next-hop router, the LSAs are no longer equivalent, and the LSA reduction feature no longer applies.



The ASBR with the higher router ID becomes unavailable or is reconfigured so that it is no longer an ASBR. In this case, the other ASBR floods the AS-external-LSAs.

About replacing inter-area-prefix-LSAs and type-7-external-LSA default routes with an AS-external-LSA default route By default, a routing switch operating as an ABR for a stub area or NSSA injects non-default, inter-area routes (inter-areaprefix-LSAs) into the stub areas and NSSAs. For NSSAs, the routing switch also injects a type-7-LSA default external route. You can further reduce LSA traffic into these areas by using no-summary. This command option configures the routing switch to: •

Replace injection of inter-area-prefix-LSAs into a stub area or NSSA with an inter-area-prefix-LSA default summary route (::/0).



Replace injection of all external routes into an NSSA with an inter-area-prefix-LSA default route.

You can enable this behavior when you first configure the stub area or NSSA, or at a later time. For the full command to use, see “Configuring a stub or NSSA area” (page 222). The no-summary command does not affect intra-area advertisements, meaning the switch still accepts summary LSAs from OSPFv3 neighbors within its area and floods them to other neighbors. The switch can form adjacencies with other routers regardless of whether summarization is enabled or disabled for areas on each switch.

Reducing AS-external-LSAs and inter-area-prefix-LSAs 265

When you use no-summary, the change takes effect immediately. If you apply the option to a previously configured area, the switch flushes all of the summary LSAs it has generated (as an ABR) from the area. NOTE: This feature applies only when the routing-switch is configured as an ABR for a stub area or NSSA. To completely prevent summary LSAs from injection into the area, use no-summary to disable the summary LSAs on each OSPFv3 router that is an ABR for the area. To implement the above operation for a stub area or NSSA, enter a command such as the following: HP Switch(ospf3)# area 40 stub metric-cost 3 no-summary For the full command syntax, see “Configuring a stub or NSSA area” (page 222).

About equal-cost multi-path routing The ECMP feature allows OSPFv3 to add routes with multiple next-hop addresses and with equal costs to a given destination in the forwarding information base (FIB) on the routing switch. For example, if multiple, equal-cost, next-hop routes exist on a routing switch for a destination in a network with the prefix 2620:e::/64, these routes would appear similar to the following in the IPv6 Route Entries Table:

266 OSPFv3 Routing

Example 177 show ipv6 route command output with multiple next-hop routes HP Switch(config)# show ipv6 route IPv6 Route Entries Destination : ::1/128 Gateway : lo0 Type: connected Sub-Type: NA

Distance: 0

Metric: 1

Destination : 2620:c::/64 Gateway : 2620:e::55:2 Type: static Sub-Type: NA

Distance: 200

Metric: 1

Destination : 2620:a::/64 Gateway : fe80::22:3%vlan22 Type: ospf3 Sub-Type: InterArea

Distance: 110 1 Metric: 2

Destination : 2620:a::/64 Gateway : fe80::22:5%vlan22 Type: ospf3 Sub-Type: InterArea

Distance: 110

Metric: 2

Destination : 2620:a::/64 Gateway : fe80::22:11%vlan22 Type: ospf3 Sub-Type: InterArea

Distance: 110

Metric: 2

Destination : 2620:b::/64 Gateway : VLAN22 Type: connected Sub-Type: NA

Distance: 0

Metric: 1

1

Multiple next-hop gateway addresses are displayed for the destination network 2620:a::/64

For a given destination network in an OSPFv3 domain, multiple ECMP next-hop routes can be one of the following types. •

Intra-area (routes to the destination in the same OSPFv3 area)



Inter-area (routes to the destination through another OSPFv3 area)



External (routes to the destination through another autonomous system)

Multiple ECMP next-hop routes cannot be a mixture of intra-area, inter-area, and external routes. For example, in Example 177 (page 267), the multiple next-hop routes to network 2620:a::/64 are all inter-area. Also, according to the distributed algorithm used in the selection of ECMP next-hop routes: •

Intra-area routes are preferred to inter-area routes.



Inter-area routes are preferred to external routes through a neighboring AS.

In addition, ECMP ensures that all traffic forwarded to a given host address follows the same path, which is selected from the possible next-hop routes. ECMP load-sharing does not affect routed traffic to different hosts on the same subnet. That is, all traffic for different hosts on the same subnet will go through the same next-hop router. For example, if subnet 2001:db8:0:1f::/64 includes two servers at 2001:db8:0:1f::1ab.101 and 2001:db8:0:1f::1ab.93, all traffic from router "A" to these servers will go through the same next-hop router.

About equal-cost multi-path routing 267

OSPFv3 Activation and Dynamic Configuration All configuration commands affecting OSPFv3 (except reconfiguring the router ID) are dynamically implemented and can be used without restarting OSPFv3 routing. To reconfigure the router ID, see “System router ID” (page 183).

General configuration steps for OSPFv3 To begin using OSPFv3 on the routing switch: 1. Enable IPv6 on at least one VLAN interface. 2. In the global config context, use ipv6 unicast-routing to enable routing (see page 221). 3. Execute router ospf3 enable to enable OSPFv3 routing (see page 221). 4. Use area in the ospf3 context to assign the areas to which the routing switch will be attached (see page 221). 5. Assign VLAN interfaces to the configured areas by moving to each VLAN context and using the command ipv6 ospf3 area ospf-area-id assigns all interfaces in the VLAN to the same area. 6. Optional: Assign loopback interfaces to OSPFv3 areas by using the interface loopback 0 - 7 ipv6 ospf3 area command. (See page 225) 7. Optional: On each routing switch used as an ASBR in your OSPFv3 domain (see page 269): a. Configure route-maps to permit route prefixes you want redistributed in your OSPFv3 domain and to deny all others. b. Configure redistribution to enable importing the static and connected routes you want to make available in the domain. 8. 9. 10. 11.

Optional: Configure ranges on ABRs to reduce inter-area RA (see page 229). Optional: Use administrative distance to influence route choices (see page 231). Optional: Enforce strict LSA operation for graceful restart helper mode (see page 232). Optional: Adjust performance by changing the IP routing interface settings, if needed. Includes cost, dead-interval, hello-interval, priority, and others (see page 232). 12. Configure virtual links for any areas not directly connected to the backbone (see page 271).

Configuration rules •

If the switch is to operate as an ASBR, you must enable redistribution; see “General configuration steps for OSPFv3” (page 268). When you do so, ASBR capability is automatically enabled. For this reason, you should first configure route policy and redistribution filters on the ASBR. Otherwise, all possible external routes will be allowed to flood the domain (see page 269).



Each IP routing interface on which you want OSPFv3 to run must be assigned to one of the defined areas. When a VLAN interface is assigned to an area, the IPv6 addresses configured on that VLAN are automatically included in the assignment.

OSPFv3 global and interface settings When first enabling OSPF, you may want to consider configuring ranges and restricting redistribution (if an ASBR is used) to avoid unwanted advertisements of external routes. You may also want to enable OSPFv3 traps to enhance troubleshooting. However, it is generally recommended that the remaining parameters with non-null default settings be left as-is until you have the opportunity to assess OSPFv3 operation and determine whether any adjustments to default settings is warranted. For information on when to use the global and per-interface commands used with OSPFv3, see “Configuring OSPFv3 on the routing switch” (page 220). For detailed information on each command, see the page listed for each command.

268 OSPFv3 Routing

NOTE: Before enabling OSPFv3, ensure that ipv6 unicast-routing is enabled. Also, either begin each command with router ospf3, or execute router ospf3 at the global CONFIG level and then execute the individual commands in that context. For example: HP Switch(config)# router ospf3 HP Switch(ospf3)# enable Use the appropriate interface context to set interface level OSPFv3 parameters for the desired interface. To access this context level, use vlan vid or interface tunnel tunnel-id either to move to the interface context level or to specify that context from the global config level. For example: HP HP HP HP HP HP

Switch(config)# vlan 20 Switch(vlan-20)# cost 15 Switch(config)# vlan 20 cost 15 Switch(config)# interface tunnel 3 Switch(tunnel-3)# ipv6 ospf3 cost 15 Switch(config)# interface tunnel 3 ipv6 ospf3 cost 15

OSPFv3 redistribution of loopback addresses When you assign an IPv6 address to a loopback interface on a routing switch, the address is listed as connected in the route table on that routing switch and is advertised to neighbors as described in Table 28 (page 269). Table 28 Route redistribution of loopback addresses Loopback address assignment

Route redistribution enabled

Route redistribution disabled

Loopback not assigned to an OSPFv3 The loopback address is advertised to The loopback address is not advertised area neighbors as an OSPFv3 External to neighbors. route. Loopback is assigned to an OSPFv3 area

The loopback address is advertised to neighbors in the same area as an OSPFv3 Intra-Area route. For all other areas it is advertised as an OSPFv3 Inter-Area route.

About configuring for external route redistribution in an OSPFv3 domain (optional) Configuring route redistribution for OSPFv3 establishes the routing switch as an ASBR (residing in a backbone, normal, or NSSA) for importing and translating different protocol routes from other IGP domains into an OSPFv3 domain. The switches support redistribution for static routes and directly connected routes. When you configure redistribution for OSPF, you can specify that static or connected routes external to the OSPFv3 domain are imported as OSPFv3 routes. The steps for configuring external route redistribution to support ASBR operation include the following: 1. Optional: Configure route-maps to permit and/or deny route prefixes for redistribution in your OSPFv3 domain. 2. Enable route redistribution. 3. Optional: Modify the default metric for redistribution. 4. Optional: Modify the redistribution metric type.

OSPFv3 redistribution of loopback addresses 269

5.

Optional: Change the administrative distance setting. NOTE: In the default configuration, redistribution is permitted for all routes from supported sources. Enable redistribution after you have configured route-maps defining the route policies you want to apply to route redistribution in the OSPFv3 domain. Otherwise, your AS may become overloaded with routes that you did not intend to redistribute.

About configuring ranges on an ABR to reduce advertising to the backbone Configuring ranges does the following to reduce inter-area advertising: Summarizing routes Enables a routing switch operating as an ABR to use a specific prefix to summarize a range of IPv6 addresses into a single RA for injection into the backbone. This results in only one address being advertised to the network instead of all the addresses within that range. This reduces LSA traffic and the resources needed to maintain routing tables. Blocking routes Prevents an ABR from advertising specific networks or subnets to the backbone area. Each OSPFv3 area supports up to eight range configurations.

About influencing route choices by changing the administrative distance default (optional) The administrative distance value can be left in its default configuration setting (110) unless a change is needed to improve OSPFv3 performance for a specific network configuration. The switch can learn about networks from various protocols. Consequently, the routes to a network may differ depending on the protocol from which the routes were learned. On the routing switches, the administrative distance for OSPFv3 routes is set at 110 for all route types (external, inter-area, and intra-area). The routing switch selects routes on the basis of route source information. To enable this operation, the administrative distance assigned to each source is used to influence route choices. You can change the distance settings in the OSPFv3 global context to enable preference of one route type over another.

About enforcing strict LSA operation for graceful restart helper mode (optional) OSPFv3 operation on the routing switches includes helper mode operation for graceful restart of OSPFv3 on a neighboring router upon receipt of a "grace LSA" from the neighbor. In the default configuration, helper mode operation in this case includes terminating graceful restart support ("strict LSA" operation) on the routing switch if it detects a topology change requiring updated LSAs during the restart period of the neighboring router. Terminating this support forces the helper routing switch to re-establish its LSAs and OSPFv3 functions on the network segment affected by the OSPFv3 restart on the neighboring router. (For more information on OSPFv3 graceful restart, see RFC 3623.) In the default OSPFv3 configuration, the default helper mode operation terminates graceful restart if topology changes affect the network segment.

270 OSPFv3 Routing

About adjusting performance by changing the VLAN interface settings (optional) The following OSPFv3 interface parameters are automatically set to their default values. No change to the defaults is usually required unless needed for specific network configurations. Parameter

Default

Page

cost

1

232

dead-interval

40 seconds

232

hello-interval

10 seconds

233

priority

1

233

retransmit-interval

5 seconds

233

transit-delay

1 second

233

passive

disabled

237

Settings are configured on a per-interface basis. NOTE: Most of these parameters also apply to virtual link configurations. However, when used on a virtual link configuration, the OSPFv3 context requirement is different and the parameters are applied only to the interfaces included in the virtual link. See “About adjusting virtual link performance by changing the interface settings” (page 272).

About configuring an ABR to use a virtual link to the backbone All OSPFv3 ABRs (area border routers) must have either a direct, physical or indirect, virtual link to the OSPFv3 backbone area (0.0.0.0 or 0). If an ABR does not have a physical link to the area backbone, it can use a virtual link to provide a logical connection to another ABR having a direct physical connection to the area backbone. Both ABRs must belong to the same area, and this area becomes a transit area for traffic to and from the indirectly connected ABR. NOTE: A backbone area can be purely virtual with no physical backbone links. Also note that virtual links can be linked in a series. If so, one end may not be physically connected to the backbone. Because both ABRs in a virtual link connection are in the same OSPFv3 area, they use the same transit area ID. This setting is configured using area area-id virtual-link router-id in the router ospf3 context and should match the area ID value configured on both ABRs in the virtual link. The ABRs in a virtual link connection also identify each other with a neighbor router setting: •

On the ABR having the direct connection to the backbone area, the neighbor router is the router ID (in decimal or 32-bit dotted decimal format) of the router interface needing a logical connection to the backbone.



On the opposite ABR (the one needing a logical connection to the backbone), the neighbor router is the router ID (in decimal or 32-bit dotted decimal format) of the ABR that is directly connected to the backbone.

About adjusting performance by changing the VLAN interface settings (optional)

271

NOTE: By default, the router ID is the lowest numbered IPv4 address or (user-configured) IPv4 loopback interface configured on the device. For more information or to change the router ID, see “System router ID” (page 183). When you establish an area virtual link, you must configure it on both of the ABRs (both ends of the virtual link).

About adjusting virtual link performance by changing the interface settings The following OSPFv3 interface parameters are automatically set to their default values for virtual links. No change to the defaults is usually required unless needed for specific network conditions. This is a subset of the parameters described under “Adjusting performance by changing the VLAN interface settings” (page 232). (The cost and priority settings are not configurable for a virtual link, and the commands for reconfiguring the settings are accessed in the router OSPFv3 context instead of the VLAN context.) NOTE: The parameter settings described in this section for virtual links must be the same on the ABRs at both ends of a given link. Parameter

Default

Page

dead-interval

40 seconds

232

hello-interval

10 seconds

233

retransmit-interval 5 seconds transit-delay

1 second

233 233

OSPFv3 passive OSPFv3 sends LSAs to all other routers on the same VLAN interface. With OSPFv3 configured as passive on a VLAN interface, the routing switch is identified as a route in the OSPFv3 domain, but does not form an adjacency to any other router and does not send or receive OSPFv3 traffic on the subject VLAN interface. (A VLAN configured as passive operates similar to a VLAN connected to a stub network and does advertise the interface as a stub link into OSPFv3.) Up to 128 active interfaces and a combined total of 512 active and passive interfaces are supported on the routing switch.

272 OSPFv3 Routing

11 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels Table 29 Summary of Commands Command syntax

Description

Default

[no] interface tunnel 1 - 128 tunnel [ enable | disable ] [no] tunnel name string

Configuring a Tunnel Interface

[no] tunnel mode [ 6in4 | unspecified ]

Configuring the Tunnel Mode

n/a

275

[no] tunnel source [ ipv4-addr | ipv6-addr ]

Configuring the Tunnel Source

n/a

275

[no] tunnel destination [ ipv4-addr | IPv6-addr ]

Configuring the Tunnel Destination

n/a

276

[no] tunnel mtu [ 1280 - 9198 ]

Configuring the Static MTU

1280

276

[no] tunnel tos [ 0 - 63 | copy | use-qos ]

Configuring a Value for TOS

copy

276

[no] tunnel ttl [ 0 - 255 | copy ]

Configuring a Value for Time-to-Live (TTL)

n/a enabled n/a

CLI page reference 274

64 seconds 277

Overview IPv6 over IPv4 tunneling is a way to establish point-to-point tunnels by encapsulating IPv6 packets within IPv4 headers so that they can be carried over the IPv4 routing infrastructure. IPv6 over IPv4 tunneling provides a mechanism for utilizing the existing IPv4 routing infrastructure to carry IPv6 traffic between IPv6 networks. There are a number of IPv6 tunneling mechanisms. Currently only tunneling IPv6 traffic over an IPv4 network through 6in4 manually configured tunnel endpoints is supported. Tunnels are an additional routing interface type, similar to a VLAN interface or a loopback interface. Routing into 6in4 tunnels is supported for: •

A standard route table lookup



Static Routes



Policy Based Routing (PBR)



Running OFPFv3 over the point-to-point tunnel interface

See RFC 4213 Basic Transition Mechanisms for IPv6 Hosts and Routers for more information about tunneling. Tunneling can be used for: •

Router to router—IPv4 routers connected by an IPv4 infrastructure can tunnel IPv6 packets among themselves. The tunnel spans one segment of the end-to-end path.



Host to router—IPv4 and IPv6 hosts can tunnel IPv6 packets to an intermediary IPv6 or IPv4 router that is reachable through an IPv4 infrastructure. The tunnel spans the first segment of the end-to-end path.

Overview 273



Host to host—IPv6 or IPv4 hosts that are interconnected by an IPv4 infrastructure can tunnel IPv6 packets among themselves. The tunnel spans the entire end-to-end path.



Route to host— IPv6 or IPv4 routers can tunnel IPv6 packets to their final destination IPv6 or IPv4 host. This tunnel spans only the last segment of the end-to-end path.

Configured tunnels are usually used in the router-to-router configuration because the tunnel endpoints need to be explicitly configured. The tunnel endpoint includes: •

The entry node of the tunnel (the encapsulator), which creates an encapsulating IPv4 header and sends the encapsulated packet. Which packets to tunnel is determined by a routing table lookup based on the IPv6 address.



The exit node of the tunnel (the decapsulator):



receives the encapsulated packet



reassembles the packet if needed



removes the IPv4 encapsulating header



processes the IPv6 packet in the usual manner

The decapsulator matches received packets to the tunnels it has configured, and only processes packets where the IPv4 source and destination addresses match the endpoint addresses of the configured tunnels. A tunnel’s IPv4 address must be the same on both the encapsulator and the decapsulator. IPv4 routing switches route the packet based on the IPv4 header. IPv6 traffic can travel the tunnel in either direction. Each end node can be either the encapsulator or the decapsulator depending on the flow of the IPv6 traffic. Figure 27 Conceptual Example of a Tunnel

A tunnel is treated as a single point-to-point link; the encapsulator and decapsulator behave as IPv6 neighbors on that link. The encapsulator and decapsulator assign IPv6 link-local addresses to the interface and may also assign IPv6 global addresses. Neighbor discovery and duplicate address detection are implemented as they are on any other IPv6 interface.

Configuring a Tunnel Interface An IPv6 address is configured on the tunnel interface in the same way that it would be on other IP routing interfaces, such as VLANs. IPv4 addresses are configured as the tunnel source and tunnel destination endpoint addresses. 274

IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

To create a tunnel, enter this command in the global config context.

Syntax [no] interface tunnel 1 - 128 Creates a tunnel. Tunnel interface context is entered. The no form of the command removes the tunnel configuration. To enable or disable the tunnel, enter this command in tunnel context.

Syntax tunnel [ enable | disable ] Enables or disables the tunnel. The enable command only succeeds if all mandatory parameters such as source and destination addresses for the tunnel are configured. If disable is specified, the tunnel configuration is not removed. Default: Enabled To optionally configure a name for the tunnel, enter this command in tunnel context.

Syntax [no] tunnel name string Optional; Provides a name for the tunnel. The name must be unique for all existing tunnels. The no form of the command removes the name for the tunnel. Example 178 Creating, Enabling, and Naming a Tunnel HP Switch(config)# interface tunnel 3 HP Switch(tunnel-3)# tunnel enable HP Switch(tunnel-3)# tunnel name Redtunnel

Configuring the Tunnel Mode The tunnel mode configures the tunnel encapsulation type. The only mode currently supported is 6in4 mode.

Syntax [no] tunnel mode [ 6in4 | unspecified ] Configures the type of tunnel. 6in4 IPv6 packets encapsulated and transported over an IPv4 network. unspecified Set the tunnel to be unspecified. Example 179 Configuring Tunnel Mode HP Switch(tunnel-3)# tunnel mode 6in4

Configuring the Tunnel Source When encapsulating a packet, the source IP address is used in the encapsulating IPv4 header. When decapsulating a packet, this address is matched against the destination IP address in the encapsulating IPv4 header to determine if the packet was received on a valid, configured tunnel. The command is executed in tunnel context. Configuring a Tunnel Interface 275

Syntax [no] tunnel source [ ipv4-addr | ipv6-addr ] Configures the IPv4 or IPv6 address of the source (local) end of the tunnel. Must not be the same address as the tunnel destination. Tunnel mode must be configured before tunnel source. You cannot configure the same source and destination address pair on more than one tunnel interface.

Configuring the Tunnel Destination When an encapsulating packet is sent into a tunnel, the tunnel destination address is used in the encapsulating IPv4 header. When decapsulating a packet, this address is matched against the source IP address in the encapsulating IPv4 header to determine if the packet was received on a valid, configured tunnel. The command is executed in tunnel context.

Syntax [no] tunnel destination [ ipv4-addr | ipv6-addr ] Configures the IPv4 or IPv6 address of the remote end of the tunnel. Must not be the same address as the tunnel source. Tunnel mode must be configured before tunnel destination. Example 180 Configuring Destination and Source Addresses HP Switch(tunnel-3)# tunnel source 20.30.30.3 HP Switch(tunnel-3)# tunnel destination 10.20.20.2

Configuring the Static MTU Only the static tunnel MTU option is supported. Enter this command in tunnel context. NOTE:

The MTU value should be the same for the source and the destination switch.

Syntax [no] tunnel mtu [ 1280 - 9198 ] Configures the static MTU for the tunnel. Default: 1280 Example 181 Configuring a Static MTU HP Switch(tunnel-3)# tunnel mtu 1500

Configuring a Value for TOS To configure a value to use for the TOS byte in the encapsulating IPv4 header when encapsulating a packet to send over the tunnel interface, enter this command in tunnel context. If no value is specified, the value of the TOS byte is determined from the traffic class field of the IPv6 header.

Syntax [no] tunnel tos [ 0 - 63 | copy | use-qos ] Configures a value to use for the TOS byte. 0 - 63 Range of static values.

276 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

copy The TOS bits are copied from the IPv6 header. This is the default. use-qos Use the value returned by the QoS classifier. Example 182 Configuring a TTL for the Packet HP Switch(tunnel-3)# tunnel tos use-qos

Configuring a Value for Time-to-Live (TTL) Use this command to configure the TTL in the encapsulating IPv4 header when encapsulating a packet to send over the tunnel. Enter this command in tunnel context.

Syntax [no] tunnel ttl [ 0 - 255 | copy ] Configures the time-to-live. copy When specified, the value of the TTL field from the IPv6 header is used in the IPv4 header. Default : 64 seconds Example 183 Configuring a TTL for the Packet HP Switch(tunnel-3)# tunnel ttl 100

Example: Manual 6in4 Tunneling This example creates an IPv6 6in4 tunnel, which allows IPv6 hosts in one network to exchange IPv6 data with hosts in another IPv6 network by using the IPv4 tunneling infrastructure. In this example: •

All IPv6 traffic between hosts in the 3000::/64 and the 4000::/64 networks is tunneled through Switch B and Switch C, respectively.



Switch B and Switch C are the manually configured endpoints.



IPv6 traffic entering the tunnel endpoints is encapsulated with an IPv4 header based on the 6in4 configuration, and then routed through the IPv4 network using Switch A as the first hop.



Static IPv6 routes are used to route into the tunnel interfaces.

Figure 28 Example of Manual 6in4 Tunneling

Example: Manual 6in4 Tunneling 277

Configure Switch B 1.

Configure the Tunnel Endpoint for Switch B The tunnel endpoint for Switch B is configured with: •

Mode—6in4



Source—encapsulation information for the IPv4 header indicating the source IPv4 address



Destination—encapsulation information for the IPv4 header indicating the destination IPv4 address



IPv6 Address—IPv6 network assigned to the interface (much like VLAN assignments). Enables IPv6 on the interface. It is required in order to route over the interface.

Example 184 Configuring a Tunnel Endpoint HP HP HP HP HP

Switch(config)# interface tunnel 1 Switch(tunnel-1)# tunnel mode 6in4 Switch(tunnel-1)# tunnel source 20.0.0.1 Switch(tunnel-1)# tunnel destination 30.0.0.1 Switch(tunnel-1)# ipv6 address 5000::1/64

5000::1/64 IPv6 network assigned to the interface. Enables IPv6 on the interface. 2.

Configure VLAN 2 for the IPv6 Hosts on the 3000::/64 Network This step configures IPv6 address 3000::1/64 on VLAN 2. When IPv6 traffic from IPv6 host 3000::2/64 that is destined for the IPv6 4000::/64 network is transported over VLAN 2, it will be routed into the tunnel. The command is executed in VLAN context. Example 185 Configuring the VLAN for Routing Traffic into the Tunnel HP Switch(config)# vlan 2 HP Switch(vlan-2)# ipv6 address 3000::1/64

3.

Configure IPv6 Routing to Route into the Tunnel IPv6 unicast routing is enabled. The static route is configured to route packets destined for the 4000::/64 network into the tunnel. Example 186 Configuring IPv6 Routing to Route into Tunnel HP Switch(config)# ipv6 unicast routing HP Switch(config)# ipv6 route 4000::/64 1 1 2

4.

5000::2 2

Configures static route for destinations in the 4000::/64 network IPv6 address of tunnel endpoint

Configure the VLAN Used to Reach the First Hop Router in the IPv4 Network This step configures the tunnel endpoint IPv4 address 20.0.0.1/24 (VLAN 1) in VLAN 2 context so that traffic from VLAN 2 can reach the first hop router in the IPv4 network through the tunnel. The command is performed in VLAN context.

278 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

Example 187 Configuring the IPv4 Tunnel Endpoint Address HP Switch(vlan-2)# ip address 20.0.0.1/24

5.

Configure the IPv4 Default Gateway The IPv4 default gateway 20.0.0.2 is configured. Example 188 Configuring the IPv4 Gateway HP Switch(config)# ip default-gateway 20.0.0.2

Configure Switch C Configure Switch C in a manner similar to Switch B, using the appropriate IPv6 and IPv4 addresses. 1. Configure the Tunnel Endpoint for Switch C The tunnel endpoint for Switch C is configured with a mode, IPv4 source address, IPv4 destination address, and an IPv6 interface address. Example 189 Configuring a Tunnel Endpoint (opposite end) HP HP HP HP HP

2.

Switch(config)# interface tunnel 1 Switch(tunnel-1)# tunnel mode 6in4 Switch(tunnel-1)# tunnel source 30.0.0.1 Switch(tunnel-1)# tunnel destination 20.0.0.1 Switch(tunnel-1)# ipv6 address 5000::2/64

Configure VLAN 2 for the IPv6 Hosts on the 4000::/64 Network This step configures IPv6 address 4000::1/64 on VLAN 2. When IPv6 traffic from IPv6 host 4000::2/64 that is destined for the IPv6 3000::/64 network is transported over VLAN 2, it will be routed into the tunnel. This step is performed in VLAN context. Example 190 Configuring the VLAN for Routing Traffic into the Tunnel HP Switch(config)# vlan 2 HP Switch(vlan-2)# ipv6 address 4000::1/64

3.

Configure IPv6 Routing to Route into the Tunnel IPv6 unicast routing is enabled. The static route is configured to route packets destined for the 3000::/64 network into the tunnel. Example 191 Configuring IPv6 Routing to Route Into Tunnel HP Switch(config)# ipv6 unicast routing HP Switch(config)# ipv6 route 3000::/64 5000::1

3000::/64 Configures static route for destinations in the 3000::/64 network 5000::1 IPv6 address of tunnel endpoint

Example: Manual 6in4 Tunneling 279

4.

Configure the VLAN Used to Reach the First Hop Router in the IPv4 Network This step configures the tunnel endpoint IPv4 address 30.0.0.1/24 (VLAN1) in VLAN 2 context so that traffic from VLAN 2 can reach the first hop router in the IPv4 network through the tunnel. The command is executed in VLAN context. Example 192 Configuring the IPv4 Tunnel Endpoint Address HP Switch(vlan-2)# ip address 30.0.0.1/24

5.

Configure the IPv4 Default Gateway The default gateway 30.0.0.2 is configured. Example 193 Configuring the IPv4 Default Gateway HP Switch(config)# ip default-gateway 30.0.0.2

Example: Tunneling Using Policy-Based Routing (PBR) The following example uses the configuration shown in Example 193 (page 280). The routing configuration uses PBR to route into the tunnel. The configuration steps are similar to the prior example, with the addition of the PBR configuration.

Configure Switch B 1.

Configure the Tunnel Endpoint for Switch B The tunnel endpoint for Switch B is configured with a mode, source and destination, and an IPv6 address. Example 194 Configuring a Tunnel Endpoint HP HP HP HP HP

2.

Switch(config)# interface tunnel 1 Switch(tunnel-1)# tunnel mode 6in4 Switch(tunnel-1)# tunnel source 20.0.0.1 Switch(tunnel-1)# tunnel destination 30.0.0.1 Switch(tunnel-1)# ipv6 address 5000::1/64

Configure VLAN 2 for the IPv6 Hosts on the 3000::/64 Network This step is performed in VLAN context. Example 195 Configuring the VLAN for Routing Traffic into the Tunnel HP Switch(config)# vlan 2 HP Switch(vlan-2)# ipv6 address 3000::1/64

3.

Configure the VLAN Used to Reach the First Hop Router in the IPv4 Network This command is performed in VLAN context.

280 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

Example 196 Configuring the IPv4 Tunnel Endpoint Address HP Switch(vlan-2)# ip address 20.0.0.1/24

4.

Configure the IPv4 Default Gateway and Enable Unicast Routing Example 197 Configuring the IPv4 Default Gateway and Enabling Unicast Routing HP Switch(config)# ip default-gateway 20.0.0.2 HP Switch(config)# ipv6 unicast routing

5.

Configure IPv6 PBR-Based Routing Execute these steps to configure the IPv6 PBR-based routing to route into the tunnel, and apply it to the inbound VLAN. Example 198 Configuring IPv6 PBR-Based Routing HP Switch(config)# class ipv6 PBR_Class HP Switch(config-class)# match ipv6 any 4000::/64 HP Switch(config-class)# exit HP HP HP HP

Switch(config)# policy pbr PBR_Policy Switch(policy-pbr)# class ipv6 PBR_Class Switch(policy-pbr)# action interface tunnel 1 Switch(policy-pbr)# exit

HP Switch(config)# vlan 2 HP Switch(vlan-2)# service-policy PBR_Policy in

Configure Switch C 1.

Configure the Tunnel Endpoint for Switch C The tunnel endpoint for Switch C is configured with a mode, source and destination, and an IPv6 address. Example 199 Configuring a Tunnel Endpoint (opposite end) HP HP HP HP HP

2.

Switch(config)# interface tunnel 1 Switch(tunnel-1)# tunnel mode 6in4 Switch(tunnel-1)# tunnel source 30.0.0.1 Switch(tunnel-1)# tunnel destination 20.0.0.1 Switch(tunnel-1)# ipv6 address 5000::2/64

Configure VLAN 2 for the IPv6 Hosts on the 4000::/64 Network This step is performed in VLAN context. Example 200 Configuring the VLAN for Routing Traffic into the Tunnel HP Switch(config)# vlan 2 HP Switch(vlan-2)# ipv6 address 4000::1/64

Example: Tunneling Using Policy-Based Routing (PBR)

281

3.

Configure the VLAN Used to Reach the First Hop Router in the IPv4 Network This step is performed in VLAN context. Example 201 Configuring the IPv4 Tunnel Endpoint Address HP Switch(vlan-2)# ip address 30.0.0.1/24

4.

Configure the IPv4 Default Gateway and Enable Unicast Routing Example 202 Configuring the IPv4 Default Gateway and Configuring Unicast Routing HP Switch(config)# ip default-gateway 30.0.0.2 HP Switch(config)# ipv6 unicast routing

5.

Configure IPv6 PBR-Based Routing Execute these steps to configure the IPv6 PBR-based routing to route into the tunnel, and apply it to the inbound VLAN. Example 203 Configuring IPv6 PBR-based Routing HP Switch(config)# class ipv6 PBR_Class HP Switch(config-class)# match ipv6 any 3000::/64 HP Switch(config-class)# exit HP HP HP HP

Switch(config)# policy pbr PBR_Policy Switch(policy-pbr)# class ipv6 PBR_Class Switch(policy-pbr)# action interface tunnel 1 Switch(policy-pbr)# exit

HP Switch(config)# vlan 2 HP Switch(vlan-2)# service-policy PBR_Policy in

Displaying Tunnel Configuration and Status Information The show interface tunnel list command displays information about the configuration and status of the specified tunnels.

282 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

Example 204 Tunnel Configuration and Status Information for Multiple Tunnels HP Switch(config)# show interface tunnel 3,4 Tunnel Configuration : Tunnel Tunnel Name Tunnel Status Source Address Destination Address Mode TOS TTL

: : : : : : : :

tunnel-3 Redtunnel Enabled 120.22.33.44 121.23.34.2 6in4 5 IPv6 100 MTU

: Enabled : 1282

Current Tunnel Status : Tunnel State Destination Address Route Next Hop IP Next Hop Interface Next Hop IP Link Status Source Address IP Datagrams Received IP Datagrams Transmitted

: : : : : : : :

Down 0.0.0.0/0 15.255.128.1 vlan-1 Up Not Configured 0 0

Tunnel Configuration : Tunnel Tunnel Name Tunnel Status Source Address Destination Address Mode TOS TTL

: : : : : : : :

tunnel-4 Blue_tunnel Enabled 20.30.30.4 10.20.20.3 6in4 10 64

Current Tunnel Status : Tunnel State Destination Address Route Next Hop IP Next Hop Interface Next Hop IP Link Status Source Address

: : : : : :

IPv6 MTU

: Disabled : 1280

Down 0.0.0.0/0 15.255.128.1 vlan-1 Up Not Configured

Show Interface Tunnel Brief Example 205 Brief Output for Tunnel Configuration Information HP Switch(config)# show interface tunnel brief Status - Tunnel Information Brief Tunnel : tunnel-3 Source Address : 120.22.33.44 Destination Address : 121.23.34.2 Configured Tunnel Status : Enabled

Mode

: 6in4

Current Tunnel Status : Down

Show IPv6 ND RA Prefix for a Specific Tunnel Displaying Tunnel Configuration and Status Information 283

This command displays IPv6 neighbor discovery prefix information for the specified tunnel.

284 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

Example 206 IPv6 Neighbor Discovery Prefix Information for a Tunnel HP Switch(tunnel-3)# show ipv6 nd ra prefix tunnel 3 IPv6 Neighbor Discovery Prefix Information Tunnel Name IPv6 Prefix Valid Lifetime Preferred Lifetime On-link Flag Autonomous Flag Advertise Flag

: : : : : : :

Tunnel3 Default 15 days 14 days On On On

Show IP Counters for a Tunnel

Displaying Tunnel Configuration and Status Information 285

Example 207 Output Showing Counters for a Tunnel HP Switch(config)# show ip counters tunnel 3 Address Family : IPv4 Interface : Tunnel 3 IP In Datagrams Received IP In Octets Received IP In Datagrams Broadcast Received IP In Octets Broadcast Received IP In Datagrams Multicast Received IP In Octets Multicast Received IP In Datagrams Discarded Datagram Header Error IP In Datagrams Discarded No Route IP In Datagrams Discarded Invalid Address IP In Datagrams Discarded Unknown Protocol IP In Datagrams Discarded Truncation IP In Datagrams Discarded Processing Error IP In Datagrams Forwarding Required IP In Datagrams Delivery to Protocols Successful IP Datagrams Reassembly Required IP Datagrams Reassembly Successful IP Datagrams Reassembly Failed IP Out Datagrams Transmitted IP Out Octets Transmitted IP Out Datagrams Broadcast Transmitted IP Out Octets Broadcast Transmitted IP Out Datagrams Multicast Transmitted IP Out Octets Multicast Transmitted IP Out Datagrams Discarded Processing Error IP Out Datagrams Forwarded IP Out Datagrams Transmit Requests from Protocols IP Out Datagrams Fragmentation Required IP Out Datagrams Fragmentation Successful IP Out Datagrams Fragmentation Failed IP Out Datagrams Fragments Created IP In Datagrams Received IP In Octets Received IP In Datagrams Broadcast Received IP In Octets Broadcast Received IP In Datagrams Multicast Received IP In Octets Multicast Received IP In Datagrams Discarded Datagram Header Error IP In Datagrams Discarded No Route IP In Datagrams Discarded Invalid Address IP In Datagrams Discarded Unknown Protocol IP In Datagrams Discarded Truncation IP In Datagrams Discarded Processing Error IP In Datagrams Forwarding Required IP In Datagrams Delivery to Protocols Successful IP Datagrams Reassembly Required IP Datagrams Reassembly Successful IP Datagrams Reassembly Failed IP Out Datagrams Transmitted IP Out Octets Transmitted IP Out Datagrams Broadcast Transmitted IP Out Octets Broadcast Transmitted IP Out Datagrams Multicast Transmitted IP Out Octets Multicast Transmitted IP Out Datagrams Discarded Processing Error IP Out Datagrams Forwarded IP Out Datagrams Transmit Requests from Protocols IP Out Datagrams Fragmentation Required IP Out Datagrams Fragmentation Successful 286 IPv6 Tunneling Over IPv4 Using Manually Configured Tunnels

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0

IP Out Datagrams Fragmentation Failed IP Out Datagrams Fragments Created

: 0 : 0

Displaying Tunnel Configuration and Status Information 287

12 IPv6 Diagnostic and Troubleshooting Introduction Table 30 Summary of Commands Command syntax

Description

ipv6 icmp error-interval 0 - 2147483647 [ bucket-size 1 - 200 ] no ipv6 icmp error-interval

ICMP Rate-Limiting

ping6 [ ipv6-address | hostname | switch-number ]

Ping for IPv6 – Ping6

Default

100 10 n/a

CLI page reference 289

290

ping6 [ link-local-address %vlan vid | hostname | switch-number ] [ [ [ [ [ [

repetitions 1 - 10000 ] timeout 1 - 60 ] data-size 0 - 65507 ] data-fill 0 - 1024 ] source [ ipv6-addr | vid ]] oobm ]

1 1 second 0 0 n/a

traceroute6 [ ipv6-address | hostname ] [ [ [ [ [ -

Traceroute for IPv6

minttl 1 - 255 ] maxttl 1 - 255 ] timeout 1 - 120 ] probes 1 - 5 ] source ipv6-addr | vid | loopback 0 7 | oobm ]

291 1 30 5 seconds 3 n/a

[ dstport 1 - 34000 ] [ srcport 1 - 34000 ]

n/a n/a

traceroute6 [ link-local-address %vlan vid | hostname ] [ [ [ [ [

minttl 1 - 255 ] maxttl 1 - 255 ] timeout 1 - 120 ] probes 1 - 5 ] source ipv6-addr | vid | oobm

1 30 5 seconds 3 n/a

]

[no] ip dns server-address priority 1 - 3 ip-addr [ oobm ] [no] ip dns domain-name domain-name-suffix

DNS Configuration

n/a

293

[no] debug ipv6 [ debug-type ]

Debug Command

n/a

296

[no] logging syslog-ipv4-address

Logging Command

n/a

298

The IPv6 ICMP feature enables control over the error and informational message rate for IPv6 traffic, which can help mitigate the effects of a Denialof- service attack. Ping6 enables verification of access to a specific IPv6 device, and traceroute6 enables tracing the route to an IPv6-enabled device on the network.

288 IPv6 Diagnostic and Troubleshooting

ICMP Rate-Limiting ICMP rate-limiting controls the rate at which ICMPv6 generates error and informational messages for features such as: •

neighbor solicitations



neighbor advertisements



multicast listener discovery (MLD)



path MTU discovery (PMTU)



duplicate address discovery (DAD)



neighbor unreachability detection (NUD)



router discovery



neighbor discovery (NDP)

ICMPv6 error message generation is enabled by default. The rate of message generation can be adjusted, or message generation can be disabled. Controlling the frequency of ICMPv6 error messages can help to prevent DoS (Denial- of- Service) attacks. With IPv6 enabled on the switch, you can control the allowable frequency of these messages with ICMPv6 rate-limiting.

Syntax ipv6 icmp error-interval 0 - 2147483647 [ bucket-size 1 - 200 ] no ipv6 icmp error-interval This command is executed from the global configuration level, and uses a “token bucket” method for limiting the rate of ICMP error and informational messages. Using this method, each ICMP message uses one token, and a message can be sent only if there is a token available. In the default configuration, a new token can be added every 100 milliseconds, and a maximum of 10 tokens are allowed in the token bucket. If the token bucket is full, a new token cannot be added until an existing token is used to enable sending an ICMP message. You can increase or decrease both the the frequency with which used tokens can be replaced and (optionally) the number of tokens allowed to exist. error-interval Specifies the time interval in milliseconds between successive token adds. Increasing this value decreases the rate at which tokens can be added. A setting of 0 disables ICMP messaging. Default : 100; Range: 0 - 2147483647 bucket-size This optional keyword specifies the maximum number of tokens allowed in the token bucket at any time. Decreasing this value decreases the maximum number of tokens that may be available at any time. Default : 10; Range: 1 - 200 You can change the rate at which ICMP messages are allowed by changing the error-interval with or without a corresponding change in the bucket-size. The no ipv6 icmp error-interval command resets both the error-interval and the bucket-size values to their defaults. Use the show run command to view the current ICMP error interval settings. For example, the following command limits ICMP error and informational messages to no more than 20 every 1 second: ICMP Rate-Limiting 289

Switch(config)# ipv6 icmp error-interval 1000000 bucket-size 20

Ping for IPv6 (Ping6) The Ping6 test is a point-to-point test that accepts an IPv6 address or IPv6 host name to see if an IPv6 switch is communicating properly with another device on the same or another IPv6 network. A ping test checks the path between the switch and another device by sending IP packets (ICMP Echo Requests). To use a ping6 command with an IPv6 host name or fully qualified domain names, see “DNS Resolver for IPv6” (page 293). You can issue single or multiple ping tests with varying repetitions and timeout periods to wait for a ping reply. Replies to each ping test are displayed on the console screen. To stop a ping test before it finishes, press [Ctrl] [C]. For more information about using a ping test, see the “Troubleshooting” appendix in the current Management and Configuration Guide for your switch.

Syntax ping6 [ ipv6-address | hostname | switch-number ] [ repetitions 1 - 10000 ] [ timeout 1 - 60 ] [ data-size 0 - 65507 ] [ data-fill 0 - 1024 ] [ source [ ipv6-addr | vid ]] [ oobm ]

Syntax ping6 [ link-local-address %vlan vid | hostname | switch-number ] [ repetitions 1 - 10000 ] [ timeout 1 - 60 ] [ data-size 0 - 65507 ] [ data-fill 0 - 1024 ] [ source [ ipv6-addr | vid ]] [ oobm ] Pings the specified IPv6 host by sending ICMP version 6 (ICMPv6) echo request packets to the specified host. ipv6-address IPv6 address of a destination host device. link-local-address %vlan vid IPv6 link-local address, where %vlan vid specifies the VLAN ID number. hostname Host name of an IPv6 host device configured on an IPv6 DNS server. switch-number Number of an IPv6-based switch that is a member of a switch stack (IPv6 subnet). Valid values: 1 - 16. oobm For switches that have a separate out-of-band management (OOBM) port, oobm specifies that the traffic originates from the out-of-band management port. repetitions 1 - 10000 Number of times that IPv6 ping packets are sent to the destination IPv6 host. Default: 1. timeout 1 - 60 Number of seconds within which a response is required from the destination host before the ping test times out. Valid values : 1 - 60. Default: 1 second. data-size 0 - 65471 Size of data (in bytes) to be sent in ping packets. 290 IPv6 Diagnostic and Troubleshooting

Valid values: 0 - 65471. Default: 0. data-fill 0 - 1024 Text string used as data in ping packets. Range: up to 1024 alphanumeric characters; Default: 0 source [ ipv6-addr | vid ] The IPv6 address of the pinging device or the VLAN-ID on which the ping is being sent. Default : 0 (no text is used). Example 208 IPv6 Ping Tests HP Switch# ping6 fe80::2:1%vlan10 fe80:0000:0000:0000:0000:0000:0002:0001 is alive, time = 975 ms HP Switch# ping6 2001:db8::a:1c:e3:3 repetitions 3 2001:0db8:0000:0000:000a:001c:00e3:0003 is alive, iteration 1, time = 15 ms 2001:0db8:0000:0000:000a:001c:00e3:0003 is alive, iteration 2, time = 15 ms 2001:0db8:0000:0000:000a:001c:00e3:0003 is alive, iteration 3, time = 15 ms 3 packets transmitted, 3 packets received, 0% packet loss round-trip (ms) min/avg/max = 15/15/15 HP Switch# ping6 2001:db8::214:c2ff:fe4c:e480 repetitions 3 timeout 2 2001:db8:0000:0000:0214:c2ff:fe4c:e480 is alive, iteration 1, time = 15 ms 2001:db8:0000:0000:0214:c2ff:fe4c:e480 is alive, iteration 2, time = 10 ms 2001:db8:0000:0000:0214:c2ff:fe4c:e480 is alive, iteration 3, time = 15 ms HP Switch# ping6 2001:db8::10 Request timed out.

Traceroute for IPv6 The traceroute6 command enables you to trace the route from a switch to a host device that is identified by an IPv6 address or IPv6 host name. In the command output, information on each (router) hop between the switch and the destination IPv6 address is displayed. To use a traceroute6 command with an IPv6 host name or fully qualified domain names, see “DNS Resolver for IPv6” (page 293). Note that each time you perform a traceroute operation, the traceroute command uses the default settings unless you enter different values with each instance of the command. Replies to each traceroute operation are displayed on the console screen. To stop a traceroute operation before it finishes, press [Ctrl] [C]. For more information about how to configure and use a traceroute operation, see the “Troubleshooting” appendix in the Management and Configuration Guide.

Syntax traceroute6 [ ipv6-address | hostname ] [ minttl 1 - 255 ] [ maxttl 1 255 ] [ timeout 1 - 120 ] [ probes 1 - 5 ] [source ipv6-addr | vid | loopback 0 - 7 | oobm ] [ dstport 1 - 34000 ] [ srcport 1 - 34000 ]

Syntax traceroute6 [ link-local-address %vlanvid | hostname ] [ minttl 1 - 255 ] [ maxttl 1 - 255 ] [ timeout 1 - 120 ] [ probes 1 - 5 ] [ source [ ipv6-addr | vid | oobm ]] Traceroute for IPv6 291

Lists the IPv6 address of each hop in the route to the specified destination host device with the time (in microseconds) required for a packet reply to be received from each next-hop device. ipv6-address IPv6 address of a destination host device. link-local-address %vlan vlan-id IPv6 link-local address, where %vlan vlan-id specifies the VLAN ID number. hostname Host name of an IPv6 host device configured on an IPv6 DNS server. oobm For switches that have a separate out-of-band management (OOBM) port, oobm specifies that the traffic originates from the out-of-band management port. minttl Minimum number of hops allowed for each probe packet sent along the route. Default : 1; Range : 1 - 255 •

If the minttl value is greater than the actual number of hops, the traceroute output displays only the hops equal to or greater than the configured minttl threshold value. The hops below the threshold value are not displayed.



If the minttl value is the same as the actual number of hops, only the final hop is displayed in the command output.



If the minttl value is less than the actual number of hops, all hops to the destination host are displayed.

maxttl Maximum number of hops allowed for each probe packet sent along the route. Valid values: 1 - 255. Default: 30 If the maxttl value is less than the actual number of hops required to reach the host, the traceroute output displays only the IPv6 addresses of the hops detected by the configured maxttl value. timeout Number of seconds within which a response is required from the IPv6 device at each hop in the route to the destination host before the traceroute operation times out. Default : 5 seconds; Range : 1 - 120 probes Number of times a traceroute is performed to locate the IPv6 device at any hop in the route to the specified host before the operation times out. Default : 3; Range: 1 - 5 source [ ipv6-addr | vid ] The source IPv6 address or VLAN of the traceroute device or the VLAN-ID on which the traceroute packet is being sent. dstport 1 - 34000 Destination port. srcport 1 - 34000 Source port.

292 IPv6 Diagnostic and Troubleshooting

Example 209 IPv6 Traceroute Probes HP Switch# traceroute6 2001:db8::10 traceroute to 2001:db8::10 1 hop min, 30 hops max, 5 sec. timeout, 3 probes 1 2001:db8::a:1c:e3:3 0 ms 0 ms 0 ms 2 2001:db8:0:7::5 7 ms 3 ms 0 ms 3 2001:db8::214:c2ff:fe4c:e480 0 ms 1 ms 0 ms 4 2001:db8::10 0 ms 1 ms 0 ms

First three hops Intermediate router hops with the time (in milliseconds) for the switch to receive a response from each of the three probes sent to each router. Last hop Destination IPv6 address HP Switch# traceroute6 2001:db8::10 maxttl 7 traceroute to fe80::1:2:3:4 1 hop min, 7 hops max, 5 sec. timeout, 3 probes 1 2001:db8::a:1c:e3:3 0 ms 0 ms 0 ms 2 2001:db8:0:7::5 0 ms 0 ms 0 ms 3 * 2001:db8::214:c2ff:fe4c:e480 * 4 * * * 5 * * * 6 * * * 7 * * *

At hop 3, the first and third probes timed out, but the second probe reached the router. Each timed-out probe is displayed with an asterisk (*). The four remaining probes within the configured seven-hop maximum (maxttl) also timed out without finding a next-hop router or the destination IPv6 address.

DNS Resolver for IPv6 The Domain Name System (DNS) resolver is designed for local network domains where it enables use of a host name or fully qualified domain name to support DNS-compatible commands from the switch. Beginning with software release K.13.01, DNS operation supports these features: •

dual-stack operation: IPv6 and IPv4 DNS resolution



DNS-compatible commands: ping, ping6, traceroute, and traceroute6



multiple, prioritized DNS servers (IPv4 and IPv6)

DNS Configuration Up to three DNS servers can be configured. The addresses must be prioritized, and can be for any combination of IPv4 and IPv6 DNS servers. NOTE: This section describes the commands for configuring DNS operation for IPv6 DNS applications. For further information and examples on using the DNS feature, see “DNS Resolver” in appendix, “Troubleshooting”, in the current Management and Configuration Guide for your switch.

Syntax [no] ip dns server-address priority 1 - 3 ip-addr [ oobm ] Used at the global config level to configure the address and priority of a DNS server. Allows for configuring up to three servers providing DNS service. (The servers must all be accessible to the switch.) The command allows both IPv4 and IPv6 servers in any combination and any order of priority.

DNS Resolver for IPv6 293

priority 1 - 3 Identifies the order in which the specified DNS server will be accessed by a DNS resolution attempt. A resolution attempt tries each configured DNS server address, in ascending order of priority, until the attempt is successful or all configured server options have been tried and failed. To change the priority of an existing server option, you must remove the option from the switch configuration and re-enter it with the new priority. If another server address is configured for the new priority, you must also remove that address from the configuration before re-assigning its priority to another address. The no form of the command removes the specified address from the server address list configured on the switch. ip-addr Specifies the address of an IPv6 or IPv4 DNS server. oobm For switches that have a separate out-of-band management (OOBM) port, this parameter specifies that communication with the DNS server goes through that OOBM port.

Syntax [no] ip dns domain-name domain-name-suffix Used at the global config level to configure the domain suffix that is automatically appended to the host name entered with a command supporting DNS operation. Configuring the domain suffix is optional if you plan to use fully qualified domain names in all cases instead of just entering host names. You can configure up to three addresses for DNS servers in the same or different domains. However, you can configure only one domain name suffix. This means that a fully qualified domain name must be used to resolve addresses for hosts that do not reside in the same domain as the one you configure with this command. That is, if the domain name suffix and the address of a DNS server for that same domain are both configured on the switch, then you need to enter only the host name of the desired target when executing a command that supports DNS operation. But if the DNS server used to resolve the host name for the desired target is in a different domain than the domain configured with this command, then you need to enter the fully qualified domain name for the target. The no form of the command removes the configured domain name suffix. For example, suppose you want to configure the following on the switch: •

the address 2001:db8::127:10 which identifies a DNS server in the domain named mygroup.hpnetworking.net



a priority of 1 for the above server



the domain suffix mygroup.hpnetworking.net

Assume that the above, configured DNS server supports an IPv6 device having a host name of “mars-1” (and an IPv6 address of fe80::215:60ff:fe7a:adc0) in the “mygroup.hpnetworking.net” domain. In this case you can use the device's host name alone to ping the device because the mygroup.hpnetworking.net domain has been configured as the domain name on the switch and the address of a DNS server residing in that domain is also configured on the switch. The commands for these steps are as follows:

294 IPv6 Diagnostic and Troubleshooting

Example 210 Configuring for a Local DNS Server and Pinging a Registered Device HP Switch(config)# ip dns server priority 1 2001:db8::127:10 HP Switch(config)# ip dns domain-name mygroup.hpnetworking.net HP Switch(config)# ping6 mars-1 fe80::215:60ff:fe7a:adc0 is alive, time = 1 ms

However, for the same “mars-1” device, if mygroup.hpnetworking.net was not the configured domain name, you would have to use the fully qualified domain name for the device named mars-1: HP Switch# ping6 mars-1.mygroup.hpnetworking.net

For further information and examples on using the DNS feature, see “DNS Resolver” in the Troubleshooting appendix, in the current Management and Configuration Guide for your switch.

Viewing the Current Configuration Use the show ip dns command to view the current DNS server configuration. Use the show run command to view both the current DNS server addresses and the current DNS domain name in the active configuration.

Operating Notes In software release K.13.01, DNS addressing is not configurable from a DHCPv6 server.

Debug/Syslog for IPv6 The Debug/System logging (Syslog) for IPv6 feature provides logging functions similar to those of the IPv4 version, allowing you to record IPv4 and IPv6 Event Log and debug messages on a remote device to troubleshoot switch or network operation. For example, you can send messages about routing mis-configurations and other network protocol details to an external device, and later use them to debug network-level problems. NOTE: This section describes the commands for Debug/Syslog configuration in an IPv6 environment. For information on using the Debug/Syslog feature in an IPv4 environment, see “Debug/Syslog Operation” in the Troubleshooting appendix in the current Management and Configuration Guide for your switch.

Configuring Debug and Event Log Messaging To specify the types of debug and Event Log messages that you want to send to an external device: •

Use the debug ipv6... command to send messaging reports for the following types of switch events:



DHCPv6 client



DHCPv6 relay



forwarding



neighbor discovery

Debug/Syslog for IPv6 295





OSPFv3



packets

Use the logging [ severity severity-level | system-module system-module ] command to select a subset of Event Log messages to send to an external device for debugging purposes according to:



Severity level



System module

Debug Command Syntax [no] debug ipv6 [ debug-type ] Configures the types of IPv6 messages that are sent to Syslog servers or other configured debug destinations, where debug-type is any of the following event types: (none) all IPv6 events dhcpv6-client [ events | packets ] one of the following IPv6 client debug message types events DHCPv6 client events packets DHCPv6 client packets dhcpv6-relay [ events | packets ] one of the following IPv6 relay debug message types events DHCPv6 relay events packets DHCPv6 relay packets forwarding IPv6 forwarding events nd IPv6 neighbor discovery events ospf3 one of the following OSPFv3 message types: (none) all OSPFv3 debug events adj adjacency changes event events flood flooding

296 IPv6 Diagnostic and Troubleshooting

lsa-generation link state advertisement generation packet one of the following OSPFv3 packet types: (none) all OSPFv3 packets sent or received DD DD packets sent or received Hello Hello packets sent or received LSA LSA packets sent or received LSR LSR packets sent or received LSU LSU packets sent or received retransmission retransmissions spf SPF computations packet all IPv6 packet messages The no debug ipv6... form of the command stops the sending of debug messages of the specified type.

Configuring Debug Destinations An IPv6-based Debug/Syslog destination device can be a Syslog server (up to six maximum) and/or a console session: •

Use the debug destination [ logging | session | buffer ] command to enable (and disable) Syslog messaging on a Syslog server or to a CLI session for the debug message types configured with the debug and logging commands (see “Configuring Debug and Event Log Messaging” (page 295)). debug destination logging enables the configured debug message types to be sent to Syslog servers configured with the logging [ syslog-ipv4addr | syslog-ipv6-addr ] command. debug destination session enables the configured debug message types to be sent to the CLI session that executed this command. The session can be on any one terminal emulation device with serial, Telnet, or SSH access to the CLI at the Manager level prompt. debug destination buffer enables the configured debug message types to be sent to a buffer in switch memory.



Use the logging [ syslog-ipv6-addr ] command to configure the Syslog server at the specified IPv6 destination address.

Debug/Syslog for IPv6 297

Configuring an IPv6 Syslog Server Syslog for IPv6 is a client-server logging tool that allows a client switch to send event notification messages to n IPv6 networked device operating with Syslog server software. Messages sent to a Syslog server can be stored to a file for later debugging analysis. To use the Syslog for IPv6 feature, you must install and configure a Syslog server application on an IPv6 networked host accessible to the switch. See the documentation for the Syslog server application for instructions. To configure an IPv6 Syslog server, use the logging [ syslog-ipv6-addr ] command as described below. When you configure a Syslog server, Event Log messages are automatically enabled to be sent to the server. To reconfigure this setting, use the following commands: •

Use the debug command to specify additional debug message types.



Use the logging command to configure the system module or severity level used to filter the Event Log messages sent to configured Syslog servers. For more information, see “Configuring Debug and Event Log Messaging” (page 295).

Logging Command Syntax [no] logging syslog-ipv4-address Enables or disables Syslog messaging to the specified IPv4 address. You can configure up to six addresses. If you configure an address when none are already configured, this command enables destination logging (Syslog) and the Event debug type. Therefore, at a minimum, the switch begins sending Event Log messages to configured Syslog servers. If other debug message types are configured, they are also sent to the Syslog server. no logging Removes all currently configured Syslog logging destinations from the running configuration. no logging syslog-ipv4-address Removes only the specified Syslog logging destination from the running configuration. NOTE: The no logging command does not delete the Syslog server addresses stored in the startup configuration. To delete Syslog addresses in the startup configuration, you must enter the no logging command followed by the write memory command. To verify the deletion of a Syslog server address, display the startup configuration by entering the show config command. To block the messages sent to configured Syslog servers from the currently configured debug message type, enter the no debug [ debug-type ] command. To disable Syslog logging on the switch without deleting configured server addresses, enter the no debug destination logging command. For complete information on how to configure a Syslog server and Debug/ Syslog message reports, see the “Troubleshooting” appendix in the Management and Configuration Guide.

298 IPv6 Diagnostic and Troubleshooting

Displaying a Debug/Syslog for Configuration Use the show debug command to display the currently configured settings for: •

Debug message types and Event Log message filters (severity level and system module) sent to debug destinations



IPv4/IPv6 debug destinations (Syslog servers or CLI session) and Syslog server facility to be used

Example 211 Syslog Configuration to Receive Event Log Messages at Specified System Module and Severity Levels on an IPv6 Syslog Server Examples of show debug command output that displays a configured IPv6 Syslog server. HP Switch(config)# show debug Debug Logging Destination: None Enabled debug types: None are enabled

Displays the default debug configuration when no Syslog server IP addresses or debug types are configured.

HP Switch(config)# logging fe80:215:60ff:fe7a:adc0 HP Switch(config)# write memory HP Switch(config)# show debug Debug Logging Destination: Logging -fe80:215:60ff:fe7a:adc0 Facility=user Severity=debug System module=all-pass Enabled debug types: event

When you configure a Syslog IPv6 address with the logging command, by default, the switch enables debug messaging to the Syslog address and the user facility on the Syslog server, and sends Event Log messages of all severity levels from all system modules.

Debug/Syslog for IPv6 299

IPv6 Glossary For IPv6 ACL terminology, see “IPv6 ACL Terminology” (page 95). DAD

Duplicate Address Detection (see “Duplicate address detection (DAD) for statically configured addresses” (page 31)).

Default Route

The route selected for forwarding traffic when no other route is satisfactory for the intended destination. The IPv6 default route is ::/0.

Device Identifier

The low-order bits in an IPv6 address that identify a specific device. For example, in the link-local address 2001:db8:a10:101:212:79ff:fe88:a100/64, the bits forming 212:79ff:fe88:a100 comprise the device identifier.

DHCPv6

Dynamic Host Configuration Protocol, IPv6 version.

DoS

Denial-of-Service.

EUI-64

Extended Unique Identifier.

Global Address

Also known as global unicast address, these addresses are identified with the prefix format 001 (2000::/3). IPv6 global addresses are equivalent to IPv4 public addresses, and are whole routable and reachable in the IPv6 internet fragment.

Global Config Context

The manager configuration context indicated by (config)# in the CLI prompt and used for system-level configuration commands. See also “Manager Privileges” in the chapter titled “Using the Command Line Interface (CLI)” in the Basic Operation Guide.

IPv4

Internet Protocol version 4 32-bit address, which provides apr ox i mat ely 4 billion hosts.

IPv6

Internet Protocol version 6 128-bit address, which creates a larger address space.

IPv6 Host

A device that transmits and receives IPv6 traffic per-VLAN, but does not route traffic between VLANs.

Link-local Address

Link-local addresses are only valid on a single link or subnet. They always begin with the prefix “FE80::/10”.

Local Network

An interface on which a pair of nodes can communicate by switching within the same VLAN or subnet; i.e., for IPv6 nodes, communication requiring only link-local addresses. See also “Remote Network”. See also Remote Network.

Loopback Address

The loopback address is used by a node to send a packet to itself. The notation is ::1.

Manual Address Configuration

Configures an IPv6 address by using the CLI to manually enter a static address. Referred to as “Static Address Configuration” in this guide. See also Static Address Configuration.

MLD

Multicast Listener Discovery (see “Multicast Listener Discovery (MLD) Snooping” (page 71)).

MTU

Maximum Transmission Unit. The largest frame size allowed on a given path or device.

Multicast Address

Multicast addresses can be listened to by multiple nodes at one, even on the same link. They always begin with “FF00:/8”. The last 112 bits are the multicast group ID.

ND

Neighbor Discovery.

Null Route

A static route that drops traffic instead of forwarding it to a destination.

RA

Router Advertisement (see “Router Advertisements (RAs)” (page 181)).

Remote DHCPv6 Server

A DHCPv6 server that is on a different VLAN or subnet than a client needing DHCPv6 services.

Remote Network

An interface that can be reached by routing traffic between VLANs or subnets; i.e., for IPv6 nodes, communication requiring global unicast or multicast addresses. See also Local Network.

Routing Table

A list of routes to destinations the routing switch is aware of. Where multiple dynamic routes exist to the same destination, the routing table includes only the dynamic route determined to be the best choice. However, a static route normally supersedes a dynamic route to the same destination due to the low default administrative distance typically assigned to static routes.

300 IPv6 Glossary

SLAAC

Stateless Address Autoconfiguration.

Static Address

A permanently configured IPv6 address, as opposed to an autoconfigured address.

Static Address Configuration

Configures an IPv6 address by using the CLI to manually enter the address instead of using an automatically generated or DHCPv6-assigned address. Same as “Manual Address Configuration”. See also Manual Address Configuration.

Static Route

A static route uses a defined path and does not allow changes in the path to compensate for changes in routing conditions.

Unique link-local Address

“fd08::/8” is the unique local unicast address.

Unspecified Address

The unspecified address is used to indicate that an interface doesn’t yet have an address. The notation is “::”.

301

Index Symbols %vlan< vid > suffix;IPv6 link-local suffix;suffix, link-local address, 44, 46, 56 2740 RFC 2740, 259 802.1X ACL, IPv6, effect on, 102 port-based access not recommended, 102

A ABR definition, 261 OSPFv3, 261 ACL end, 120 filtering process, 111 rules, operation, 112 traffic not filtered, 111 VLANs, 112 ACL, IPv4 802.1X port-based not recommended, 102 deny any, implicit, IPv6, 101 dual-stack operation, 93 limit, 120, 161 monitoring, 156 RADIUS-assigned, limit, 120, 161 scalability, 120, 161 statistics counters RACL counter operation, 167 statistics counters, ACE, 156 ACL, IPv6 802.1X client limit, 102 802.1X port-based not recommended, 102 802.1X, effect on, 102 ACE after match not used, 111, 119 insert in list, 132 limit, 112 minimum number, 171 not used, 108 ACE, order in list see sequence, ACEs ACE:general rules, 120 AppleTalk, 111 application, 93 application methods, 110 application points, 106, 110 applications, 99, 103, 106, 107, 115, 159 assign to VLAN, 120 assigning, 115 assigning: to a VLAN, 130 assignment not deleted, 132 basic structure, 117 CIDR mask, 160 302 Index

command summary, 95 concurrent IPv4 and IPv6 ACLs, 99 configuration planning, 106 configure, 121 configured but not used, 120 configuring, 115 offline, 105 copy operation appends, 150 create, 94 defined, 93, 116 delete, 95 deleting an ACL, 172 deleting from config, 132 deny any any, implicit, supersede;supersede implicit deny any any, 117 deny any, implicit, 105, 107, 109, 110, 111, 120, 163 destination on the switch, 107 disable, 95 display, 95 ACLs and assignments, 164 assignments, 142, 143 configuration details, 141 content of an ACL, 145 data types, 149 summary, configured ACLs, 140 DSCP setting, 106 DSCP, ToS setting, 106 dual stack, 103 dual-stack operation, 93 duplicate sequence number, 121 dynamic, 93, 107, 160 dynamic port (RADIUS) ACL, 93 dynamic port ACL application, 101 dynamic port ACL operation defined, 99 dynamic port joins to a VLAN, 113 editing, 162 offline, 150 effect of replacing, 120 empty, 130 empty ACL, 129, 163 enable, 95 established, 116, 127 exit statement, 120 features, common to all, 105 filtering methods, 99 filtering process, 108 hit count see statistics, ACE ICMP traffic, 106 identifier, maximum length, 129, 130, 131 implicit deny see deny any, implicit interface assignment, options, 107 IPv6 routing required, 159 IPX, 111 length, prefix, 105

limit, 120, 161 log function, 105 log message see ACL, IPv6, logging logging, 105, 106 described, 164 session, 105 logging: notes, 170 mask CIDR, 160 match, always, 120 match, ignored, 111 maximum allowed, 112 mirroring, 99 monitoring, 156 multiple ACLs on interface, 103 multiple applications, 103 multiple lists on an interface, 104 multiple on same interface, 103 name, maximum length, 117, 129, 130, 131 non-IP traffic, 111 nonexistent identifier, assign, 120 number of entries, 105 offline editing, 150 operator, comparison;, 126 operator,comparison, 126 packet screened by multiple lists, 104 permit with multiple ACLs, 103 permit any forwarding, 111 permit/deny options, 116 permit/deny options, defined, 116 planning, 106 port ACL, 159 operation defined, 99 port ACL, IPv6, 159 see also static port ACL and dynamic port ACL port-based 802.1X, 102 port-based security, 102, 103 ports affected, 113 precedence DSCP, 106 prefix length, 105 protocol options, 116 purpose , 94 RACL RACL applications, 99 routing requirement, 107 screening switched traffic, 105 RACL:operation defined, 99 RADIUS-assigned, 93, 101, 107, 160 RADIUS-assigned ACL implicit deny IPv6, 101 multiple clients connected, 101 RADIUS-assigned, IPv6 denied traffic, 101 RADIUS-assigned, limit, 120, 161 remark remove from an ACE, 138 removing from a VLAN, 130

replacing, 112 replacing active ACEs, 120 resequence, 95 resource monitor, 171 rules, configuration, 112 SA or DA on the switch, 113 scalability, 120, 161 security use, 94 sequence number, 94, 163 use to delete ACE, 134 use to insert ACE, 132 sequence number, duplicate, 121 sequence number:out-of-range, 133 static port, 159 static port ACL, 95 static VLAN requirement, 107, 112, 113 static, defined, 159 statistics counters ACE, 156 structure, 117 switch-generated traffic, outbound, 170 switched packets, 113 Syslog see ACL, logging TCP control bits, 116, 160 TCP, established, 116 traffic generated by the switch, 107 traffic to/from the switch, 113 traffic types filtered, 94, 110 troubleshooting, 156 trunk adding port, 113 type, 119, 139, 142, 143, 144, 146 user-based 802.1X, 102 user-based security, 102, 103 VACL defined, 159 operation defined, 99 VACL applications, 100 VACL:configure , 95 VLAN ACL, IPv6 see VACL where applied to traffic, 107 ACL, IPv6:RACL configure, 95 ACL,IPv6 policies, 109 security use, 111 static port ACL application, 101 address configuration IPv6 global unicast, 21 IPv6 link-local, 20 IPv6 link-local autoconfiguration;VLAN:IPv6 link-local address autoconfiguration, 15 administrative distance OSPFv3, 270 advertisement OSPFv3, 233, 259 all-nodes, used in IPv6 DAD, 33 area, OSPFv3 configuring, 221 303

definition, 263 authorized IP managers access privilege, 58 access-method, 58 binary expressions of hexadecimal blocks binary expressions of IPv6 address, 66, 69 configuration command, 58 configuration examples, 67 configuring access privilege, 65 displaying configuration, 59 feature description, 65 IP mask used to configure single station, 58 IP masks used to configure multiple stations, 59, 66 precedence among security settings, 65 using IP masks, 58, 66 autoconfigured address effect of static address;static address configuration:effect of autoconfig, 22 autoconfigured unicast address DHCPv6 precedence, 20 Autonomous system OSPFv3, 263 autorun TFTP download of key file;TFTP:downloading key file, 49 TFTP download of trusted certificate;TFTP:downloading trusted certificate, 49

C command file TFTP download and running command script;TFTP: downloading command, 49 command output TFTP upload on remote device;TFTP:uploading command output, 50 configuration IPv6 routing parameters, 183 static IPv6 routes, 187 configuration file TFTP download;TFTP:downloading configuration file, 50 TFTP upload on remote device;TFTP:uploading configuration file, 50 configuring advertisement, 221 copy TFTP transfers, 48 crash data file TFTP upload on remote device;TFTP:uploading crash data file, 50 crash log TFTP upload on remote device;TFTP:uploading crash log, 50

D DAD configuration;neighbor solicitations:used in duplicate address detection, 33

304 Index

detecting duplicate unicast addresses on an interface, 14, 16, 18, 21, 31 performed on all IPv6 unicast addresses, 34 DCHPv6 client authentication, 31 default settings area, stub, metric-type, type2, 223 ip router-id, disabled, 220 IPv6:access-list resequence interval, 10, 135 IPv6:DAD, enabled, 23 IPv6:IPv6 operation, disabled, 13 IPv6:MLD default mode, auto, 85 IPv6:MLD, disabled, 72 IPv6:nd dad-attempts, 3 (enabled), 34 IPv6:nd ns-interval, 1000 ms, 34 IPv6:nd reachable-time, 3000 ms, 34 loopback interface, le0, 31 range, not set, 229 restart strict-lsa, enabled, 232 route-map, none, 221 router ospf3, disable, 221 router ospf3, disabled, 221 default VLAN IPv6 management interface, 180 IPv6 routing interface, 180 deprecated address;preferred lifetime;preferred address, 24 DHCPv6 configuring service requirements, 195 does not assign link-local address, 17 mutually exclusive with autoconfigured global unicast address, 16 mutually exclusive with static global unicast address, 20 precedence over autoconfig address, 20 server-assigned global unicast address;address configuration:IPv6 global unicast using DHCPv6;, 17 supported with DHCPv4 on same VLAN, 20 DHCPv6 Relay enabling;, 211 helper addresses, 215 minimum requirements;DHCPv6 Relay:enabling, 215 packet forwarding, 215 DHCPv6 Relay: configuring;, 211 DNS Configuration RA options, 35 dual-stack operation, 93

E ECMP feature description, 266 election DR (designated router), 261 EUI in IPv6 address autoconfiguration;MAC address:in IPv6 interface identifier, 16, 21

used in IPv6 address autoconfiguration;MAC address:used in IPv6 interface identifier, 15 event log TFTP upload on remote device;TFTP:uploading event log, 50

F fast leave MLD configuration, 76, 77 used in MLD snooping, 87 FE80 link-local address prefix, 15

G global unicast address autoconfigured is mutually exclusive with DHCP server-assigned address, 16 deprecation;anycast address:deprecation, 30 preferred lifetime;preferred lifetime:of global unicast address, 16, 18, 21 preferred lifetime;preferred lifetime:use of IPv6 address as source or destination;anycast address:preferred lifetime, 30 valid lifetime;valid lifetime:of global unicast address, 16, 18 valid lifetime;valid lifetime:use of deprecated IPv6 address as source or destination;anycast address:valid lifetime, 30

H hop limit IPv6 routing, 176

I IANA, protocol numbers, 122, 128 IP masks for multiple authorized manager stations, 59, 66 for single authorized manager station, 58 used in configuring authorized IP management, 58, 66 IP Preserve configuring;IPv6:IP Preserve, 54 DHCP-assigned address, 54 feature description, 54 IP routing changing router ID, 183 required for ACLs, 159 IPv6 configuration overview, 13 DHCPv6 server-assigned address, 13 DHCPv6 server-assigned address;VLAN:DHCPv6 server-assigned address;routing:DHCPv6 server-assigned address, 17 disabling, 32 displaying IPv6 configuration;VLAN:displaying IPv6 configuration, 27 displaying IPv6 routing table;routing:displaying IPv6 routing table, 28 displaying IPv6 routing table;routing:displaying IPv6 routing table;VLAN:displaying IPv6 routing

table;IPv6:show ipv6 routers output empty;routing:show ipv6 routers output empty, 29 enabling commands:displayed in IPv6 configuration, 27 global unicast address autoconfiguration;routing:IPv6 global unicast address autoconfiguration;global unicast address:autoconfiguration, 16 global unicast address manual configuration;global unicast address:manual configuration, 21 link-local address autoconfiguration;link-local address:autoconfiguration;, 15 link-local address manual configuration;link-local address:manual configuration;, 20 management address, 180 neighbor cache, clear;clear: neighbor cache, 56 neighbor cache, view;neighbor cache, view;, 39 neighbor discovery;neighbor discovery:IPv6 similar to IPv4 ARP, 32 routing between different VLANs;routing:IPv6 traffic between different VLANs, 35 routing interface, 180 routing table, 177 selecting default router on a VLAN;VLAN:selecting default IPv6 router;routing:selecting default IPv6 router, 37 Telnet6, view configuration;Telnet6:view configuration;, 44 IPv6 address binary expression, 66 ipv6 enable;IPv6 enabling commands, 14 IPv6 Router Advertisement configuration, 36 Global Configuration, 35 Interface Configuration, 36 RA options for DNS, 35 running configuration, 36 suppressing RDNSS, 36 suppressing SNSSL, 36 IPv6 routing administrative distance, 181 concurrent static and dynamic, 181 configuration :DHCPv6 service requirements, 195 configuration:displaying, 204 configuration:hop limit, 176 configuration:intervals between RA transmissions, 195 configuration:RA default router lifetime, 196 configuration:RA hop limit, 196 configuration:RAs on the same VLAN, 209 configuration:router advertisements, 208 enabling or disabling, 194 IPv6 static routes:administrative distance, 177, 187 IPv6 static routes:blackhole, 191 IPv6 static routes:configuring, 187 IPv6 static routes:default route, 184 IPv6 static routes:null interface, 189 IPv6 static routes:null route, 187 IPv6 static routes:one per destination, 189 IPv6 static routes:VLAN state, 191 305

loopback interface, 262 management interface, 180 minimum software release, 173 null routes, 187 parameter configuring, 183 router advertisement, enable, 194 router advertisement, enabled, 209 router advertisement:advertisement values, 208 router advertisement:basics, 208 router advertisement:default router lifetime, 196 router advertisement:DHCPv6 service requirements, 195 router advertisement:displaying configuration, 204 router advertisement:global context commands, 194 router advertisement:hop limit for host-generated packets, 196 router advertisement:intervals between RA transmissions, 195 router advertisement:setting up policies, 208 router advertisement:VLAN context neighbor discovery, 194 router ID, 261, 262 routing table, 177 routing table:adding routes, 187 static routes: administrative distance, 191 static routes:metric, 191 VLAN interface, 179 IPv6 routing:configuration RA global context commands, 194 VLAN context ND, 194 IPv6: address binary expression, 69

L link configuration, 271 loopback interface configuration;port:loopback interface configuration;IP address:loopback interface configuration, 22 displaying configuration;port:displaying loopback interface, 23 router priority, 262

M MAC address used in IPv6 link-local autoconfiguration;VLAN:link-local address autoconfiguration;, 15 management access address IPv6; address, management access:IPv6, 180 management interface, IPv6; IPv6 management interface, 180 MIB support:SNMP;SNMP:supported MIBs, 57 mirroring, 99 ACL, classifier-based, 105 ACL, deprecated, 105 MLD blocking multicast packet forwarding, 74, 86 displaying configuration, 77, 78 displaying statistics, 81, 82 306 Index

forwarding multicast packets, 73, 85 reducing multicast flooding, 72, 84 snooping at port level, 72 multicast MLD snooping reduces multicast flooding, 72, 84

O OSPF software license requirements, 174 OSPFv3 ABR, connection requirement, 221 ABR, range configuration, 268 administrative distance, 268, 270 area, 221, 263, 268 area configuration, 221 ASBR, 261, 268 ASBR, advertising, 265 autonomous system, 263 autonomous system boundary router, 261 backbone area, 263, 264 configuration steps, 220 cost, 268, 271 dead-interval, 268, 271 DR (designated router), 261 election, 261 enabling, 220, 221 external route cost options, 223 graceful restart helper, 268 hello-interval, 268, 271 Inter-Area-Prefix LSA default summary route, 223 Inter-Area-Prefix-LSA, 223 interface parameters, 271 loopback interface, 262, 268 loopback interface, redistribution, 229, 269 no-summary, 223 no-summary, effect, 265 normal area, 263 NSSA, 260, 264 parameters, 272 passive, 272 Premium License, 219 priority, 268, 271 range, blocking, 231 redistribution, loopback interface, 229, 269 retransmit-interval, 271 RFC 2740, 220 RFC 3101, 220 RFC compliance, 220 router ID, 251, 261, 262, 265, 268, 272 router ID, , 252 show passive information, 237 stub area, 263, 264 transit area, 271 transit area ID, 235, 236 transit-delay, 271 translator, NSSA, 260 Type-7-LSA default external route, 223 virtual, 271 virtual link, 221, 272

P parameters OSPFv3 interface, 271 virtual link, 272 port loopback interface configuration;loopback interface:default;loopback interface:multiple interfaces supported, 31 MLD snooping, 79 port ACL, 159 port monitoring, ACL, 99 port-level MLD snooping, 72, 74 Premium License OSPF, 174 OSPFv3, 258

R RADIUS dynamic port ACL, 101 RADIUS-assigned ACLs, 101 RADIUS-assigned, 101 see also dynamic port acl RADIUS-assigned ACLs, 101, 160 rate-limiting ACL, static, classifier-based, 105 ACL, static, deprecated, 105 resource monitor see Management and Configuration Guide retransmit interval, 233 RFC 2740 2740, 220 OSPFv3, 220, 259 RFC 3101 OSPF NSSA, 220 OSPFv3, 220, 259 RFC 6106, 35 RFCs RFC 2740, 220, 259 RFC 3101, 220 router ID, changing;configuration router ID, 183 routing IPv6 global unicast address autoconfiguration, 37 IPv6 Routing field, 26 routing switch, 174 running-config TFTP upload on remote device;TFTP:uploading running-config file, 50

S SCP/SFTP session limit, 70 security precedence of authorized IP manager settings, 65 sFlow;flow sampling;traffic monitoring sFlow, 57 show ipv6, 16, 17, 20, 22 show ipv6;IPv6

displaying IPv6 configuration, 23 show run IPv6 output, 27 SNMP configuring SNMPv1/v2c trap receiver, 52 configuring SNMPv3 management station, 52 displaying SNMPv3 management station configuration, 53 displaying trap configuration;traps:displaying configuration;notifications:displaying configuration, 52 features supported for IPv6;IPv6:SNMP support, 57 remote monitoring (RMON), 57 SNMPv1 and v2c traps;traps:supported in IPv6, 57 SNMPv2c informs;inform messages, 57 SNMPv3 notifications;notifications:supported in IPv6, 57 SNTP mode, 45 poll interval, 45 priority, 45 protocol version, 45 server address, 45 SNTP server, 47 software image TFTP download;TFTP: downloading software images, 50 TFTP upload on remote device;TFTP:uploading software image file, 50 solicited-node used in IPv6 neighbor discovery, 32 SSH filetransfer, 51 overview, 70 SSHv2 restriction, 64 version 1, 64 startup-config TFTP upload on remote device;TFTP:uploading startup-config file, 50 static ACL, 159 static IPv6 routes configuring, 187 Syslog see ACL, IPv6, logging

T tables IPv6 route, 177 Telnet viewing current use;IPv6:Telnet: view current use, 42 Telnet6 enable/disable inbound;IPv6:Telnet6: access, 43 Telnet6;outbound Telnet6, 41 TFTP auto-TFTP feature;auto-TFTP:downloading software images;TFTP:disabled;auto-TFTP:disabled, 51 downloading public-key file;public-key file:TFTP download, 50 downloading startup-config file;startup-config:TFTP download, 50 307

enabling client functionality, 48 enabling server functionality, 48 TFTP6 auto-TFTP;auto-TFTP:for IPv6, 51 enable client or server, 48 upload file to server;, 50 time sync mode, 45 Timepv6 manual configuration, 47 transit area OSPFv3, 271 tunnel configuring information, 220

V VACL defined, 159 VLAN displaying IPv6 configuration, 26 displaying MLD configuration, 77, 78, 79 displaying MLD statistics, 81, 82 global unicast address autoconfiguration;address configuration:IPv6 global unicast, 16 global unicast address manual configuration, 21 interface, 271 link-local address manual configuration, 20 MLD snooping, 71, 73, 74, 85 neighbor discovery operation, 32 VLAN interface Pv6:description, 179

308 Index

Loading...

HP Switch Software IPv6 Configuration Guide - HPE Support Center

HP Switch Software IPv6 Configuration Guide K/KA.15.14 nl Abstract This switch software guide is intended for network administrators and support pers...

3MB Sizes 0 Downloads 0 Views

Recommend Documents

HP OneView 2.0 User Guide - HPE Support Center
up-front planning considerations and how to use the HP OneView appliance UI or REST APIs to configure, manage, monitor,

HP ProLiant Servers Troubleshooting Guide - HPE Support Center
Mar 19, 2002 - o HP BladeSystem infrastructure error codes ("HP BladeSystem p-Class infrastructure error codes" on page

HP TP SMS User Interface Guide - HPE Support Center
server name) for packet capture (PCAP) files, which can then be viewed through the SMS client. ...... For inspection and

DECnet-Plus Planning Guide - HPE Support Center
Nov 2, 1996 - 9 DECdns Namespace Design Example. 9.1. Part 1: Planning ... 10–8. 10.6. DECdts Planning Worksheet . ...

Setup and Installation Guide - HPE Support Center
Feb 2, 2013 - HP BladeSystem c7000 Carrier-Grade. Enclosure. Setup and Installation Guide. Abstract. This document conta

HPE iLO 4 User Guide - HPE Support Center
This guide provides information about configuring, updating, and operating HPE ProLiant Gen8 and Gen9 servers and ......

CLI Reference Guide - HPE Support Center
Jun 2, 2014 - automatic hardware initialization and configuration, and its value resides in the node_config file. A comp

HP Printer Guide - software support systems
software and documentation, getting started guide, support flyer, Smart Install flyer. Warranty. One-year ... **Declared

Compatibility Matrix - HPE Support Center
File Access Auditing applications. Application ...... B.11.31.1603. AD300A2. 4Gb FC HBA. PCI-e. 5.03.15. B.11.31.1705. A

IPv6 Multicast Deployment and Configuration Guide
This field is roughly the equivalent of IPv4 Scoped Zones. However, in IPv6 this concept is accomplished by using the sc