Cisco IOS Quality of Service Solutions Command Reference

ip rsvp precedence

To enable the router to mark the IP Precedence value of the type of service (ToS) byte for packets in a Resource Reservation Protocol (RSVP) reserved path using the specified values for packets that either conform to or exceed the RSVP flowspec, use the ip rsvp precedence command in interface configuration mode. To remove existing IP Precedence settings, use the no form of this command.

ip rsvp precedence {conform precedence-value | exceed precedence-value}

no ip rsvp precedence [conform | exceed]

Syntax Description

Command Default

The IP Precedence bits of the ToS byte are left unmodified when this command is not used. The default state is equivalent to execution of the no ip rsvp precedence command.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

Packets in an RSVP reserved path are divided into two classes: those that conform to the reservation flowspec and those that correspond to a reservation but that exceed, or are outside, the reservation flowspec.

The ip rsvp precedence command allows you to set the IP Precedence values to be applied to packets belonging to these two classes. You must specify the IP Precedence value for at least one class of traffic when you use this command. You can use a single instance of the command to specify values for both classes, in which case you can specify the conform and exceed keywords in either order.

As part of its input processing, RSVP uses the ip rsvp precedence command to set the IP Precedence bits on conforming and nonconforming packets. If per-VC DWRED is configured, the system uses the IP Precedence and ToS bit settings on the output interface in its packet drop process. The IP Precedence setting of a packet can also be used by interfaces on downstream routers.

Execution of the ip rsvp precedence command causes IP Precedence values for all preexisting reservations on the interface to be modified.

Note RSVP must be enabled on an interface before you can use this command; that is, use of the ip rsvp bandwidth command must precede use of the ip rsvp precedence command. RSVP cannot be configured with VIP-distributed Cisco Express Forwarding (dCEF).

RSVP receives packets from the underlying forwarding mechanism. Therefore, before you use the ip rsvp precedence command to set IP Precedence, one of the following features is required:

•Weighted fair queueing (WFQ) must be enabled on the interface.

•RSVP switched virtual circuits (SVCs) must be used.

•NetFlow must be configured to assist RSVP.

Note Use of the no form of this command is not equivalent to giving the ip rsvp precedence 0 command, which sets all precedence on the packets to 0, regardless of previous precedence setting.

Examples

The following example shows how to set the IP Precedence value to 3 for all traffic on the ATM interface 0 that conforms to the RSVP flowspec and to 2 for all traffic that exceeds the flowspec:

interface atm0

 ip rsvp precedence conform 3 exceed 2

The following example shows how to set the IP Precedence value to 2 for all traffic on ATM interface 1 that conforms to the RSVP flowspec. The IP Precedence values of those packets that exceed the flowspec are not altered in any way:

interface ATM1

 ip rsvp precedence conform 2

Related Commands

ip rsvp qos

To enable Resource Reservation Protocol (RSVP) quality of service (QoS) flows on a router running Multiprotocol Label Switching traffic engineering (MPLS TE), use the ip rsvp qos command in global configuration mode. To disable RSVP QoS flows, use the no form of this command.

ip rsvp qos

no ip rsvp qos

Syntax Description

This command has no arguments or keywords.

Command Default

RSVP QoS flows are not enabled.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

If RSVP QoS flows and MPLS TE are enabled, the router processes and installs RSVP label switched path (LSP) and IPv4 messages such as PATH and RESV. If RSVP QoS flows and MPLS TE are then disabled with IPv4 and LSP states installed, all installed IPv4 states are immediately cleared. LSP states remain unmodified. Further refreshes or new IPv4 RSVP messages are forwarded unmodified.

Use the show ip rsvp command to display the status of the ip rsvp qos command.

Examples

The following example configures RSVP QoS flows on a router running MPLS TE:

Router> enable

Router# configure terminal

Router(config)# ip rsvp qos

Related Commands

ip rsvp reservation

To enable a router to simulate receiving Resource Reservation Protocol (RSVP) RESV messages from a downstream host, use the ip rsvp reservation command in global configuration mode. To disable this function, use the no form of this command.

ip rsvp reservation session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port next-hop-address next-hop-interface {ff | se | wf} {load | rate} bandwidth burst-size [identity alias]

no ip rsvp reservation session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port next-hop-address next-hop-interface {ff | se | wf} {load | rate} bandwidth burst-size [identity alias]

Syntax Description

Command Default

The router does not simulate receiving RSVP RESV messages.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp reservation command to make the router simulate receiving RSVP RESV messages from a downstream host and to proxy RSVP RESV messages for that host. By giving a local (loopback) next-hop address and next-hop interface, you can also use this command to proxy RSVP for the router that you are configuring or you can use the ip rsvp reservation-host command.

An alias must reference an RSVP identity that you created by using the ip rsvp identity command. The policy-locator string associated with this identity is signaled in the RESV message. This identity overrides any application ID that is contained in the matching PATH message.

If the matching PATH message has an application ID, but you have not specified an application ID using the ip rsvp reservation command, the RESV message will not contain an application ID. However, the RESV message proxied by the ip rsvp listener command does put the matching PATH message application ID into the proxied RESV message.

Examples

The following example shows how to specify the use of a Shared Explicit style of reservation and the controlled load service, with token buckets of 100 or 150 kbps and a maximum queue depth of 60 or 65 kbps:

Router(config)# ip rsvp reservation 192.168.0.2 172.16.1.1 udp 20 30 172.16.4.1 Ethernet1 
se load 100 60

Router(config)# ip rsvp reservation 192.168.0.2 172.16.2.1 tcp 20 30 172.16.4.1 Ethernet1 
se load 150 65 

The following example shows how to specify the use of a Wildcard Filter style of reservation and the guaranteed bit rate service, with token buckets of 300 or 350 kbps, a maximum queue depth of 60 or 65 kbps, and an application ID:

Router(config)# ip rsvp reservation 192.168.0.3 0.0.0.0 udp 20 0 172.16.4.1 Ethernet1 wf 
rate 300 60 identity xyz

Router(config)# ip rsvp reservation 192.168.1.1 0.0.0.0 udp 20 0 172.16.4.1 Ethernet1 wf 
rate 350 65 identity xyz

Note that the wildcard filter does not admit the specification of the sender; it accepts all senders. This action is denoted by setting the source address and port to zero. If, in any filter style, the destination port is specified to be zero, RSVP does not permit the source port to be anything else; it understands that such protocols do not use ports or that the specification applies to all ports.

Related Commands

ip rsvp reservation-host

To enable a router to simulate a host generating Resource Reservation Protocol (RSVP) RESV messages, use the ip rsvp reservation-host command in global configuration mode. To disable this function, use the no form of this command.

ip rsvp reservation-host session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port {ff | se | wf} {load | rate} bandwidth burst-size [identity alias] [vrf vrf-name]

no ip rsvp reservation-host session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port {ff | se | wf} {load | rate} bandwidth burst-size [identity alias] [vrf vrf-name]

Syntax Description

Command Default

The router does not simulate a host generating RSVP RESV messages.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Note The syntax of the command depends on your platform and release. The vrf vrf-name keyword and argument combination is not supported on ASR 1000 Series Aggregation Services Routers.

Use the ip rsvp reservation-host command to make a router simulate a host generating its own RSVP RESV messages. This command is similar to the ip rsvp reservation command, which can cause a router to generate RESV messages on behalf of another host. The main differences between the ip rsvp reservation-host and ip rsvp reservation commands follow:

•When you enter the ip rsvp reservation-host command, the session-ip-address argument must be a local address configured on an interface on the router. Therefore, you cannot proxy a reservation on behalf of a flow that is destined for another host. Also, you cannot use this command to generate reservation messages for multicast sessions.

•Because the message is assumed to originate from the router that you are configuring, you do not specify a next-hop or incoming interface for the RSVP RESV message when entering the ip rsvp reservation-host command.

•Use the ip rsvp reservation-host command for debugging and testing purposes because you cannot use it to proxy RSVP for non-RSVP-capable hosts or for multicast sessions.

An alias must reference an RSVP identity that you created by using the ip rsvp identity command. The policy-locator string associated with this identity is signaled in the RESV message. This identity overrides any application ID that is contained in the matching PATH message.

If the matching PATH message has an application ID, but you have not specified an application ID using the ip rsvp reservation-host command, the RESV message does not contain an application ID. However, the RESV message proxied by the ip rsvp listener command does put the matching PATH message application ID into the proxied RESV message.

Examples

The following example specifies the use of a Shared Explicit style of reservation and the controlled load service, with token buckets of 100 or 150 kbps, 60 or 65 kbps maximum queue depth, and an application ID:

Router(config)# ip rsvp reservation-host 10.1.1.1 10.30.1.4 udp 20 30 se load 100 60 
identity xyz

Router(config)# ip rsvp reservation-host 10.40.2.2 10.22.1.1 tcp 20 30 se load 150 65 
identity xyz

Related Commands

ip rsvp resource-provider

To configure a resource provider for an aggregate flow, use the ip rsvp resource-provider command in interface configuration mode. To disable a resource provider for an aggregate flow, use the no form of this command.

ip rsvp resource-provider {none | wfq interface | wfq pvc}

no ip rsvp resource-provider

Syntax Description

Command Default

WFQ (the wfq interface keyword) is the default resource provider that Resource Reservation Protocol (RSVP) configures on the interface.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

Note The syntax of the command depends on your platform and image. The wfq interface and wfq pvc keywords are not supported on Cisco ASR 1000 series routers.

Use the ip rsvp resource-provider command to configure the resource provider with which you want RSVP to interact when it installs a reservation.

To ensure that a flow receives quality of service (QoS) guarantees when using WFQ on a per-flow basis, configure wfq interface or wfq pvc as the resource provider. To ensure that a flow receives QoS guarantees when using class-based weighted fair queueing (CBWFQ) for data packet processing, configure none as the resource provider.

Note Resource provider was formerly called QoS provider.

Examples

In the following example, the ip rsvp resource-provider command is configured with wfq pvc as the resource provider, ensuring that a flow receives QoS guarantees when using WFQ on a per-flow basis:

Router# configure terminal

Router(config)# interface atm 6/0

Router(config-if)# ip rsvp resource-provider wfq pvc

In the following example, the ip rsvp resource-provider command is configured with none as the resource provider, ensuring that a flow receives QoS guarantees when using CBWFQ for data-packet processing:

Router# configure terminal

Router(config)# interface atm 6/0

Router(config-if)# ip rsvp resource-provider none

Related Commands

ip rsvp sender

To enable a router to simulate receiving Resource Reservation Protocol (RSVP) PATH messages, use the ip rsvp sender command in global configuration mode. To disable this function, use the no form of this command.

ip rsvp sender session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port previous-hop-ip-address previous-hop-interface bandwidth burst-size [identity alias]

no ip rsvp sender session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port previous-hop-ip-address previous-hop-interface bandwidth burst-size [identity alias]

Syntax Description

Command Default

The router does not simulate receiving RSVP PATH messages.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp sender command to make the router simulate the receiving of RSVP PATH messages from an upstream host and to proxy RSVP PATH messages from that host. By including a local (loopback) previous-hop address and previous-hop interface, you can also use this command to proxy RSVP for the router that you are configuring.

An alias must reference an RSVP identity that you created by using the ip rsvp identity command. The policy-locator string associated with this identity is supplied in the PATH message.

Examples

The following example sets up the router to act as though it is receiving RSVP PATH messages using UDP over loopback interface 1:

Router(config)# ip rsvp sender 192.168.0.1 172.16.2.1 udp 20 30 172.16.2.1 loopback1 50 5 
identity xyz

Router(config)# ip rsvp sender 192.168.0.2 172.16.2.1 udp 20 30 172.16.2.1 loopback1 50 5 
identity xyz

Related Commands

ip rsvp sender-host

To enable a router to simulate a host generating a Resource Reservation Protocol (RSVP) PATH message, use the ip rsvp sender-host command in global configuration mode. To disable this function, use the no form of this command.

ip rsvp sender-host session-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port bandwidth burst-size [identity alias] [vrf vrf-name]

no ip rsvp sender-host ssession-ip-address sender-ip-address {ip-protocol | tcp | udp} session-dest-port sender-source-port bandwidth burst-size [identity alias] [vrf vrf-name]

Syntax Description

Command Default

The router does not simulate RSVP PATH message generation.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Note The syntax of the command depends on your platform and release. The vrf vrf-name keyword and argument combination is not supported on ASR 1000 Series Aggregation Services Routers.

Use the ip rsvp sender-host command to make a router simulate a host generating its own RSVP PATH messages. This command is similar to the ip rsvp sender command, which can cause a router to generate RSVP PATH messages on behalf of another host. The main differences between the ip rsvp sender-host and ip rsvp sender commands follow:

•When you enter the ip rsvp sender-host command, the sender-ip-address argument must be a local address configured on an interface of the router.

•Because the message is assumed to originate from the router that you are configuring, you do not specify a previous-hop or incoming interface for the RSVP PATH message when entering the ip rsvp sender-host command.

•Use the ip rsvp sender-host command for debugging and testing purposes because you cannot use it to proxy RSVP for non-RSVP-capable hosts.

An alias must reference an RSVP identity that you created by using the ip rsvp identity command. The policy-locator string associated with this identity is signaled in the RESV message. This identity overrides any application ID that is contained in the matching PATH message.

Examples

The following example sets up the router to act like a host that sends traffic to the given address:

Router(config)# ip rsvp sender-host 10.0.0.7 10.0.0.1 udp 1 1 10 10 identity xyz

Related Commands

ip rsvp signalling dscp

To specify the differentiated services code point (DSCP) value to be used on all Resource Reservation Protocol (RSVP) messages that are transmitted on an interface, use the ip rsvp signalling dscp command in interface configuration mode. To restore the default DSCP value on all RSVP messages transmitted on an interface, use the no form of this command.

ip rsvp signalling dscp value

no ip rsvp signalling dscp

Syntax Description

Command Default

The default DSCP value is 0.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You must configure the DSCP per interface, not per flow. The DSCP determines the priority that a packet receives from various hops as it travels to its destination.

The DSCP applies to all RSVP flows that are installed on a specific interface. You can configure each interface independently for DSCP.

Examples

The following example shows how to configure the ip rsvp signalling dscp command with a DSCP value of 6:

Router> enable

Router# configure terminal

Router(config)# interface gigabitEthernet 0/0/0

Router(config-if)# ip rsvp signalling dscp 6

Router(config-if)# end

Related Commands

ip rsvp signalling fast-local-repair notifications

To configure the number of per flow notifications that Resource Reservation Protocol (RSVP) processes during a fast local repair (FLR) procedure before suspending, use the ip rsvp signalling fast-local-repair notifications command in global configuration mode. To set the number of notifications to its default, use the no form of this command.

ip rsvp signalling fast-local-repair notifications number

no ip rsvp signalling fast-local-repair notifications

Syntax Description

Command Default

Notifications are sent by the Routing Information Base (RIB) and processed by RSVP. If the ip rsvp signalling fast-local-repair notifications command is not configured, RSVP processes 1000 notifications, suspends the notifications, and then resumes processing of another 1000 notifications.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Upon a route change, RIB builds a list of notifications, one per affected flow, and notifies RSVP by sending an event including these notifications. Therefore, these events can contain thousands of elements, depending on the number of path state blocks (PSBs) affected.

RSVP processes, by default, 1000 notifications at a time and then suspends if required, to prevent the CPU from being overwhelmed. However, you can configure this number using the ip rsvp signalling fast-local-repair notifications command.

Examples

The following example shows how to configure the number of flows that are repaired before RSVP suspends to 100:

Router(config)# ip rsvp signalling fast-local-repair notifications 100

Related Commands

ip rsvp signalling fast-local-repair rate

To configure the repair rate that Resource Reservation Protocol (RSVP) uses for a fast local repair (FLR) procedure, use the ip rsvp signalling fast-local-repair rate command in global configuration mode. To set the repair rate to its default, use the no form of this command.

ip rsvp signalling fast-local-repair rate messages-per-second

no ip rsvp signalling fast-local-repair rate

Syntax Description

Command Default

If this command is not configured, the RSVP message pacing rate is used.

Note The RSVP message pacing rate is enabled by default in Cisco IOS Release 12.2 and later releases.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

The default repair rate is based on the RSVP message pacing rate.

If you configure the FLR rate by using the ip rsvp signalling fast-local-repair rate command, and RSVP message pacing is enabled, the lower FLR rate and the RSVP message pacing rate takes effect. If you disable the RSVP rate limit by using the no ip rsvp signalling rate-limit command, then the FLR rate is used. However, if you disable the RSVP rate limit and do not configure an FLR rate, then RSVP performs no message pacing and messages are sent back-to-back. This action is not recommended because the point of local repair (PLR) may flood the downstream node with PATH messages causing some of them to be dropped.

The repair rate is determined at notification time, and this same rate is used during the time of the repair even if you change either the RSVP message pacing rate or the FLR rate during this time.

Examples

The following example shows how to configure a repair rate of 100 messages per second:

Router(config)# ip rsvp signalling fast-local-repair rate 100

Related Commands

ip rsvp signalling fast-local-repair wait-time

To configure the delay that Resource Reservation Protocol (RSVP) uses before starting a fast local repair (FLR) procedure, use the ip rsvp signalling fast-local-repair wait-time command in interface configuration mode. To set the delay to its default, use the no form of this command.

ip rsvp signalling fast-local-repair wait-time interval

no ip rsvp signalling fast-local-repair wait-time

Syntax Description

Command Default

This command is disabled by default; therefore, no delay is configured.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

Use the ip rsvp signalling fast-local-repair wait-time command to configure the delay desired in starting an FLR procedure. If you do not configure a delay, then path refreshes are triggered immediately after RSVP receives a route change notification from the routing information base (RIB).

Examples

The following example configures a delay of 100 ms:

Router(config-if)# ip rsvp signalling fast-local-repair wait-time 100

Related Commands

ip rsvp signalling hello (configuration)

To enable hello globally on a router, use the ip rsvp signalling hello command in global configuration mode. To disable hello globally on a router, use the no form of this command.

ip rsvp signalling hello

no ip rsvp signalling hello

Syntax Description

This command has no arguments or keywords.

Command Default

No hellos are enabled.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

To enable hello globally on a router, you must enter this command. You also must enable hello on the interface.

Examples

In the following example, hello is enabled globally on a router:

Router(config)# ip rsvp signalling hello 

Related Commands

ip rsvp signalling hello (interface)

To enable hello on an interface where you need Fast Reroute protection, use the ip rsvp signalling hello command in interface configuration mode. To disable hello on an interface where you need Fast Reroute protection, use the no form of this command

ip rsvp signalling hello

no ip rsvp signalling hello

Syntax Description

This command has no arguments or keywords.

Command Default

No hellos are enabled.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You must configure hello globally on a router and on the specific interface.

Examples

In the following example, hello is enabled on an interface:

Router(config-if)# ip rsvp signalling hello 

Related Commands

ip rsvp signalling hello dscp

To set the differentiated services code point (DSCP) value that is in the IP header of a Resource Reservation Protocol (RSVP) traffic engineering (TE) hello message sent from an interface, use the ip rsvp signalling hello dscp command in interface configuration mode. To set the DSCP value to its default, use the no form of this command.

ip rsvp signalling hello [fast-reroute] dscp num

no ip rsvp signalling hello [fast-reroute] dscp

Syntax Description

Command Default

The default DSCP value is 48.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

If a link is congested, it is recommended that you set the DSCP to a value higher than 0 to reduce the likelihood that hello messages will be dropped.

You configure the DSCP per interface, not per flow.

The DSCP applies to the RSVP hellos created on a specific interface. You can configure each interface independently for DSCP.

If you issue the ip rsvp signalling hello dscp command without the optional fast-reroute keyword, the command applies to Fast Reroute hellos. This command is provided for backward compatibility; however, we recommend that you use the ip rsvp signalling hello fast-reroute dscp command.

Examples

In the following example, hello messages sent from this interface have a DSCP value of 30 and Fast Reroute capability is enabled by specifying the fast-reroute keyword:

Router(config-if)# ip rsvp signalling hello fast-reroute dscp 30

In the following example, hello messages sent from this interface have a DSCP value of 30 and Fast Reroute capability is enabled by default:

Router(config-if)# ip rsvp signalling hello dscp 30

Related Commands

ip rsvp signalling hello graceful-restart

To enable the Resource Reservation protocol (RSVP) traffic engineering (TE) graceful restart capability on a neighboring router, use the ip rsvp signalling hello graceful-restart command in interface configuration mode. To disable the graceful restart capability, use the no form of this command.

ip rsvp signalling hello graceful-restart

no ip rsvp signalling hello graceful-restart

Syntax Description

This command has no arguments or keywords.

Command Default

Neighboring routers have only node hello enabled.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

Use the ip rsvp signalling hello graceful-restart command to enable support for graceful restart on routers helping their neighbors recover TE tunnels following stateful switchover (SSO).

Note This command is optional. Use it when node hello is not supported.

Examples

The following example configures graceful restart on POS interface 1/0/0 of a neighboring router with the IP address 10.0.0.1:

Router# configure terminal  
Enter configuration commands, one per line. End with CTTL/Z. 
Router(config)# interface POS1/0/0 
Router(config-if)# ip rsvp signalling hello graceful-restart

Related Commands

ip rsvp signalling hello graceful-restart dscp

To set the differentiated services code point (DSCP) value that is in the IP header of a Resource Reservation Protocol (RSVP) traffic engineering (TE) graceful restart hello message, use the ip rsvp signalling hello graceful-restart dscp command in global configuration mode. To set the DSCP valueto its default, use the no form of this command.

ip rsvp signalling hello graceful-restart dscp num

no ip rsvp signalling hello graceful-restart dscp

Syntax Description

Command Default

The default DSCP value is 48.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

If a link is congested, set the DSCP to a value higher than 0 to reduce the likelihood that hello messages get dropped.

The DSCP applies to the RSVP hellos created on a specific router. You can configure each router independently for the DSCP.

Examples

In the following example, hello messages have a DSCP value of 30:

Router(config)# ip rsvp signalling hello graceful-restart dscp 30

Related Commands

ip rsvp signalling hello graceful-restart mode

To enable Resource Reservation Protocol (RSVP) traffic engineering (TE) graceful restart capability on a Route Processor (RP), use the ip rsvp signalling hello graceful-restart mode command in global configuration mode. To disable graceful restart capability, use the no form of this command.

Cisco IOS 12.0(29)S, 12.2(33)SRA, 12.2(33)SXH, and Later Releases

ip rsvp signalling hello graceful-restart mode {help-neighbor | full}

no ip rsvp signalling hello graceful-restart mode

Cisco IOS T and XE Trains

ip rsvp signalling hello graceful-restart mode help-neighbor

no ip rsvp signalling hello graceful-restart mode help-neighbor

Syntax Description

Command Default

Graceful restart is disabled.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling hello graceful-restart mode help-neighbor command to enable support capability for a neighboring router to restart after a failure.

Use the ip rsvp signalling hello graceful-restart mode full command to enable support capability for a router to begin self-recovery or help its neighbor to restart on platforms that support stateful switchover (SSO), such as Cisco 7600 series routers, provided that you have installed and configured a standby RP.

Examples

The following example shows how to configure an RP with support capability to perform self-recovery after a failure:

Router(config)# ip rsvp signalling hello graceful-restart mode full

Related Commands

ip rsvp signalling hello graceful-restart mode help-neighbor

Note Effective with Cisco IOS Release 12.2(33)SRA, the ip rsvp signalling hello graceful-restart mode help-neighbor command is replaced by the ip rsvp signalling hello graceful-restart mode command. See the ip rsvp signalling hello graceful-restart mode command for more information.

To enable Resource Reservation Protocol (RSVP) traffic engineering (TE) graceful restart capability on a neighboring router, use the ip rsvp signalling hello graceful-restart mode help-neighbor command in global configuration mode. To disable graceful restart capability, use the no form of this command.

ip rsvp signalling hello graceful-restart mode help-neighbor

no ip rsvp signalling hello graceful-restart mode help-neighbor

Syntax Description

This command has no arguments or keywords.

Command Default

Graceful restart is disabled.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling hello graceful-restart mode help-neighbor command to restart a neighboring router.

Examples

In the following example, graceful restart is enabled:

Router(config)# ip rsvp signalling hello graceful-restart mode help-neighbor

Related Commands

ip rsvp signalling hello graceful-restart neighbor

To enable Resource Reservation Protocol (RSVP) traffic engineering (TE) graceful restart capability on a neighboring router, use the ip rsvp signalling hello graceful-restart neighbor command in interface configuration mode. To disable graceful restart capability, use the no form of this command.

ip rsvp signalling hello graceful-restart neighbor ip-address

no ip rsvp signalling hello graceful-restart neighbor ip-address

Syntax Description

Command Default

No neighboring routers have graceful restart capability enabled until you issue this command.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

Use the ip rsvp signalling hello graceful-restart neighbor command to enable support for graceful restart on routers helping their neighbors recover TE tunnels following stateful switchover (SSO).

Note You must issue this command on every interface of the neighboring router that you want to help restart.

Examples

The following example configures graceful restart on POS interface 1/0/0 of a neighboring router with the IP address 10.0.0.1:

Router# configure terminal

Enter configuration commands, one per line.  End with CNTL/Z.

Router(config)# interface POS1/0/0

Router(config-if)# ip rsvp signalling hello graceful-restart neighbor 10.0.0.1

Related Commands

ip rsvp signalling hello graceful-restart refresh interval

To configure the Resource Reservation Protocol (RSVP) traffic engineering (TE) refresh interval in graceful restart hello messages, use the ip rsvp signalling hello grateful-restart refresh interval command in global configuration mode. To set the interval to its default value, use the no form of this command.

ip rsvp signalling hello graceful-restart refresh interval interval-value

no ip rsvp signalling hello graceful-restart refresh interval

Syntax Description

Command Default

1000 milliseconds (10 seconds)

Command Modes

Global configuration (config)

Command History

Usage Guidelines

A node periodically generates a hello message that contains a Hello Request object for all its neighbors. The frequency of those hello messages is determined by the hello interval.

Note If you change the default value for this command and you are also using the ip rsvp signalling refresh interval command, ensure that the value for the ip rsvp signalling hello graceful-restart refresh interval command is less than the value for the ip rsvp signalling refresh interval command. Otherwise, some or all of the label-switched paths (LSPs) may not be recovered after a stateful switchover (SSO) has occurred. We recommend that the value for the ip rsvp signalling refresh interval command be twice the value for the ip rsvp signalling hello graceful-restart refresh interval command.

Examples

The following example shows how to configure hello requests to be sent to a neighbor every 5000 ms:

Router(config)# ip rsvp signalling hello graceful-restart refresh interval 5000

Related Commands

ip rsvp signalling hello graceful-restart refresh misses

To specify how many sequential Resource Reservation Protocol (RSVP) traffic engineering (TE) graceful restart hello acknowledgments (ACKs) a node can miss before the node considers communication with its neighbor lost, use the ip rsvp signalling hello graceful-restart refresh misses command in global configuration mode. To return the missed refresh limit to its default value, use the no form of this command.

ip rsvp signalling hello graceful-restart refresh misses msg-count

no ip rsvp signalling hello graceful-restart refresh misses

Syntax Description

Command Default

The default number of sequential hello acknowledgments is 4.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

A hello message comprises a hello message, a Hello Request object, and a Hello ACK object. Each request is answered by an acknowledgment. If a link is congested or a router has a heavy load, set this number to a value higher than the default value to ensure that hello does not falsely declare that a neighbor is down.

Note If you change the default value for this command and you are also using the ip rsvp signalling hello refresh misses command, ensure that the value for the ip rsvp signalling hello graceful-restart refresh misses command is less than the value for the ip rsvp signalling hello refresh misses command. Otherwise, some or all of the label-switched paths (LSPs) may not be recovered after a stateful switchover (SSO) has occurred. We recommend that the value for the ip rsvp signalling hello refresh misses command be twice the value for the ip rsvp signalling hello graceful-restart refresh misses command.

Examples

The following example shows how to set the router so that, if the node does not receive five sequential hello acknowledgments, the node declares that its neighbor is down:

Router(config)# ip rsvp signalling hello graceful-restart refresh misses 5

Related Commands

ip rsvp signalling hello graceful-restart send

To configure the time for Resource Reservation Protocol (RSVP) label switched paths (LSPs) in a Multiprotocol Label Switching (MPLS) traffic engineering (TE) network to recover or restart after a stateful switchover (SSO) occurs, use the ip rsvp signalling hello graceful-restart send command in global configuration mode. To keep the default recovery and restart times, use the no form of this command.

ip rsvp signalling hello graceful-restart send {recovery-time ms | restart-time ms}

no ip rsvp signalling hello graceful-restart send {recovery-time ms | restart-time ms}

Syntax Description

Command Default

The default recovery and restart times of 120,000 and 30,000 ms, respecively, are in effect until you change them.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling hello graceful-restart send command to give LSPs a longer time to recover or restart after an SSO occurs. Otherwise, the LSPs may not all come back up and your network performance is negatively affected.

Examples

In the following example, a recovery time of 300,000 ms is configured:

Router(config)# ip rsvp signalling hello graceful-restart send recovery-time 300000

Related Commands

ip rsvp signalling hello refresh interval

To configure the Resource Reservation Protocol (RSVP) traffic engineering (TE) hello refresh interval, use the ip rsvp signalling hello refresh interval command in interface configuration mode. To set the refresh interval to its default value, use the no form of this command.

ip rsvp signalling hello [fast-reroute] refresh interval interval-value

no ip rsvp signalling hello [fast-reroute] refresh interval

Syntax Description

Command Default

The default frequency at which a node sends hello messages to a neighbor is 200 msec.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You can configure the hello request interval on a per-interface basis. A node periodically generates a hello message containing a Hello Request object for each neighbor whose status is being tracked. The frequency of those hello messages is determined by the hello interval.

If you issue the ip rsvp signalling hello refresh interval command without the optional fast-reroute keyword, the command applies to Fast Reroute hellos. This command is provided for backward compatibility; however, we recommend that you use the ip rsvp signalling hello fast-reroute refresh interval command.

Examples

In the following example, hello requests are sent to a neighbor every 5000 milliseconds and Fast Reroute capability is enabled by specifying the fast-reroute keyword:

Router(config-if)# ip rsvp signalling hello fast-reroute refresh interval 5000 

In the following example, hello requests are sent to a neighbor every 5000 milliseconds and Fast Reroute capability is enabled by default:

Router(config-if)# ip rsvp signalling hello refresh interval 5000

Related Commands

ip rsvp signalling hello refresh misses

To specify how many Resource Reservation Protocol (RSVP) traffic engineering (TE) hello acknowledgments a node can miss in a row before the node considers that communication with its neighbor is down, use the ip rsvp signalling hello refresh misses command in interface configuration mode. To return the missed refresh limit to its default value, use the no form of this command.

ip rsvp signalling hello [fast-reroute] refresh misses msg-count

no ip rsvp signalling hello [fast-reroute] refresh misses

Syntax Description

Command Default

The default number of sequential hello acknowledgments is 4.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

A hello comprises a hello message, a Hello Request object, and a Hello ACK object. Each request is answered by an acknowledgment. If a link is very congested or a router has a very heavy load, set this number to a value higher than the default value to ensure that hello does not falsely declare that a neighbor is down.

If you issue the ip rsvp signalling hello refresh misses command without the optional fast-reroute keyword, the command applies to Fast Reroute hellos and Fast Reroute capability is enabled by default. This command is provided for backward compatibility; however, we recommend that you use the ip rsvp signalling hello fast-reroute refresh misses command.

Examples

The following example shows how to configure the router so that, if the node does not receive five hello acknowledgments in a row, the node declares that its neighbor is down and Fast Reroute is enabled by specifying the fast-reroute keyword:

Router(config-if)# ip rsvp signalling hello fast-reroute refresh misses 5

The following example shows how to configure the router so that, if the node does not receive five hello acknowledgments in a row, the node declares that its neighbor is down and Fast Reroute is enabled by default:

Router(config-if)# ip rsvp signalling hello refresh misses 5

Related Commands

ip rsvp signalling hello reroute dscp

To set the differentiated services code point (DSCP) value that is in the IP header of a Resource Reservation Protocol (RSVP) traffic engineering (TE) reroute hello (for state timeout) message sent from an interface, use the ip rsvp signalling hello reroute dscp command in interface configuration mode. To set the DSCP value to its default, use the no form of this command.

ip rsvp signalling hello reroute dscp num

no ip rsvp signalling hello reroute dscp

Syntax Description

Command Default

The default DSCP value is 48.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

If a link is congested, you should set the DSCP to a value higher than 0 to reduce the likelihood that hello messages get dropped.

You configure the DSCP per interface, not per flow.

The DSCP applies to the RSVP hellos created on a specific interface. You can configure each interface independently for DSCP.

Examples

The following example shows how to configure the interface so that hello messages sent from this interface have a DSCP value of 30:

Router(config-if)# ip rsvp signalling hello reroute dscp 30

Related Commands

ip rsvp signalling hello reroute refresh interval

To configure the Resource Reservation Protocol (RSVP) traffic engineering (TE) reroute hello (for state timeout) refresh interval, use the ip rsvp signalling hello reroute refresh interval command in interface configuration mode. To set the refresh interval to its default value, use the no form of this command.

ip rsvp signalling hello reroute refresh interval interval-value

no ip rsvp signalling hello reroute refresh interval

Syntax Description

Command Default

The default frequency at which a node sends hello messages to a neighbor is 1000 milliseconds (10 seconds).

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You can configure the hello request interval on a per-interface basis. A node periodically generates a hello message containing a Hello Request object for each neighbor whose status is being tracked. The frequency of those hello messages is determined by the hello interval. For some routers, if you set the interval to a value less than the default value, CPU usage may be high.

Examples

The following example shows how to configure the router so that hello requests are sent to a neighbor every 5000 milliseconds and Fast Reroute capability is enabled by default:

Router(config-if)# ip rsvp signalling hello reroute refresh interval 5000

Related Commands

ip rsvp signalling hello reroute refresh misses

To specify how many Resource Reservation Protocol (RSVP) traffic engineering (TE) reroute hello (for state timeout) acknowledgments (ACKs) a node can miss in a row before the node considers communication with its neighbor is down, use the ip rsvp signalling hello reroute refresh misses command in interface configuration mode. To return the missed refresh limit to its default value, use the no form of this command.

ip rsvp signalling hello reroute refresh misses msg-count

no ip rsvp signalling hello reroute refresh misses

Syntax Description

Command Default

The default is 4.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

A hello comprises a hello message, a Hello Request object, and a Hello ACK object. Each request is answered by an acknowledgment. If a link is very congested or a router has a very heavy load, set this number to a value higher than the default value to ensure that hello does not falsely declare that a neighbor is down.

Examples

The following example shows how to configure the router so that, if the node does not receive five hello acknowledgments in a row, the node declares that its neighbor is down:

Router(config-if)# ip rsvp signalling hello reroute refresh misses 5

Related Commands

ip rsvp signalling hello statistics

To enable hello statistics on a router, use the ip rsvp signalling hello statistics command in global configuration mode. To disable hello statistics on a router, use the no form of this command.

ip rsvp signalling hello statistics

no ip rsvp signalling hello statistics

Syntax Description

This command has no arguments or keywords.

Command Default

No hello statistics are enabled.

Command Modes

Global configuration (config)

Command History

Examples

In the following example, hello statistics are enabled on a router:

Router(config)# ip rsvp signalling hello statistics 

Related Commands

ip rsvp signalling initial-retransmit-delay

To configure the minimum amount of time that a Resource Reservation Protocol (RSVP)-configured router waits for an acknowledgment (ACK) message before retransmitting the same message, use the ip rsvp signalling initial-retransmit-delay command in global configuration mode. To reset the delay value to its default, use the no form of this command.

ip rsvp signalling initial-retransmit-delay delay-value

no ip rsvp signalling initial-retransmit-delay

Syntax Description

Defaults

The default value is 1000 ms (1.0 sec).

Command Modes

Global configuration

Command History

Usage Guidelines

Use the ip rsvp signalling initial-retransmit-delay command to configure the minimum amount of time that a router waits for an ACK message before retransmitting the same message.

If an ACK is not received for a state, the first retransmit occurs after the initial retransmit interval. If no ACK is received after the first retransmit, a second retransmit occurs. The message continues to be retransmitted, with the gap between successive retransmits being twice the previous interval, until an ACK is received. Then the message drops into normal refresh schedule if it needs to be refreshed (Path and Resv messages), or is processed (Error or Tear messages). If no ACK is received after five retransmits, the message is discarded as required.

Examples

The following command shows how to set the initial-retransmit-delay to 2 seconds:

Router(config)# ip rsvp signalling initial-retransmit-delay 2000

The following command shows how to reset the initial-retransmit-delay to the default (1.0 sec):

Router(config)# no ip rsvp signalling initial-retransmit-delay

ip rsvp signalling patherr state-removal

To reduce the amount of Resource Reservation Protocol (RSVP) traffic messages in a network, use the ip rsvp signalling patherr state-removal command in global configuration mode. To disable this function, use the no form of this command.

ip rsvp signalling patherr state-removal [neighbor acl]

no ip rsvp signalling patherr state-removal

Syntax Description

Defaults

Disabled

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling patherr state-removal command to allow routers to delete Path state automatically when forwarding a PathError message, thereby eliminating the need for a subsequent PathTear message.

This command is most effective when all network nodes support this feature. All nodes need to have the latest version of Cisco IOS software configured.

This command applies only to label-switched path (LSP) flows.

Examples

The following command shows how to enable ip rsvp signalling patherr state-removal:

Router(config)# ip rsvp signalling patherr state-removal

The following command shows how to disable ip rsvp signalling patherr state-removal:

Router(config)# no ip rsvp signalling patherr state-removal

The following example shows how to enable ip rsvp signalling patherr state-removal based on an access control list (ACL):

Router(config)# ip rsvp signalling patherr state-removal neighbor 98

The following example shows how to disable ip rsvp signalling patherr state-removal based on an ACL:

Router(config)# no ip rsvp signalling patherr state-removal neighbor 98

ip rsvp signalling rate-limit

To control the transmission rate for Resource Reservation Protocol (RSVP) messages that are sent to a neighboring router during a specified amount of time, use the ip rsvp signalling rate-limit command in global configuration mode. To disable this function, use the no form of this command.

Releases Before Cisco IOS Release 12.4(20)T

ip rsvp signalling rate-limit [burst number] [maxsize bytes] [period ms]

no ip rsvp signalling rate-limit

Cisco IOS 12.0S Releases, 12.2S Releases, XE 2 Releases, Release 12.4(20)T, and Later T Releases

ip rsvp signalling rate-limit [burst number] [limit number] [maxsize bytes] [period ms]

no ip rsvp signalling rate-limit

Syntax Description

Command Default

If you do not enter this command, the default values are used.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling rate-limit command to prevent a burst of RSVP traffic engineering signaling messages from overflowing the input queue of a receiving router, which would cause the router to drop some messages. Dropped messages substantially delay the completion of signaling.

This command replaces the ip rsvp msg-pacing command.

Examples

The following command shows how 6 messages with a message queue of 500 bytes are sent every 10 ms to any neighboring router:

Router(config)# ip rsvp signalling rate-limit burst 6 maxsize 500 period 10

Related Commands

ip rsvp signalling refresh interval

To specify the interval between sending refresh messages for each Resource Reservation Protocol (RSVP) state, use the ip rsvp signalling refresh interval command in global configuration mode. To set the interval to its default value, use the no form of the command.

ip rsvp signalling refresh interval interval-value

no ip rsvp signalling refresh interval

Syntax Description

Command Default

30000 milliseconds (30 seconds)

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling refresh interval command to specify the interval between sending refresh messages for each RSVP state.

The RSVP protocol relies on a soft-state mechanism to maintain state consistency in the face of network losses. This mechanism is based on continuous refresh messages to keep a state current. Each RSVP router is responsible for sending periodic refresh messages to its neighbors.

Note If you change the default value for this command and you are also using the ip rsvp signalling hello graceful-restart refresh interval command, ensure that the value for the ip rsvp signalling hello graceful-restart refresh interval command is less than the value for the ip rsvp signalling refresh interval command. Otherwise, some or all of the label-switched paths (LSPs) may not be recovered after a stateful switchover (SSO) has occurred. We recommend that the value for the ip rsvp signalling refresh interval command be twice the value for the ip rsvp signalling hello graceful-restart refresh interval command.

Examples

The following example shows how to specify a refresh interval of 60000 milliseconds (60 seconds):

Router(config)# ip rsvp signalling refresh interval 60000

The following example shows how to return the refresh interval to the default value of 30 seconds:

Router(config)# no ip rsvp signalling refresh interval

Related Commands

ip rsvp signalling refresh misses

To specify the number of successive refresh messages that can be missed before Resource Reservation Protocol (RSVP) removes a state from the database, use the ip rsvp signalling refresh misses command in global configuration mode. To return the missed refresh limit to its default value, use the no form of this command.

ip rsvp signalling refresh misses msg-count

no ip rsvp signalling refresh misses

Syntax Description

Command Default

4 messages

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling refresh misses command to specify the number of successive refresh messages that can be missed before RSVP regards the router state as expired and removes that state from the database.

Note If you change the default value for this command and you are also using the ip rsvp signalling hello graceful-restart refresh misses command, ensure that the value for the ip rsvp signalling hello graceful-restart refresh misses command is less than the value for the ip rsvp signalling refresh misses command. Otherwise, some or all of the label-switched paths (LSPs) may not be recovered after a stateful switchover (SSO) has occurred. We recommend that the value for the ip rsvp signalling refresh misses command be twice the value for the ip rsvp signalling hello graceful-restart refresh misses command.

Examples

The following example shows how to specify a missed refresh limit of 6 messages:

Router(config)# ip rsvp signalling refresh misses 6

The following example shows how to return the refresh misses limit to the default value of 4:

Router(config)# no ip rsvp signalling refresh misses

Related Commands

ip rsvp signalling refresh reduction

To enable Resource Reservation Protocol (RSVP) refresh reduction, use the ip rsvp signalling refresh reduction command in global configuration mode. To disable refresh reduction, use the no form of this command.

ip rsvp signalling refresh reduction

no ip rsvp signalling refresh reduction

Syntax Description

This command has no arguments or keywords.

Defaults

Disabled

Command Modes

Global configuration (config)

Command History

Usage Guidelines

RSVP refresh reduction is a set of extensions to reduce the messaging load imposed by RSVP and to help it scale to support larger numbers of flows.

The following features of the refresh reduction standard (RFC 2961) are supported and will be turned on with this command:

•Setting the refresh-reduction-capable bit in message headers

•Message-Identifier (ID) usage

•Reliable messaging with rapid retransmit, acknowledgement (ACK) messages, and MESSAGE_ID objects

•Summary refresh extension

•Bundle messages (reception only)

Refresh reduction requires the cooperation of the neighbor to operate; for this purpose, the neighbor must also support the standard. If the router detects that a directly connected neighbor is not supporting the refresh reduction standard (either through observing the refresh-reduction-capable bit in messages received from the next hop, or by sending a MESSAGE_ID object to the next hop and receiving an error), refresh reduction will not be used on this link irrespective of this command.

Examples

The following example shows how to enable RSVP refresh reduction:

Router(config)# ip rsvp signalling refresh reduction

The following example shows how to disable RSVP refresh reduction:

Router(config)# no ip rsvp signalling refresh reduction

Related Commands

ip rsvp signalling refresh reduction ack-delay

To configure the maximum amount of time that a Resource Reservation Protocol (RSVP)-configured router holds on to an acknowledgment (ACK) message before sending it, use the ip rsvp signalling refresh reduction ack-delay command in global configuration mode. To reset the ack-delay value to its default, use the no form of this command.

ip rsvp signalling refresh reduction ack-delay delay-value

no ip rsvp signalling refresh reduction ack-delay

Syntax Description

Defaults

The default value is 250 ms (0.25 sec).

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp signalling refresh reduction ack-delay command to configure the maximum amount of time that an RSVP-configured router keeps an ACK message before sending it.

Examples

The following example shows how to set the ack-delay value to 1 second:

Router(config)# ip rsvp signalling refresh reduction ack-delay 1000

The following example shows how to set the ack-delay value to the default value:

Router(config)# no ip rsvp signalling refresh reduction ack-delay

ip rsvp source

To configure a Resource Reservation Protocol (RSVP) router to populate an address other than the native interface address in the previous hop (PHOP) address field of the PHOP object when forwarding a PATH message onto that interface, use the ip rsvp source command in interface configuration mode. To keep the native interface address in the PHOP address field, use the no form of this command.

ip rsvp source {address ip-address | interface type number}

no ip rsvp source

Syntax Description

Command Default

The native interface address is written in the PHOP address field.

Command Modes

Interface configuration (config-if)

Command History

Examples

The following example shows how to configure the IP address 10.1.3.13 for the PHOP address field:

Router# configure terminal

Enter configuration commands, one per line.  End with CNTL/Z.

Router(config)# interface ethernet 0/0

Router(config-if)# ip rsvp bandwidth

Router(config-if)# ip rsvp source address 10.1.3.13

Router(config-if)# end

The following example shows how to configure the loopback interface 0 as the interface whose address is used in the PHOP address field:

Router# configure terminal

Enter configuration commands, one per line.  End with CNTL/Z.

Router(config)# interface ethernet 1/0

Router(config-if)# ip rsvp bandwidth

Router(config-if)# ip rsvp source interface loopback 0

Router(config-if)# end

Related Commands

ip rsvp svc-required

To enable creation of a switched virtual circuit (SVC) to service any new Resource Reservation Protocol (RSVP) reservation made on the interface or subinterface of an Enhanced ATM port adapter (PA-A3), use the ip rsvp svc-required command in interface configuration mode. To disable SVC creation for RSVP reservations, use the no form of this command.

ip rsvp svc-required

no ip rsvp svc-required

Syntax Description

This command has no arguments or keywords.

Command Default

Disabled

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

This command applies exclusively to the RSVP-ATM QoS Interworking feature.

Usually reservations are serviced when RSVP classifies packets and a queueing mechanism schedules them for transmission to manage congestion. Traditionally, RSVP is used with weighted fair queueing (WFQ). When RSVP is coupled with WFQ, all of the packets visible to WFQ are also visible to RSVP, which allows RSVP to identify and take action on packets important to it. In this case, WFQ provides bandwidth guarantees.

However, when the ip rsvp svc-required command is used to configure an interface or subinterface, a new SVC is established and used to service each new reservation on the interface. ATM SVCs are used to provide bandwidth guarantees and NetFlow is used on input interfaces to make data packets visible to RSVP.

Note When RSVP is enabled, all packets are processed by the Route Switch Processor (RSP).

This command must be executed on both ends of an SVC driven by RSVP. This command is supported only for the Enhanced ATM port adapter (PA-A3) and its subinterfaces.

Note For this command to take effect, NetFlow must be enabled. Therefore, the ip route-cache flow command must precede this command in the configuration.

Use the show ip rsvp interface command to determine whether this command is in effect for any interface or subinterface.

Examples

The following example shows how to signal RSVP that reservations made on ATM interface 2/0/0 will be serviced by creation of an SVC:

interface atm2/0/0

 ip rsvp svc-required

Related Commands

ip rsvp tos

To enable the router to mark the five low-order type of service (ToS) bits of the IP header ToS byte for packets in a Resource Reservation Protocol (RSVP) reserved path using the specified values for traffic that either conforms to or exceeds the RSVP flowspec, use the ip rsvp tos command in interface configuration mode. To remove existing settings for the ToS bits, use the no form of this command; if neither the conform nor exceed keyword is specified, all settings for the ToS bits are removed.

ip rsvp tos {conform tos-value} {exceed tos-value}

no ip rsvp tos [conform] [exceed]

Syntax Description

Command Default

The ToS bits of the ToS byte are left unmodified when this command is not used. (The default behavior is equivalent to use of the no ip rsvp tos command.)

Command Modes

Interface configuration

Command History

Usage Guidelines

Packets in an RSVP reserved path are divided into two classes: those that conform to the reservation flowspec and those that correspond to a reservation but that exceed, or are outside, the reservation flowspec.

The ip rsvp tos command allows you to set the ToS values to be applied to packets belonging to these two classes. You must specify the ToS value for at least one class of traffic when you use this command. You can use a single instance of the command to specify values for both classes, in which case you can specify the conform and exceed keywords in either order.

As part of its input processing, RSVP uses the ip rsvp tos command configuration to set the ToS bits of the ToS byte on conforming and nonconforming packets. If per-virtual circuit (VC) VIP-distributed Weighted Random Early Detection (DWRED) is configured, the system uses the ToS bit and IP Precedence bit settings on the output interface in its packet drop process. The ToS bit and IP Precedence bit settings of a packet can also be used by interfaces on downstream routers.

Execution of the ip rsvp tos command causes ToS bit values for all preexisting reservations on the interface to be modified.

Note RSVP must be enabled on an interface before you can use this command; that is, use of the ip rsvp bandwidth command must precede use of the ip rsvp tos command. RSVP cannot be configured with VIP-distributed Cisco Express Forwarding (dCEF).

Note The ip rsvp tos command sets bits 0 to 4 so that in combination with the IP Precedence bit settings every bit in the ToS byte is set. Use of these bits is made with full knowledge of the fact that certain canonical texts that address the ToS byte specify that only bits 1 to 4 are used as the ToS bits.

RSVP receives packets from the underlying forwarding mechanism. Therefore, to use the ip rsvp tos command to set the ToS bits, one of the following features is required:

•Weighted fair queueing (WFQ) must be enabled on the interface.

•RSVP switched virtual circuits (SVCs) must be used.

•NetFlow must be configured to assist RSVP.

Note Use of the no form of this command is not equivalent to giving the ip rsvp tos 0 command, which sets all precedence on the packets to 0, regardless of previous precedence setting.

Examples

The following example sets the ToS bits value to 4 for all traffic on ATM interface 1 that conforms to the RSVP flowspec. ToS bits on packets exceeding the flowspec are not altered.

interface atm1

 ip rsvp tos conform 4

Related Commands

ip rsvp transport

To create a Resource Reservation Protocol (RSVP) transport session, use the ip rsvp transport command in global configuration mode. To disable the RSVP transport session, use the no form of this command.

ip rsvp transport {client client-id | statistics}

no ip rsvp transport {client client-id | statistics}

Syntax Description

Command Default

RSVP is configured as transport protocol.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

You can use the ip rsvp transport command to configure RSVP to be used as transport mechanism for the clients. The client-id is used for identification of the client that initiates the RSVP as a transport protocol. The statistics keyword is used to record statistics for RSVP TP sessions. The statistics recorded includes information passed by RSVP to the RSVP TP client as part of callback. The maximum amount of information that can be recorded is 32 MB.

The ip rsvp transport command enables a router to simulate a host generating RSVP PATH message. This command is used for testing and debugging purposes.

Examples

The following example shows how to identify a client to establish an RSVP transport session:

Router> enable

Router# configure terminal

Router(config)# ip rsvp transport client 12

Related Commands

ip rsvp transport sender-host

To register a transport client ID with Resource Reservation Protocol (RSVP), use the ip rsvp transport sender-host command in global configuration mode. To disable the static RSVP host path configuration, use the no form of this command.

ip rsvp transport sender-host [tcp | udp] destination-address source-address ip-protocol dest-port source-port client-id init-id instance-id [vrf vrf-name] [data data-value]

no ip rsvp transport sender-host [tcp | udp] destination-address source-address ip-protocol dest-port source-port client-id init-id instance-id [vrf vrf-name] [data data-value]

Syntax Description

Command Default

The static RSVP host path is configured.

Command Modes

Global configuration (config)

Command History

Usage Guidelines

Use the ip rsvp transport sender-host command to configure the RSVP transport proxy path. When this command is configured, RSVP sends PATH messages downstream.

Examples

The following example shows how to configure an RSVP sender host path:

Router> enable

Router# configure terminal

Router(config)# ip rsvp transport sender-host 10.1.1.1 10.2.1.1 2 3 4 3 192.168.1.2 2 vrf 
vrf1 data d1

Related Commands

ip rsvp tunnel overhead-percent

To manually override the Resource Reservation Protocol (RSVP) percentage bandwidth, use the ip rsvp tunnel overhead-percent command in interface configuration mode. To restore the tunnel overhead percentage to its default values, use the no form of this command.

ip rsvp tunnel overhead-percent percentage

no ip rsvp tunnel overhead-percent

Syntax Description

Command Default

The percentage overhead for generic routing encapsulation (GRE) or multipoint generic routing encapsulation (mGRE) interfaces is 4 percent. The percentage overhead for GRE and mGRE with IPsec interfaces ranges from 4 to 15 percent, with an average of 10 percent.

Command Modes

Interface configuration mode (config-if)

Command History

Usage Guidelines

During the bandwidth admission control, the Cisco IOS software must consider the additional IP overhead introduced because of tunneling and a possible encryption over these tunnels. The default values for the overhead depends on the average size of an Internet packet. However, you can manually override the default values by using the ip rsvp tunnel overhead-percent command.

For example, when the Cisco IOS software gets a reservation request for 100 bytes, and if the outbound interface is a GRE or an mGRE interface, then a bandwidth reservation request for 104 bytes is made available locally on that tunnel interface. In case the GRE or mGRE interface is in protected mode, 110 bytes is requested on the respective link. This IP overhead does not affect the bandwidth signaled via RSVP.

Examples

The following example shows how to configure the router to manually override the percentage bandwidth:

Router(config)# interface tunnel 1

Router(config-if)# ip rsvp tunnel overhead-percent 20

Related Commands

ip rsvp udp-multicasts

To instruct the router to generate User Datagram Protocol (UDP)-encapsulated Resource Reservation Protocol (RSVP) multicasts whenever it generates an IP-encapsulated multicast packet, use the ip rsvp udp-multicasts command in interface configuration mode. To disable this function, use the no form of this command.

ip rsvp udp-multicasts [multicast-address]

no ip rsvp udp-multicasts [multicast-address]

Syntax Description

Command Default

The generation of UDP multicasts is disabled. If a system sends a UDP-encapsulated RSVP message to the router, the router begins using UDP for contact with the neighboring system. The router uses multicast address 224.0.0.14 and starts sending to UDP port 1699. If the command is entered with no specifying multicast address, the router uses the same multicast address.

Command Modes

Interface configuration

Command History

Usage Guidelines

Use this command to instruct a router to generate UDP-encapsulated RSVP multicasts whenever it generates an IP-encapsulated multicast packet. Some hosts require this trigger from the router.

RSVP cannot be configured with VIP-distributed Cisco Express Forwarding (dCEF).

Examples

The following example shows how to reserve up to 7500 kbps on Ethernet interface 2, with up to 1 Mbps per flow. The router is configured to use UDP encapsulation with the multicast address 224.0.0.14:

interface ethernet 2

 ip rsvp bandwidth 7500 1000

 ip rsvp udp-multicasts 224.0.0.14

Related Commands

ip rtp compression-connections

To specify the total number of Real-Time Transport Protocol (RTP) header compression connections that can exist on an interface, use the ip rtp compression-connections command in interface configuration mode. To restore the default value, use the no form of this command.

ip rtp compression-connections number

no ip rtp compression-connections

Syntax Description

Command Default

For PPP and High-Level Data Link Control (HDLC) interfaces, the default is 16 compression connections.

For Frame Relay interfaces, the default is 256 compression connections.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You should configure one connection for each RTP call through the specified interface.

Each connection sets up a compression cache entry, so you are in effect specifying the maximum number of cache entries and the size of the cache. Too few cache entries for the specified interface can lead to degraded performance, and too many cache entries can lead to wasted memory.

Note Both ends of the serial connection must use the same number of cache entries.

Examples

The following example changes the number of RTP header compression connections supported to 150:

Router> enable

Router# configure terminal

Router(config)# interface Serial1/0.0

Router(config-if)# encapsulation ppp

Router(config-if)# ip rtp header-compression

Router(config-if)# ip rtp compression-connections 150

Router(config-if)# end

Related Commands

ip rtp header-compression

To enable Real-Time Transport Protocol (RTP) header compression, use the ip rtp header-compression command in interface configuration mode. To disable RTP header compression, use the no form of this command.

ip rtp header-compression [passive | iphc-format | ietf-format] [periodic-refresh]

no ip rtp header-compression [passive | iphc-format | ietf-format] [periodic-refresh]

Syntax Description

Command Default

Disabled

For PPP interfaces, the default format for header compression is the IPHC format.

For High-Level Data Link Control (HDLC) and Frame Relay interfaces, the default format for header compression is the original proprietary Cisco format. The maximum number of compression connections for the proprietary Cisco format is 256.

Command Modes

Interface configuration (config-if)

Command History

Usage Guidelines

You can compress IP/User Datagram Protocol (UDP)/RTP headers to reduce the size of your packets. Compressing headers is especially useful for RTP because RTP payload size can be as small as 20 bytes, and the uncompressed header is 40 bytes.

The passive Keyword

By default, the ip rtp header-compression command compresses outgoing RTP traffic. If you specify the passive keyword, outgoing RTP traffic is compressed only if incoming RTP traffic on the same interface is compressed. If you do not specify the passive keyword, all outgoing RTP traffic is compressed.

The passive keyword is ignored on PPP interfaces. PPP interfaces negotiate the use of header-compression, regardless of whether the passive keyword is specified. Therefore, on PPP interfaces, the passive keyword is replaced by the IPHC format, the default format for PPP interfaces.

The iphc-format Keyword

The iphc-format keyword indicates that the IPHC format of header compression that will be used. For PPP and HDLC interfaces, when the iphc-format keyword is specified, TCP header compression is also enabled. For this reason, the ip tcp header-compression command appears in the output of the show running-config command. Since both RTP header compression and TCP header compression are enabled, both UDP packets and TCP packets are compressed.

The iphc-format keyword includes checking whether the destination port number is even and is in the ranges of 16,385 to 32,767 (for Cisco audio) or 49,152 to 65,535 (for Cisco video). Valid RTP packets that meet the criteria (that is, the port number is even and is within the specified range) are compressed using the compressed RTP packet format. Otherwise, packets are compressed using the less-efficient compressed non-TCP packet format.

The iphc-format keyword is not available for interfaces that use Frame Relay encapsulation.

Note The header compression format (in this case, IPHC) must be the same at both ends of the network. That is, if you specify the iphc-format keyword on the local router, you must also specify the iphc-format keyword on the remote router.

The ietf-format Keyword

The ietf-format keyword indicates that the IETF format of header compression will be used. For HDLC interfaces, the ietf-format keyword compresses only UDP packets. For PPP interfaces, when the ietf-format keyword is specified, TCP header compression is also enabled. For this reason, the ip tcp header-compression command appears in the output of the show running-config command. Since both RTP header compression and TCP header compression are enabled, both UDP packets and TCP packets are compressed.

With the ietf-format keyword, any even destination port number higher than 1024 can be used. Valid RTP packets that meet the criteria (that is, the port number is even and is higher than 1024) are compressed using the compressed RTP packet format. Otherwise, packets are compressed using the less-efficient compressed non-TCP packet format.

The ietf-format keyword is not available for interfaces that use Frame Relay encapsulation.

Note The header compression format (in this case, IETF) must be the same at both ends of the network. That is, if you specify the ietf-format keyword on the local router, you must also specify the ietf-format keyword on the remote router.

Support for Serial Lines

RTP header compression is supported on serial lines using Frame Relay, HDLC, or PPP encapsulation. You must enable compression on both ends of a serial connection.

Unicast or Multicast RTP Packets

This command can compress unicast or multicast RTP packets, and, hence, multicast backbone (MBONE) traffic can also be compressed over slow links. The compression scheme is beneficial only when you have small payload sizes, as in audio traffic.

Custom or Priority Queueing

When you use the ip rtp header-compression command and configure custom or priority queueing on an encapsulated HDLC or Frame Relay interface, the compressed packets may go to the default queue instead of the user-defined queue, which results in protocol flaps (loss of keepalives). Therefore, we recommend that you use the Modular Quality of Service (QoS) Command-Line Interface (CLI) (MQC) model for configuring QoS features.

Examples

The following example shows how to enable RTP header compression on the Serial1/0 interface and limits the number of RTP header compression connections to 10. In this example, the optional iphc-format keyword of the ip rtp header-compression command is specified:

Router> enable

Router# configure terminal

Router(config)# interface Serial1/0

Router(config-if)# encapsulation ppp

Router(config-if)# ip rtp header-compression iphc-format

Router(config-if)# ip rtp compression-connections 10

Router(config-if)# end

The following example shows how to enable RTP header compression on the Serial2/0 interface and limits the number of RTP header compression connections to 20. In this example, the optional ietf-format keyword of the ip rtp header-compression command is specified:

Router> enable

Router# configure terminal

Router(config)# interface Serial2/0

Router(config-if)# encapsulation ppp

Router(config-if)# ip rtp header-compression ietf-format

Router(config-if)# ip rtp compression-connections 20

Router(config-if)# end

The following example shows how to enable RTP header compression on the Serial1/0 interface and the optional periodic-refresh keyword of the ip rtp header-compression command is specified:

Router> enable

Router# configure terminal

Router(config)# interface Serial1/0

Router(config-if)# encapsulation ppp

Router(config-if)# ip rtp header-compression iphc-format periodic-refresh

Router(config-if)# ip rtp compression-connections 10

Router(config-if)# end

Related Commands

ip rtp priority

Note Effective with Cisco IOS XE Release 2.6, Cisco IOS Release 15.0(1)S, and Cisco IOS Release 15.1(3)T, the ip rtp priority command is hidden. Although this command is still available in Cisco IOS software, the CLI interactive Help does not display it if you attempt to view it by entering a question mark at the command line.

This command will be completely removed in a future release, which means that you will need to use the appropriate replacement command (or sequence of commands). For more information (including a list of replacement commands), see the Legacy QoS Command Deprecation feature document in the Cisco IOS XE Quality of Service Solutions Configuration Guide or the Legacy QoS Command Deprecation feature document in the Cisco IOS Quality of Service Solutions Configuration Guide.

Note Effective with Cisco IOS XE Release 3.2S, the ip rtp priority command is replaced by a modular QoS CLI (MQC) command (or sequence of MQC commands). For the appropriate replacement command (or sequence of commands), see the Legacy QoS Command Deprecation feature document in the Cisco IOS XE Quality of Service Solutions Configuration Guide.

To reserve a strict priority queue for a set of Real-Time Transport Protocol (RTP) packet flows belonging to a range of User Datagram Protocol (UDP) destination ports, use the ip rtp priority command in interface configuration mode. To disable the strict priority queue, use the no form of this command.

ip rtp priority starting-rtp-port-number port-number-range bandwidth

no ip rtp priority

Syntax Description

Command Default

Disabled

Command Modes

Interface configuration (config-if)

Command History