Implementing IS-IS

Integrated Intermediate System-to-Intermediate System (IS-IS), Internet Protocol Version 4 (IPv4), is a standards-based Interior Gateway Protocol (IGP). The Cisco software implements the IP routing capabilities described in International Organization for Standardization (ISO)/International Engineering Consortium (IEC) 10589 and RFC 1195, and adds the standard extensions for single topology and multitopology IS-IS for IP Version 6 (IPv6).

This module describes how to implement IS-IS (IPv4 and IPv6) on your Cisco IOS XR network.

Enable IS-IS and Configure Level 1 or Level 2 Routing

This task explains how to enable IS-IS and configure the routing level for an area.


Note

Configuring the routing level in Step 4 is optional, but is highly recommended to establish the proper level of adjacencies.


Note

Users can configure the no max-metric command only with levels 1 or 2, that is, no max-metric level {1|2} in order to view the result in the output of the show configuration command. Else, the maximum metric configuration is not displayed in the output. This behavior is observed before committing the configuration to the router.

Before you begin

Although you can configure IS-IS before you configure an IP address, no IS-IS routing occurs until at least one IP address is configured.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • By default, all IS-IS instances are automatically Level 1 and Level 2. You can change the level of routing to be performed by a particular routing instance by using the is-type router configuration command.

Step 3

net network-entity-title

Example:


RP/0/RP0/CPU0:router(config-isis)# net 47.0004.004d.0001.0001.0c11.1110.00

Configures network entity titles (NETs) for the routing instance.

  • Specify a NET for each routing instance if you are configuring multi-instance IS-IS.

  • This example configures a router with area ID 47.0004.004d.0001 and system ID 0001.0c11.1110.00.

  • To specify more than one area address, specify additional NETs. Although the area address portion of the NET differs, the systemID portion of the NET must match exactly for all of the configured items.

Step 4

is-type { level-1 | level-1-2 | level-2-only }

Example:


RP/0/RP0/CPU0:router(config-isis)# is-type level-2-only

(Optional) Configures the system type (area or backbone router).

  • By default, every IS-IS instance acts as a level-1-2 router.

  • The level-1 keyword configures the software to perform Level 1 (intra-area) routing only. Only Level 1 adjacencies are established. The software learns about destinations inside its area only. Any packets containing destinations outside the area are sent to the nearest level-1-2 router in the area.

  • The level-2-only keyword configures the software to perform Level 2 (backbone) routing only, and the router establishes only Level 2 adjacencies, either with other Level 2-only routers or with level-1-2 routers.

  • The level-1-2 keyword configures the software to perform both Level 1 and Level 2 routing. Both Level 1 and Level 2 adjacencies are established. The router acts as a border router between the Level 2 backbone and its Level 1 area.

Step 5

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 6

show isis [ instance instance-id ] protocol

Example:


RP/0/RP0/CPU0:router# show isis protocol

(Optional) Displays summary information about the IS-IS instance.

Single-Topology IPv6

Single-topology IPv6 allows IS-IS for IPv6 to be configured on interfaces along with an IPv4 network protocol. All interfaces must be configured with the identical set of network protocols, and all routers in the IS-IS area (for Level 1 routing) or the domain (for Level 2 routing) must support the identical set of network layer protocols on all interfaces.

In single-topology mode, IPv6 topologies work with both narrow and wide metric styles in IPv4 unicast topology. During single-topology operation, one shortest path first (SPF) computation for each level is used to compute both IPv4 and IPv6 routes. Using a single SPF is possible because both IPv4 IS-IS and IPv6 IS-IS routing protocols share a common link topology.

Configure Single Topology for IS-IS

After an IS-IS instance is enabled, it must be configured to compute routes for a specific network topology.

This task explains how to configure the operation of the IS-IS protocol on an interface for an IPv4 or IPv6 topology.

Before you begin


Note

To enable the router to run in single-topology mode, configure each of the IS-IS interfaces with all of the address families enabled and “single-topology” in the address-family IPv6 unicast in the IS-IS router stanza. You can use either the IPv6 address family or both IPv4 and IPv6 address families, but your configuration must represent the set of all active address families on the router. Additionally, explicitly enable single-topology operation by configuring it in the IPv6 router address family submode.

Two exceptions to these instructions exist:

  1. If the address-family stanza in the IS-IS process contains the adjacency-check disable command, then an interface is not required to have the address family enabled.

  2. The single-topology command is not valid in the ipv4 address-family submode.

The default metric style for single topology is narrow metrics. However, you can use either wide metrics or narrow metrics. How to configure them depends on how single topology is configured. If both IPv4 and IPv6 are enabled and single topology is configured, the metric style is configured in the address-family ipv4 stanza. You may configure the metric style in the address-family ipv6 stanza, but it is ignored in this case. If only IPv6 is enabled and single topology is configured, then the metric style is configured in the address-family ipv6 stanza.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

interface type interface-path-id

Example:


RP/0/RP0/CPU0:router(config)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 3

Do one of the following:

  • ipv4 address address mask
  • ipv6 address ipv6-prefix / prefix-length [ eui-64 ]
  • ipv6 address ipv6-address { / prefix-length |  link-local }
  • ipv6 enable

Example:


RP/0/RP0/CPU0:router(config-if)# ipv4 address 10.0.1.3 255.255.255.0

or 

  RP/0/RP0/CPU0:router(config-if)# ipv6 address 3ffe:1234:c18:1::/64 eui-64
  RP/0/RP0/CPU0:router(config-if)# ipv6 address FE80::260:3EFF:FE11:6770 link-local
  RP/0/RP0/CPU0:router(config-if)# ipv6 enable
or

Defines the IPv4 address for the interface. An IP address is required on all interfaces in an area enabled for IS-IS if any one interface is configured for IS-IS routing.

or

Specifies an IPv6 network assigned to the interface and enables IPv6 processing on the interface with the eui-64 keyword.

or

Specifies an IPv6 address assigned to the interface and enables IPv6 processing on the interface with the link-local keyword.

or

Automatically configures an IPv6 link-local address on the interface while also enabling the interface for IPv6 processing.

  • The link-local address can be used only to communicate with nodes on the same link.

  • Specifying the ipv6 address ipv6-prefix / prefix-length interface configuration command without the eui-64 keyword configures site-local and global IPv6 addresses.

  • Specifying the ipv6 address ipv6-prefix / prefix-length command with the eui-64 keyword configures site-local and global IPv6 addresses with an interface ID in the low-order 64 bits of the IPv6 address. Only the 64-bit network prefix for the address needs to be specified; the last 64 bits are automatically computed from the interface ID.

  • Specifying the ipv6 address command with the link-local keyword configures a link-local address on the interface that is used instead of the link-local address that is automatically configured when IPv6 is enabled on the interface.

Step 4

exit

Example:


RP/0/RP0/CPU0:router(config-if)# exit

Exits interface configuration mode, and returns the router to mode.

Step 5

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • By default, all IS-IS instances are Level 1 and Level 2. You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 6

net network-entity-title

Example:


RP/0/RP0/CPU0:router(config-isis)# net 47.0004.004d.0001.0001.0c11.1110.00

Configures NETs for the routing instance.

  • Specify a NET for each routing instance if you are configuring multi-instance IS-IS. You can specify a name for a NET and for an address.

  • This example configures a router with area ID 47.0004.004d.0001 and system ID 0001.0c11.1110.00.

  • To specify more than one area address, specify additional NETs. Although the area address portion of the NET differs, the system ID portion of the NET must match exactly for all of the configured items.

Step 7

address-family ipv6 [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis)# address-family ipv6 unicast

Specifies the IPv6 address family and enters router address family configuration mode.

  • This example specifies the unicast IPv6 address family.

Step 8

single-topology

Example:


RP/0/RP0/CPU0:router(config-isis-af)# single-topology

(Optional) Configures the link topology for IPv4 when IPv6 is configured.

  • The single-topology command is valid only in IPv6 submode. The command instructs IPv6 to use the single topology rather than the default configuration of a separate topology in the multitopology mode.

Step 9

exit

Example:


RP/0/RP0/CPU0:router(config-isis-af)# exit

Exits router address family configuration mode, and returns the router to router configuration mode.

Step 10

interface type interface-path-id

Example:


RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 11

circuit-type { level-1 | level-1-2 | level-2-only }

Example:


RP/0/RP0/CPU0:router(config-isis-if)# circuit-type level-1-2

(Optional) Configures the type of adjacency.

  • The default circuit type is the configured system type (configured through the is-type command).

  • Typically, the circuit type must be configured when the router is configured as only level-1-2 and you want to constrain an interface to form only level-1 or level-2-only adjacencies.

Step 12

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters interface address family configuration mode.

  • This example specifies the unicast IPv4 address family on the interface.

Step 13

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 14

show isis [ instance instance-id ] interface [ type interface-path-id ] [ detail ] [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router# show isis interface HundredGigE 0/9/0/0

(Optional) Displays information about the IS-IS interface.

Step 15

show isis [ instance instance-id ] topology [ systemid system-id ] [ level { 1 | 2 }] [ summary ]

Example:


RP/0/RP0/CPU0:router# show isis topology

(Optional) Displays a list of connected routers in all areas.

Configuring Single-Topology IS-IS for IPv6: Example

The following example shows single-topology mode being enabled. An IS-IS instance is created, the NET is defined, IPv6 is configured along with IPv4 on an interface, and IPv4 link topology is used for IPv6. This configuration allows POS interface 0/3/0/0 to form adjacencies for both IPv4 and IPv6 addresses. 


  router isis isp
   net 49.0000.0000.0001.00
   address-family ipv6 unicast
    single-topology
   interface tenGigE 0/11/0/0
    address-family ipv4 unicast
    !
    address-family ipv6 unicast
    !
    exit
  !
  interface tenGigE 0/11/0/0
   ipv4 address 10.0.1.3 255.255.255.0
   ipv6 address 2001::1/64

Set SPF Interval for a Single-Topology Configuration

This task explains how to make adjustments to the SPF calculation to tune router performance. This task is optional.

Because the SPF calculation computes routes for a particular topology, the tuning attributes are located in the router address family configuration submode. SPF calculation computes routes for Level 1 and Level 2 separately.

When IPv4 and IPv6 address families are used in a single-topology mode, only a single SPF for the IPv4 topology exists. The IPv6 topology “borrows” the IPv4 topology; therefore, no SPF calculation is required for IPv6. To tune the SPF calculation parameters for single-topology mode, configure the address-family ipv4 unicast command.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


Router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type router configuration command.

Step 3

address-family { ipv4 | ipv6 } [ unicast ]

Example:


Router(config-isis)#address-family ipv4 unicast

Specifies the IPv4or IPv6 address family, and enters router address family configuration mode.

Step 4

spf-interval {[ initial-wait initial | secondary-wait secondary | maximum-wait maximum ] ...} [ level { 1 | 2 }]

Example:


Router(config-isis-af)# spf-interval maximum-wait 5000 initial-wait 50 secondary-wait 200

(Optional) Controls the minimum time between successive SPF calculations.

  • This value imposes a delay in the SPF computation after an event trigger and enforces a minimum elapsed time between SPF runs.

  • If this value is configured too low, the router can lose too many CPU resources when the network is unstable.

  • Configuring the value too high delays changes in the network topology that result in lost packets.

  • The SPF interval does not apply to the running of the ISPF because that algorithm runs immediately on receiving a changed LSP.

Step 5

ispf [ level { 1 | 2 }]

Example:


Router(config-isis-af)# ispf

(Optional) Configures incremental IS-IS ISPF to calculate network topology.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 7

show isis [ instance instance-id ] [[ ipv4 | ipv6 | afi-all ] [ unicast | safi-all ]] spf-log [ level { 1 | 2 }] [ fspf | prc | nhc ] [ detail | verbose ] [ last number | first number ]

Example:


Router# show isis instance 1 ipv4 spf-log

(Optional) Displays how often and why the router has run a full SPF calculation.

Customize Routes for IS-IS

This task explains how to perform route functions that include injecting default routes into your IS-IS routing domain and redistributing routes learned in another IS-IS instance. This task is optional.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing process, and places the router in router configuration mode.

  • By default, all IS-IS instances are automatically Level 1 and Level 2. You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 3

set-overload-bit [ on-startup { delay | wait-for-bgp }] [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# set-overload-bit

(Optional) Sets the overload bit.

Note

 

The configured overload bit behavior does not apply to NSF restarts because the NSF restart does not set the overload bit during restart.

Step 4

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis)# address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters router address family configuration mode.

Step 5

default-information originate [ route-policy route-policy-name ]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# default-information originate

(Optional) Injects a default IPv4 or IPv6 route into an IS-IS routing domain.

  • The route-policy keyword and route-policy-name argument specify the conditions under which the IPv4 or IPv6 default route is advertised.

  • If the route-policy keyword is omitted, then the IPv4 or IPv6 default route is unconditionally advertised at Level 2.

Step 6

redistribute isis instance [ level-1 | level-2 | level-1-2 ] [ metric metric ] [ metric-type { internal | external }] [ policy policy-name ]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# redistribute isis 2 level-1

(Optional) Redistributes routes from one IS-IS instance into another instance.

  • In this example, an IS-IS instance redistributes Level 1 routes from another IS-IS instance.

Step 7

Do one of the following:

  • summary-prefix address / prefix-length [ level { 1 | 2 }]
  • summary-prefix ipv6-prefix / prefix-length [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# summary-prefix 10.1.0.0/16 level 1

or 


RP/0/RP0/CPU0:router(config-isis-af)# summary-prefix 3003:xxxx::/24 level 1

(Optional) Allows a Level 1-2 router to summarize Level 1 IPv4 and IPv6 prefixes at Level 2, instead of advertising the Level 1 prefixes directly when the router advertises the summary.

  • This example specifies an IPv4 address and mask.

or

  • This example specifies an IPv6 prefix, and the command must be in the form documented in RFC 2373 in which the address is specified in hexadecimal using 16-bit values between colons.

  • Note that IPv6 prefixes must be configured only in the IPv6 router address family configuration submode, and IPv4 prefixes in the IPv4 router address family configuration submode.

Step 8

maximum-paths route-number

Example:


RP/0/RP0/CPU0:router(config-isis-af)# maximum-paths 16

(Optional) Configures the maximum number of parallel paths allowed in a routing table.

Step 9

distance weight [ address / prefix-length [ route-list-name ]]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# distance 90

(Optional) Defines the administrative distance assigned to routes discovered by the IS-IS protocol.

  • A different administrative distance may be applied for IPv4 and IPv6.

Step 10

set-attached-bit

Example:


RP/0/RP0/CPU0:router(config-isis-af)# set-attached-bit

(Optional) Configures an IS-IS instance with an attached bit in the Level 1 LSP.

Step 11

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Redistributing IS-IS Routes Between Multiple Instances: Example

The following example shows usage of the set- attached-bit and redistribute commands. Two instances, instance “1” restricted to Level 1 and instance “2” restricted to Level 2, are configured.

The Level 1 instance is propagating routes to the Level 2 instance using redistribution. Note that the administrative distance is explicitly configured higher on the Level 2 instance to ensure that Level 1 routes are preferred.

Attached bit is being set for the Level 1 instance since it is redistributing routes into the Level 2 instance. Therefore, instance “1” is a suitable candidate to get from the area to the backbone. 


  router isis 1
    is-type level-2-only
   net 49.0001.0001.0001.0001.00
   address-family ipv4 unicast
    distance 116
    redistribute isis 2 level 2
  !
  interface HundredGigE 0/9/0/0
   address-family ipv4 unicast
  !
  !
  router isis 2
   is-type level-1
   net 49.0002.0001.0001.0002.00
   address-family ipv4 unicast
    set
-attached-bit

  !
  interface HundredGigE 0/9/0/0
   address-family ipv4 unicast

Set Priority for Adding Prefixes to RIB

This optional task describes how to set the priority (order) for which specified prefixes are added to the RIB. The prefixes can be chosen using an access list (ACL), prefix list, or by matching a tag value.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing process, and places the router in router configuration mode. In this example, the IS-IS instance is called isp.

Step 3

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis)# address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters router address family configuration mode.

Step 4

metric-style wide [ transition ] [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# metric-style wide level 1

Configures a router to generate and accept only wide-link metrics in the Level 1 area.

Step 5

spf prefix-priority [ level { 1 | 2 }] { critical | high | medium } { access-list-name | tag tag }

Example:


RP/0/RP0/CPU0:router(config-isis-af)# spf prefix-priority high tag 3

Installs all routes tagged with the value 3 first.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

IS-IS Interfaces

IS-IS interfaces can be configured as one of the following types:

  • Active—advertises connected prefixes and forms adjacencies. This is the default for interfaces.

  • Passive—advertises connected prefixes but does not form adjacencies. The passive command is used to configure interfaces as passive. Passive interfaces should be used sparingly for important prefixes such as loopback addresses that need to be injected into the IS-IS domain. If many connected prefixes need to be advertised then the redistribution of connected routes with the appropriate policy should be used instead.

  • Suppressed—does not advertise connected prefixes but forms adjacencies. The suppress command is used to configure interfaces as suppressed.

  • Shutdown—does not advertise connected prefixes and does not form adjacencies. The shutdown command is used to disable interfaces without removing the IS-IS configuration.

Tag IS-IS Interface Routes

This optional task describes how to associate a tag with a connected route of an IS-IS interface.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing process, and places the router in router configuration mode. In this example, the IS-IS instance is called isp.

Step 3

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis)# address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters router address family configuration mode.

Step 4

metric-style wide [ transition ] [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# metric-style wide level 1

Configures a router to generate and accept only wide link metrics in the Level 1 area.

Step 5

exit

Example:


RP/0/RP0/CPU0:router(config-isis-af)# exit

Exits router address family configuration mode, and returns the router to router configuration mode.

Step 6

interface type number

Example:


RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 7

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters address family configuration mode.

Step 8

tag tag

Example:


RP/0/RP0/CPU0:router(config-isis-if-af)# tag 3

Sets the value of the tag to associate with the advertised connected route.

Step 9

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 10

show isis [ ipv4 | ipv6 | afi-all ] [ unicast | safi-all ] route [ detail ]

Example:


RP/0/RP0/CPU0:router# show isis ipv4 route detail

Displays tag information. Verify that all tags are present in the RIB.

Tagging Routes: Example

The following example shows how to tag routes. 


  route-policy isis-tag-55
  end-policy
  !
  route-policy isis-tag-555
    if destination in (5.5.5.0/24 eq 24) then
      set tag 555
      pass
    else
      drop
    endif
  end-policy
  !
  router static
   address-family ipv4 unicast
    0.0.0.0/0 2.6.0.1
    5.5.5.0/24 Null0
   !
  !
  router isis uut
   net 00.0000.0000.12a5.00
   address-family ipv4 unicast
    metric-style wide
    redistribute static level-1 route-policy isis-tag-555
    spf prefix-priority critical tag 13
    spf prefix-priority high tag 444
    spf prefix-priority medium tag 777

Limit LSP Flooding

Limiting link-state packets (LSP) may be desirable in certain “meshy” network topologies. An example of such a network might be a highly redundant one such as a fully meshed set of point-to-point links over a nonbroadcast multiaccess (NBMA) transport. In such networks, full LSP flooding can limit network scalability. One way to restrict the size of the flooding domain is to introduce hierarchy by using multiple Level 1 areas and a Level 2 area. However, two other techniques can be used instead of or with hierarchy: Block flooding on specific interfaces and configure mesh groups.

Both techniques operate by restricting the flooding of LSPs in some fashion. A direct consequence is that although scalability of the network is improved, the reliability of the network (in the face of failures) is reduced because a series of failures may prevent LSPs from being flooded throughout the network, even though links exist that would allow flooding if blocking or mesh groups had not restricted their use. In such a case, the link-state databases of different routers in the network may no longer be synchronized. Consequences such as persistent forwarding loops can ensue. For this reason, we recommend that blocking or mesh groups be used only if specifically required, and then only after careful network design.

Control LSP Flooding for IS-IS

Flooding of LSPs can limit network scalability. You can control LSP flooding by tuning your LSP database parameters on the router globally or on the interface. This task is optional.

Many of the commands to control LSP flooding contain an option to specify the level to which they apply. Without the option, the command applies to both levels. If an option is configured for one level, the other level continues to use the default value. To configure options for both levels, use the command twice. For example: 


  RP/0/RP0/CPU0:router(config-isis)# lsp-refresh-interval 1200 level 2
  RP/0/RP0/CPU0:router(config-isis)# lsp-refresh-interval 1100 level 1

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type router configuration command.

Step 3

lsp-refresh-interval seconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-refresh-interval 10800

(Optional) Sets the time between regeneration of LSPs that contain different sequence numbers

  • The refresh interval should always be set lower than the max-lsp-lifetime command.

Step 4

lsp-check-interval seconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-check-interval 240

(Optional) Configures the time between periodic checks of the entire database to validate the checksums of the LSPs in the database.

  • This operation is costly in terms of CPU and so should be configured to occur infrequently.

Step 5

lsp-gen-interval { [ initial-wait initial | secondary-wait secondary | maximum-wait maximum ] ... } [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-gen-interval maximum-wait 15 initial-wait 5 secondary-wait 5

(Optional) Reduces the rate of LSP generation during periods of instability in the network. Helps reduce the CPU load on the router and number of LSP transmissions to its IS-IS neighbors.

  • During prolonged periods of network instability, repeated recalculation of LSPs can cause an increased CPU load on the local router. Further, the flooding of these recalculated LSPs to the other Intermediate Systems in the network causes increased traffic and can result in other routers having to spend more time running route calculations.

Step 6

lsp-mtu bytes [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-mtu 1300

(Optional) Sets the maximum transmission unit (MTU) size of LSPs.

Step 7

max-lsp-lifetime seconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis)# max-lsp-lifetime 11000

(Optional) Sets the initial lifetime given to an LSP originated by the router.

  • This is the amount of time that the LSP persists in the database of a neighbor unless the LSP is regenerated or refreshed.

Step 8

ignore-lsp-errors disable

Example:


RP/0/RP0/CPU0:router(config-isis)# ignore-lsp-errors disable

(Optional) Sets the router to purge LSPs received with checksum errors.

Step 9

interface type interface-path-id

Example:


RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 10

lsp-interval milliseconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# lsp-interval 100

(Optional) Configures the amount of time between each LSP sent on an interface.

Step 11

csnp-interval seconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# csnp-interval 30 level 1

(Optional) Configures the interval at which periodic CSNP packets are sent on broadcast interfaces.

  • Sending more frequent CSNPs means that adjacent routers must work harder to receive them.

  • Sending less frequent CSNP means that differences in the adjacent routers may persist longer.

Step 12

retransmit-interval seconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# retransmit-interval 60

(Optional) Configures the amount of time that the sending router waits for an acknowledgment before it considers that the LSP was not received and subsequently resends. 


RP/0/RP0/CPU0:router(config-isis-if)# retransmit-interval 60

Step 13

retransmit-throttle-interval milliseconds [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-if)# retransmit-throttle-interval 1000

(Optional) Configures the amount of time between retransmissions on each LSP on a point-to-point interface.

  • This time is usually greater than or equal to the lsp-interval command time because the reason for lost LSPs may be that a neighboring router is busy. A longer interval gives the neighbor more time to receive transmissions.

Step 14

mesh-group { number | blocked }

Example:


RP/0/RP0/CPU0:router(config-isis-if)# mesh-group blocked

(Optional) Optimizes LSP flooding in NBMA networks with highly meshed, point-to-point topologies.

  • This command is appropriate only for an NBMA network with highly meshed, point-to-point topologies.

Step 15

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 16

show isis interface [ type interface-path-id | level { 1 | 2 }] [ brief ]

Example:


RP/0/RP0/CPU0:router# show isis interface HundredGigE 0/9/0/0 brief

(Optional) Displays information about the IS-IS interface.

Step 17

show isis [ instance instance-id ] database [ level { 1 | 2 }] [ detail | summary | verbose ] [ * | lsp-id ]

Example:


RP/0/RP0/CPU0:router# show isis database level 1

(Optional) Displays the IS-IS LSP database.

Step 18

show isis [ instance instance-id ] lsp-log [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router# show isis lsp-log

(Optional) Displays LSP log information.

Step 19

show isis database-log [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router# show isis database-log level 1

(Optional) Display IS-IS database log information.

IS-IS Authentication

Authentication is available to limit the establishment of adjacencies by using the hello-password command, and to limit the exchange of LSPs by using the lsp-password command.

IS-IS supports plain-text authentication, which does not provide security against unauthorized users. Plain-text authentication allows you to configure a password to prevent unauthorized networking devices from forming adjacencies with the router. The password is exchanged as plain text and is potentially visible to an agent able to view the IS-IS packets.

When an HMAC-MD5 password is configured, the password is never sent over the network and is instead used to calculate a cryptographic checksum to ensure the integrity of the exchanged data.

IS-IS stores a configured password using simple encryption. However, the plain-text form of the password is used in LSPs, sequence number protocols (SNPs), and hello packets, which would be visible to a process that can view IS-IS packets. The passwords can be entered in plain text (clear) or encrypted form.

To set the domain password, configure the lsp-password command for Level 2; to set the area password, configure the lsp-password command for Level 1.

The keychain feature allows IS-IS to reference configured keychains. IS-IS key chains enable hello and LSP keychain authentication. Keychains can be configured at the router level (in the case of the lsp-password command) and at the interface level (in the case of the hello-password command) within IS-IS. These commands reference the global keychain configuration and instruct the IS-IS protocol to obtain security parameters from the global set of configured keychains.

IS-IS is able to use the keychain to implement hitless key rollover for authentication. ey rollover specification is time based, and in the event of clock skew between the peers, the rollover process is impacted. The configurable tolerance specification allows for the accept window to be extended (before and after) by that margin. This accept window facilitates a hitless key rollover for applications (for example, routing and management protocols). 

Configure Authentication for IS-IS

This task explains how to configure authentication for IS-IS. This task is optional.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 3

lsp-password { hmac-md5 | text } { clear | encrypted } password [ level { 1 | 2 }] [ send-only ] [ snp send-only ]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-password hmac-md5 clear password1 level 1

Configures the LSP authentication password.

  • The hmac-md5 keyword specifies that the password is used in HMAC-MD5 authentication.

  • The text keyword specifies that the password uses cleartext password authentication.

  • The clear keyword specifies that the password is unencrypted when entered.

  • The encrypted keyword specifies that the password is encrypted using a two-way algorithm when entered.

  • The level 1 keyword sets a password for authentication in the area (in Level 1 LSPs and Level SNPs).

  • The level 2 keywords set a password for authentication in the backbone (the Level 2 area).

  • The send-only keyword adds authentication to LSP and sequence number protocol data units (SNPs) when they are sent. It does not authenticate received LSPs or SNPs.

  • The snp send-only keyword adds authentication to SNPs when they are sent. It does not authenticate received SNPs.

Note

 

To disable SNP password checking, the snp send-only keywords must be specified in the lsp-password command.

Step 4

interface type interface-path-id

Example:


RP/0/RP0/CPU0:router(config-isis)# interface GigabitEthernet

Enters interface configuration mode.

Step 5

hello-password { hmac-md5 | text } { clear | encrypted } password [ level { 1 | 2 }] [ send-only ]

Example:


RP/0/RP0/CPU0:router(config-isis-if)#hello-password text clear mypassword

Configures the authentication password for an IS-IS interface.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Configure Keychains for IS-IS

This task explains how to configure keychains for IS-IS. This task is optional.

Keychains can be configured at the router level ( lsp-password command) and at the interface level ( hello-password command) within IS-IS. These commands reference the global keychain configuration and instruct the IS-IS protocol to obtain security parameters from the global set of configured keychains. The router-level configuration (lsp-password command) sets the keychain to be used for all IS-IS LSPs generated by this router, as well as for all Sequence Number Protocol Data Units (SN PDUs). The keychain used for HELLO PDUs is set at the interface level, and may be set differently for each interface configured for IS-IS.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 3

l sp-password keychain keychain-name [ level { 1 | 2 }] [ send-only ] [ snp send-only ]

Example:


RP/0/RP0/CPU0:router(config-isis)# lsp-password keychain isis_a level 1

Configures the keychain.

Step 4

interface type interface-path-id

Example:


RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 5

hello-password keychain keychain-name [ level { 1 | 2 }] [ send-only ]

Example:


RP/0/RP0/CPU0:router(config-isis-if)#hello-password keychain isis_b

Configures the authentication password for an IS-IS interface.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Nonstop Forwarding

On Cisco IOS XR software, IS-IS NSF minimizes the amount of time a network is unavailable to its users following the restart of the IS-IS process.

When the IS-IS process restarts, all routing peers of that device usually detect that the device went down and then came back up. This transition results in what is called a routing flap, which could spread across multiple routing domains. Routing flaps caused by routing restarts create routing instabilities, which are detrimental to the overall network performance. NSF helps to suppress routing flaps, thus reducing network instability.

NSF allows for the forwarding of data packets to continue along known routes while the routing protocol information is being restored following the process restarts. When the NSF feature is configured, peer networking devices do not experience routing flaps. To preserve routing across RP failover events, NSR must be configured in addition to NSF.

When the Cisco IOS XR router running IS-IS routing performs the process restarts, the router must perform two tasks to resynchronize its link-state database with that of its IS-IS neighbors. First, it must relearn the available IS-IS neighbors on the network without causing a reset of the neighbor relationship. Second, it must reacquire the contents of the link-state database for the network.

The IS-IS NSF feature offers two options when configuring NSF:

  • IETF NSF

  • Cisco NSF

If neighbor routers on a network segment are NSF-aware, meaning that they are running a software version that supports RFC5306, they assist a router configured with nsf ietf command that is restarting. IETF NSF enables the neighbor routers provide adjacency and link-state information to help rebuild the routing information following a failover.

In Cisco IOS XR software, Cisco NSF checkpoints (stores persistently) all the state necessary to recover from a restart without requiring any special cooperation from neighboring routers. The state is recovered from the neighboring routers, but only using the standard features of the IS-IS routing protocol. This capability makes Cisco NSF suitable for use in networks in which other routers have not used the IETF standard implementation of NSF.


Note

If you configure IETF NSF on the Cisco IOS XR router and a neighbor router does not support IETF NSF, the affected adjacencies flap, but nonstop forwarding is maintained to all neighbors that do support IETF NSF. A restart reverts to a cold start if no neighbors support IETF NSF.


Note

Currently, a user can configure an aggressive hello-interval (lower than the default of 10 seconds for peer-to-peer session). But, if NSF is configured as a recovery for RP switchover, the default hello interval has to be used so that the sessions do not run into the risk of flapping during switchover.

Using LAN adjacencies in high availability (HA) scenarios is not recommended, since there is no designated intermediate system (DIS) redundancy in the protocol and traffic will either drop or be rerouted temporarily during DIS re-election.

Configure Nonstop Forwarding for IS-IS

This task explains how to configure your router with NSF that allows the software to resynchronize the IS-IS link-state database with its IS-IS neighbors after a process restart. The process restart could be due to an:

  • RP failover (for a warm restart)

  • Simple process restart (due to an IS-IS reload or other administrative request to restart the process)

  • IS-IS software upgrade

In all cases, NSF mitigates link flaps and loss of user sessions. This task is optional.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type router configuration command.

Step 3

nsf { cisco | ietf }

Example:


RP/0/RP0/CPU0:router(config-isis)# nsf ietf

Enables NSF on the next restart.

  • Enter the cisco keyword to run IS-IS in heterogeneous networks that might not have adjacent NSF-aware networking devices.

  • Enter the ietf keyword to enable IS-IS in homogeneous networks where all adjacent networking devices support IETF draft-based restartability.

Step 4

nsf interface-expires number

Example:


RP/0/RP0/CPU0:router(config-isis)# nsf interface-expires 1

Configures the number of resends of an acknowledged NSF-restart acknowledgment.

  • If the resend limit is reached during the NSF restart, the restart falls back to a cold restart.

Step 5

nsf interface-timer seconds

Example:


RP/0/RP0/CPU0:router(config-isis) nsf interface-timer 15

Configures the number of seconds to wait for each restart acknowledgment.

Step 6

nsf lifetime seconds

Example:


RP/0/RP0/CPU0:router(config-isis)# nsf lifetime 20

Configures the maximum route lifetime following an NSF restart.

  • This command should be configured to the length of time required to perform a full NSF restart because it is the amount of time that the Routing Information Base (RIB) retains the routes during the restart.

  • Setting this value too high results in stale routes.

  • Setting this value too low could result in routes purged too soon.

Step 7

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 8

show running-config [ command ]

Example:


RP/0/RP0/CPU0:router# show running-config router isis isp

(Optional) Displays the entire contents of the currently running configuration file or a subset of that file.

  • Verify that “nsf” appears in the IS-IS configuration of the NSF-aware device.

  • This example shows the contents of the configuration file for the “isp” instance only.

ISIS NSR

Non Stop Routing (NSR) suppresses IS-IS routing changes for devices with redundant route processors during processor switchover events (RP failover or ISSU), reducing network instability and downtime. When Non Stop Routing is used, switching from the active to standby RP have no impact on the other IS-IS routers in the network. All information needed to continue the routing protocol peering state is transferred to the standby processor prior to the switchover, so it can continue immediately upon a switchover.

To preserve routing across process restarts, NSF must be configured in addition to NSR.

Configuring ISIS-NSR

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis 1

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

Step 3

nsr

Example:


RP/0/RP0/CPU0:router(config-isis)# nsr

Configures the NSR feature.

Step 4

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 5

show isis nsr adjacency

Example:


RP/0/RP0/CPU0:router
# show isis nsr adjacency
System Id Interface SNPA State Hold Changed NSF IPv4 BFD IPv6 BFD 
  R1-v1S   Nii0     *PtoP* Up   83  00:00:33 Yes  None    None

Displays adjacency information.

Step 6

show isis nsr status

Example:


RP/0/RP0/CPU0:router
router#show isis nsr status
IS-IS test NSR(v1a) STATUS (HA Ready):
                            V1 Standby V2 Active V2 Standby
SYNC STATUS:                   TRUE      FALSE(0) FALSE(0)
PEER CHG COUNT:                 1           0        0
UP TIME:                     00:03:12     not up    not up

Displays the NSR status information.

Step 7

show isis nsr statistics

Example:


RP/0/RP0/CPU0:router
router#show isis nsr statistics
IS-IS test NSR(v1a) MANDATORY STATS :
                                 V1 Active                V1 Standby         V2 Active           V2 Standby
L1 ADJ:                               0                     0                   0                     0
L2 ADJ:                               2                     2                   0                     0
LIVE INTERFACE:                       4                     4                   0                     0
PTP INTERFACE:                        1                     1                   0                     0
LAN INTERFACE:                        2                     2                   0                     0
LOOPBACK INTERFACE:                   1                     1                   0                     0
TE Tunnel:                            1                     1                   0                     0
TE LINK:                              2                     2                   0                     0
NSR OPTIONAL STATS :
L1 LSP:                               0                     0                   0                     0
L2 LSP:                               4                     4                   0                     0
IPV4 ROUTES:                          3                     3                   0                     0
IPV6 ROUTES:                          4                     4                   0                     0

Shows number of ISIS adjacencies, lsps, routes, tunnels, Te links on active and standby routers.

Multiprotocol Label Switching Traffic Engineering

Table 1. Feature History Table

Feature Name

Release Information

Feature Description
MPLS TE Preference for Tunnels

Release 7.6.1

You can now configure the MPLS TE traffic for equal-cost multipath (ECMP) such that it flows only through TE tunnels. This is useful in scenarios where the hardware has resource constraints that limit the number of mixed ECMP routes.

In earlier releases, IS-IS installed multiple ECMPs for a route in the Routing Information Base (RIB) through TE tunnels and physical interfaces by default.

This feature introduces the following command:

mpls traffic-eng tunnel preferred

The MPLS TE feature enables an MPLS backbone to replicate and expand the traffic engineering capabilities of Layer 2 ATM and Frame Relay networks. MPLS is an integration of Layer 2 and Layer 3 technologies.

For IS-IS, MPLS TE automatically establishes and maintains MPLS TE label-switched paths across the backbone by using Resource Reservation Protocol (RSVP). The route that a label-switched path uses is determined by the label-switched paths resource requirements and network resources, such as bandwidth. Available resources are flooded by using special IS-IS TLV extensions in the IS-IS. The label-switched paths are explicit routes and are referred to as traffic engineering (TE) tunnels.

Configure MPLS Traffic Engineering for IS-IS

This task explains how to configure IS-IS for MPLS TE. This task is optional.

Before you begin

Your network must support the MPLS software feature before you enable MPLS TE for IS-IS on your router.


Note

Enter the commands in the following task list on every IS-IS router in the traffic-engineered portion of your network.


Note

MPLS traffic engineering currently does not support routing and signaling of LSPs over unnumbered IP links. Therefore, do not configure the feature over those links.

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:


RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type router configuration command.

Step 3

address-family { ipv4 | ipv6 } [ unicast ]

Example:


RP/0/RP0/CPU0:router(config-isis)#address-family ipv4 unicast

Specifies the IPv4 or IPv6 address family, and enters router address family configuration mode.

Step 4

mpls traffic-eng level { 1 | 2 }

Example:


RP/0/RP0/CPU0:router(config-isis-af)# mpls traffic-eng level 1

Configures a router running IS-IS to flood MPLS TE link information into the indicated IS-IS level.

Step 5

mpls traffic-eng router-id { ip-address | interface-name interface-instance }

Example:


RP/0/RP0/CPU0:router(config-isis-af)# mpls traffic-eng router-id loopback0

Specifies that the MPLS TE router identifier for the node is the given IP address or an IP address that is associated with the given interface.

Step 6

mpls traffic-eng tunnel preferred

Example:

RP/0/RP0/CPU0:router(config-isis-af)# mpls traffic-eng tunnel preferred

(optional) Prevents IS-IS from installing routes that use both MPLS TE tunnels and physical interfaces, and limits IS-IS to use only MPLS TE tunnels for ECMP.

Step 7

metric-style wide [ level { 1 | 2 }]

Example:


RP/0/RP0/CPU0:router(config-isis-af)# metric-style wide level 1

Configures a router to generate and accept only wide link metrics in the Level 1 area.

Step 8

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 9

show isis [ instance instance-id ] mpls traffic-eng tunnel

Example:


RP/0/RP0/CPU0:router# show isis instance isp mpls traffic-eng tunnel

(Optional) Displays MPLS TE tunnel information.

Step 10

show isis [ instance instance-id ] mpls traffic-eng adjacency-log

Example:


RP/0/RP0/CPU0:router# show isis instance isp mpls traffic-eng adjacency-log

(Optional) Displays a log of MPLS TE IS-IS adjacency changes.

Step 11

show isis [ instance instance-id ] mpls traffic-eng advertisements

Example:


RP/0/RP0/CPU0:router# show isis instance isp mpls traffic-eng advertisements

(Optional) Displays the latest flooded record from Cisco Multiprotocol Label Switching Traffic Engineering.

MPLS TE Forwarding Adjacency

MPLS TE forwarding adjacency allows a network administrator to handle a traffic engineering, label switch path (LSP) tunnel as a link in an Interior Gateway Protocol (IGP) network, based on the Shortest Path First (SPF) algorithm. A forwarding adjacency can be created between routers in the same IS-IS level. The routers can be located multiple hops from each other. As a result, a TE tunnel is advertised as a link in an IGP network, with the cost of the link associated with it. Routers outside of the TE domain see the TE tunnel and use it to compute the shortest path for routing traffic throughout the network.

MPLS TE forwarding adjacency is considered in IS-IS SPF only if a two-way connectivity check is achieved. This is possible if the forwarding adjacency is bidirectional or the head end and tail end routers of the MPLS TE tunnel are adjacent.

The MPLS TE forwarding adjacency feature is supported by IS-IS. For details on configuring MPLS TE forwarding adjacency, see the MPLS Configuration Guide.

Tune Adjacencies for IS-IS

This task explains how to enable logging of adjacency state changes, alter the timers for IS-IS adjacency packets, and display various aspects of adjacency state. Tuning your IS-IS adjacencies increases network stability when links are congested. This task is optional.

For point-to-point links, IS-IS sends only a single hello for Level 1 and Level 2, which means that the level modifiers are meaningless on point-to-point links. To modify hello parameters for a point-to-point interface, omit the specification of the level options.

The options configurable in the interface submode apply only to that interface. By default, the values are applied to both Level 1 and Level 2.

The hello-password command can be used to prevent adjacency formation with unauthorized or undesired routers. This ability is particularly useful on a LAN, where connections to routers with which you have no desire to establish adjacencies are commonly found.

Procedure

Step 1

configure

Example:

RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:

RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing instance, and places the router in router configuration mode.

  • You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 3

log adjacency changes

Example:

RP/0/RP0/CPU0:router(config-isis)# log adjacency changes

Generates a log message when an IS-IS adjacency changes state (up or down).

Step 4

interface type interface-path-id

Example:

RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 5

hello-padding { disable | sometimes } [ level { 1 | 2 }]

Example:

RP/0/RP0/CPU0:router(config-isis-if)# hello-padding sometimes

Configures padding on IS-IS hello PDUs for an IS-IS interface on the router.

  • Hello padding applies to only this interface and not to all interfaces.

Step 6

hello-interval seconds [ level { 1 | 2 }]

Example:

RP/0/RP0/CPU0:router(config-isis-if)#hello-interval 6

Specifies the length of time between hello packets that the software sends.

Step 7

hello-multiplier multiplier [ level { 1 | 2 }]

Example:

RP/0/RP0/CPU0:router(config-isis-if)# hello-multiplier 10

Specifies the number of IS-IS hello packets a neighbor must miss before the router should declare the adjacency as down.

  • A higher value increases the networks tolerance for dropped packets, but also may increase the amount of time required to detect the failure of an adjacent router.

  • Conversely, not detecting the failure of an adjacent router can result in greater packet loss.

Step 8

h ello-password { hmac-md5 | text } { clear | encrypted } password [ level { 1 | 2 }] [ send-only ]

Example:

RP/0/RP0/CPU0:router(config-isis-if)# hello-password text clear mypassword

Specifies that this system include authentication in the hello packets and requires successful authentication of the hello packet from the neighbor to establish an adjacency.

Step 9

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Step 10

show isis [ instance instance-id ] adjacency t ype interface- path-id ] [ detail ] [ systemid system-id ]

Example:

RP/0/RP0/CPU0:router# show isis instance isp adjacency

(Optional) Displays IS-IS adjacencies.

Step 11

show isis adjacency-log

Example:

RP/0/RP0/CPU0:router# show isis adjacency-log

(Optional) Displays a log of the most recent adjacency state transitions.

Step 12

show isis [ instance instance-id ] interface [ type interface-path-id ] [ brief | detail ] [ level { 1 | 2 }]

Example:

RP/0/RP0/CPU0:router# show isis interface HundredGigE 0/9/0/0 brief

(Optional) Displays information about the IS-IS interface.

Step 13

show isis [ instance instance-id ] neighbors [ interface-type interface-instance ] [ summary ] [ detail ] [ systemid system-id ]

Example:

RP/0/RP0/CPU0:router# show isis neighbors summary

(Optional) Displays information about IS-IS neighbors.

MPLS Label Distribution Protocol IGP Synchronization

Multiprotocol Label Switching (MPLS) Label Distribution Protocol (LDP) Interior Gateway Protocol (IGP) Synchronization ensures that LDP has completed label exchange before the IGP path is used for switching. MPLS traffic loss can occur in the following two situations:

  • When an IGP adjacency is established, the router begins forwarding packets using the new adjacency before LDP has exchanged labels with peers on that link.

  • When an LDP session closes, the router continues to forward traffic using the link associated with the LDP peer rather than using an alternate path with an established LDP session.

This feature provides a mechanism to synchronize LDP and IS-IS to minimize MPLS packet loss. The synchronization is accomplished by changing the link metric for a neighbor IS-IS link-state packet (LSP), based on the state of the LDP session.

When an IS-IS adjacency is established on a link but the LDP session is lost or LDP has not yet completed exchanging labels, IS-IS advertises the maximum metric on that link. In this instance, LDP IS-IS synchronization is not yet achieved.


Note

In IS-IS, a link with a maximum wide metric (0xFFFFFF) is not considered for shortest path first (SPF). Therefore, the maximum wide metric of -1 (0XFFFFFE) is used with MPLS LDP IGP synchronization.

When LDP IS-IS synchronization is achieved, IS-IS advertises a regular (configured or default) metric on that link.

Configuring MPLS LDP IS-IS Synchronization

This task explains how to enable Multiprotocol Label Switching (MPLS) Label Distribution Protocol (LDP) IS-IS synchronization. MPLS LDP synchronization can be enabled for an address family under interface configuration mode. Only IPv4 unicast address family is supported. This task is optional.

Procedure

Step 1

configure

Example:

RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

router isis instance-id

Example:

RP/0/RP0/CPU0:router(config)# router isis isp

Enables IS-IS routing for the specified routing process, and places the router in router configuration mode.

  • By default, all IS-IS instances are automatically Level 1 and Level 2. You can change the level of routing to be performed by a particular routing instance by using the is-type command.

Step 3

interface type interface-path-id

Example:

RP/0/RP0/CPU0:router(config-isis)# interface HundredGigE 0/9/0/0

Enters interface configuration mode.

Step 4

address-family ipv4 unicast

Example:

RP/0/RP0/CPU0:router(config-isis-if)# address-family ipv4 unicast

Specifies the IPv4 address family and enters router address family configuration mode.

Step 5

mpls ldp sync [ level { 1 | 2 }]

Example:

RP/0/RP0/CPU0:router(config-isis-if-af)# mpls ldp sync level 1

Enables MPLS LDP synchronization for the IPv4 address family under interface HundredGigE 0/9/0/0.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

IS-IS Overload Bit Avoidance

The IS-IS overload bit avoidance feature allows network administrators to prevent label switched paths (LSPs) from being disabled when a router in that path has its Intermediate System-to-Intermediate System (IS-IS) overload bit set.

When the IS-IS overload bit avoidance feature is activated, all nodes with the overload bit set, including head nodes, mid nodes, and tail nodes, are ignored, which means that they are still available for use with label switched paths (LSPs).


Note

The IS-IS overload bit avoidance feature does not change the default behavior on nodes that have their overload bit set if those nodes are not included in the path calculation (PCALC).

The IS-IS overload bit avoidance feature is activated using the following command:

mpls traffic-eng path-selection ignore overload

The IS-IS overload bit avoidance feature is deactivated using the no form of this command:

no mpls traffic-eng path-selection ignore overload

When the IS-IS overload bit avoidance feature is deactivated, nodes with the overload bit set cannot be used as nodes of last resort.

Configure IS-IS Overload Bit Avoidance

This task describes how to activate IS-IS overload bit avoidance.

Before you begin

The IS-IS overload bit avoidance feature is valid only on networks that support the following features:

  • MPLS

  • IS-IS

Procedure

Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters mode.

Step 2

mpls traffic-eng path-selection ignore overload

Example:


RP/0/RP0/CPU0:router(config)# mpls traffic-eng path-selection ignore overload

Activates IS-IS overload bit avoidance.

Configuring IS-IS Overload Bit Avoidance: Example

The following example shows how to activate IS-IS overload bit avoidance: 


  config
   mpls traffic-eng path-selection ignore overload

The following example shows how to deactivate IS-IS overload bit avoidance: 


  config
   no mpls traffic-eng path-selection ignore overload

IS-IS Penalty for Link Delay Anomaly


Note
For information on configuring the link delay anomaly threshold values, refer to Link Anomaly Detection with IGP Penalty in the Segment Routing Configuration Guide.

When you configure Link Anomaly Detection in SR-PM, PM sets an anomaly bit (A-bit). When IGP receives the A-bit, IGP can automatically increase the IGP metric of the link by a user-defined amount (IGP penalty). This updated IGP metric is advertised in the network to make this link undesirable or unusable. When the link recovers, PM resets the A-bit.


Note

When node is reloaded, the default or configured IGP metric (without penalty) is advertised until a new measurement is available.

Configuration

RP/0/RSP0/CPU0:ios(config)# router isis 100
RP/0/RSP0/CPU0:ios(config-isis)# interface GigabitEthernet 0/1/0/1
RP/0/RSP0/CPU0:ios(config-isis-if)# address-family ipv4 unicast
RP/0/RSP0/CPU0:ios(config-isis-if-af)# metric fallback anomaly delay increment 25
RP/0/RSP0/CPU0:ios(config-isis-if-af)# exit
RP/0/RSP0/CPU0:ios(config-isis-if)# exit
RP/0/RSP0/CPU0:ios(config-isis)# interface GigabitEthernet 0/1/0/2
RP/0/RSP0/CPU0:ios(config-isis-if)# address-family ipv4 unicast
RP/0/RSP0/CPU0:ios(config-isis-if-af)# metric fallback anomaly delay multiplier 2

Running Configuration

router isis 100
 interface GigabitEthernet0/1/0/1
  address-family ipv4 unicast
   metric fallback anomaly delay increment 25
  !
 !
 interface GigabitEthernet0/1/0/2
  address-family ipv4 unicast
   metric fallback anomaly delay multiplier 2
  !
 !
!

References for IS-IS

This section provides additional conceptual information on IS-IS. It includes the following topics:

IS-IS Functional Overview

Small IS-IS networks are typically built as a single area that includes all routers in the network. As the network grows larger, it may be reorganized into a backbone area made up of the connected set of all Level 2 routers from all areas, which is in turn connected to local areas. Within a local area, routers know how to reach all system IDs. Between areas, routers know how to reach the backbone, and the backbone routers know how to reach other areas.

The IS-IS routing protocol supports the configuration of backbone Level 2 and Level 1 areas and the necessary support for moving routing information between the areas. Routers establish Level 1 adjacencies to perform routing within a local area (intra-area routing). Routers establish Level 2 adjacencies to perform routing between Level 1 areas (interarea routing).

Each IS-IS instance can support either a single Level 1 or Level 2 area, or one of each. By default, all IS-IS instances automatically support Level 1 and Level 2 routing. You can change the level of routing to be performed by a particular routing instance using the is-type command.

Multiple IS-IS instances can exist on the same physical interface. However, you must configure different instance-id for every instance that shares the same physical interface.

Alternatively, you can also create dot1q sub-interfaces and configure each dot1q sub-interface to different IS-IS instances.

Default Routes

You can force a default route into an IS-IS routing domain. Whenever you specifically configure redistribution of routes into an IS-IS routing domain, the software does not, by default, redistribute the default route into the IS-IS routing domain. The default-information originate command generates a default route into IS-IS, which can be controlled by a route policy. You can use the route policy to identify the level into which the default route is to be announced, and you can specify other filtering options configurable under a route policy. You can use a route policy to conditionally advertise the default route, depending on the existence of another route in the routing table of the router.

Overload Bit on Router

The overload bit is a special bit of state information that is included in an LSP of the router. If the bit is set on the router, it notifies routers in the area that the router is not available for transit traffic. This capability is useful in four situations:

  1. During a serious but nonfatal error, such as limited memory.

  2. During the startup and restart of the process. The overload bit can be set until the routing protocol has converged. However, it is not employed during a normal NSF restart or failover because doing so causes a routing flap.

  3. During a trial deployment of a new router. The overload bit can be set until deployment is verified, then cleared.

  4. During the shutdown of a router. The overload bit can be set to remove the router from the topology before the router is removed from service.

Overload Bit Configuration During Multitopology Operation

Because the overload bit applies to forwarding for a single topology, it may be configured and cleared independently for IPv4 and IPv6 during multitopology operation. For this reason, the overload is set from the router address family configuration mode. If the IPv4 overload bit is set, all routers in the area do not use the router for IPv4 transit traffic. However, they can still use the router for IPv6 transit traffic.

Attached Bit on an IS-IS Instance

The attached bit is set in a router that is configured with the is-type command and level-1-2 keyword. The attached bit indicates that the router is connected to other areas (typically through the backbone). This functionality means that the router can be used by Level 1 routers in the area as the default route to the backbone. The attached bit is usually set automatically as the router discovers other areas while computing its Level 2 SPF route. The bit is automatically cleared when the router becomes detached from the backbone.


Note

If the connectivity for the Level 2 instance is lost, the attached bit in the Level 1 instance LSP would continue sending traffic to the Level 2 instance and cause the traffic to be dropped.

To simulate this behavior when using multiple processes to represent the level-1-2 keyword functionality, you would manually configure the attached bit on the Level 1 process.

IS-IS Support for Route Tags

The IS-IS Support for route tags feature provides the capability to associate and advertise a tag with an IS-IS route prefix. Additionally, the feature allows you to prioritize the order of installation of route prefixes in the RIB based on a tag of a route. Route tags may also be used in route policy to match route prefixes (for example, to select certain route prefixes for redistribution).

Flood Blocking on Specific Interfaces

With this technique, certain interfaces are blocked from being used for flooding LSPs, but the remaining interfaces operate normally for flooding. This technique is simple to understand and configure, but may be more difficult to maintain and more error prone than mesh groups in the long run. The flooding topology that IS-IS uses is fine-tuned rather than restricted. Restricting the topology too much (blocking too many interfaces) makes the network unreliable in the face of failures. Restricting the topology too little (blocking too few interfaces) may fail to achieve the desired scalability.

To improve the robustness of the network in the event that all nonblocked interfaces drop, use the csnp-interval command in interface configuration mode to force periodic complete sequence number PDUs (CSNPs) packets to be used on blocked point-to-point links. The use of periodic CSNPs enables the network to become synchronized.

Maximum LSP Lifetime and Refresh Interval

By default, the router sends a periodic LSP refresh every 15 minutes. LSPs remain in a database for 20 minutes by default. If they are not refreshed by that time, they are deleted. You can change the LSP refresh interval or maximum LSP lifetime. The LSP interval should be less than the LSP lifetime or else LSPs time out before they are refreshed. In the absence of a configured refresh interval, the software adjusts the LSP refresh interval, if necessary, to prevent the LSPs from timing out.

Mesh Group Configuration

Configuring mesh groups (a set of interfaces on a router) can help to limit flooding. All routers reachable over the interfaces in a particular mesh group are assumed to be densely connected with each router having at least one link to every other router. Many links can fail without isolating one or more routers from the network.

In normal flooding, a new LSP is received on an interface and is flooded out over all other interfaces on the router. With mesh groups, when a new LSP is received over an interface that is part of a mesh group, the new LSP is not flooded over the other interfaces that are part of that mesh group.

Multi-Instance IS-IS

Table 2. Feature History Table

Feature Name

Release Information

Feature Description

32 IS-IS Instances

 Release 7.6.1

You can now configure up to 32 IS-IS instances, thus enhancing the ability to isolate resources within your router and on the network. This ability enables you to configure more instances consuming network-wide resources at different rates, giving you more flexibility to manage your networks efficiently.

In earlier releases, you could configure up to16 IS-IS instances.

You can configure up to 32 IS-IS instances per router, as long as segment-routing mpls is not configured under router isis . Multiple IS-IS instances can share a single interface if the instances you configure are with different instance IDs.

If segment-routing mpls is configured under router isis , then IS-IS takes the connection to Label Switching Database (LSD) component within the router. Configuring 32 IS-IS instances may exceed the 32 connections that are allowed to LSD. These connections are given out on a first-come-first-serve basis. It is possible for the IS-IS instances to take them all and prevent other clients like LDP, BGP, and so on, from getting connections. This may have adverse effects on the system.


Note

IS-IS does not connect to LSD unless it needs one due to segment-routing mpls configuration. So, you can configure any number of connections up to 32, as long as the segment-routing mpls is not configured. Ensure caution if SR-MPLS is in use.

Use the show mpls lsd clients command to determine how many IS-IS instances you can configure. To do this, bring up the system without any IS-IS configuration and observe the number of LSD clients.

For example:

RP/0/0/CPU0:r100#show mpls lsd clients 
Wed Mar 16 08:10:32.646 PDT
ID Services                                Location    
-- ------------------------------------ ------------
0  LSD(A)                                  0/0/CPU0
1  Static(A)                               0/0/CPU0
2  L2VPN(A)                                0/0/CPU0
3  PIM6:pim6(A)                            0/0/CPU0
4  Application-Controller:XTC(A)           0/0/CPU0
5  PIM:pim(A)                              0/0/CPU0
6  BFD(A)                                  0/0/CPU0
7  TE-Control(A)                           0/0/CPU0
8  LDP(A)                                  0/0/CPU0

In the example, nine client connections are being used, leaving 23 for use by IS-IS instances or other clients.

Because the Routing Information Base (RIB) treats each of the IS-IS instances as equal routing clients, you must be careful when redistributing routes between IS-IS instances. The RIB does not know to prefer Level 1 routes over Level 2 routes. For this reason, if you are running Level 1 and Level 2 instances, you must enforce the preference by configuring different administrative distances for the two instances.

Label Distribution Protocol IGP Auto-configuration

Label Distribution Protocol (LDP) Interior Gateway Protocol (IGP) auto-configuration simplifies the procedure to enable LDP on a set of interfaces used by an IGP instance. LDP IGP auto-configuration can be used on a large number interfaces (for example, when LDP is used for transport in the core) and on multiple IGP instances simultaneously.

This feature supports the IPv4 address family for the default VPN routing and forwarding (VRF) instance.

LDP IGP auto-configuration can also be explicitly disabled on individual interfaces under LDP using the igp auto-config disable command. This allows LDP to receive all IGP interfaces except the ones explicitly disabled.

See the MPLS configuration guide for information on configuring LDP IGP auto-configuration.

MPLS LDP-IGP Synchronization Compatibility with LDP Graceful Restart

LDP graceful restart protects traffic when an LDP session is lost. If a graceful restart-enabled LDP session fails, MPLS LDP IS-IS synchronization is still achieved on the interface while it is protected by graceful restart. MPLS LDP IGP synchronization is eventually lost under the following circumstances:

  • LDP fails to restart before the LDP graceful restart reconnect timer expires.

  • The LDP session on the protected interface fails to recover before the LDP graceful restart recovery timer expires.

MPLS LDP-IGP Synchronization Compatibility with IGP Nonstop Forwarding

IS-IS nonstop forwarding (NSF) protects traffic during IS-IS process restarts and route processor (RP) failovers. LDP IS-IS synchronization is supported with IS-IS NSF only if LDP graceful restart is also enabled over the interface. If IS-IS NSF is not enabled, the LDP synchronization state is not retained across restarts and failovers.