Configuring Redundant Pseudowires for Layer 2 Circuits and VPLS
A redundant pseudowire can act as a backup connection between PE routers and CE devices, maintaining Layer 2 circuit and VPLS services after certain types of failures. This feature can help improve the reliability of certain types of networks (metro for example) where a single point of failure could interrupt service for multiple customers. Redundant pseudowires cannot reduce traffic loss to zero. However, they provide a way to gracefully recover from pseudowire failures in such a way that service can be restarted within a known time limit.
For an overview of how redundant pseudowires work, see Redundant Pseudowires for Layer 2 Circuits and VPLS.
To configure pseudowire redundancy for Layer 2 circuits and VPLS, complete the procedures in the following sections:
Configuring Pseudowire Redundancy on the PE Router
You configure pseudowire redundancy on the PE router acting
as the egress for the primary and standby pseudowires using the backup-neighbor statement.
To configure pseudowire redundancy on the PE router, include
the backup-neighbor statement:
backup-neighbor { community name; psn-tunnel-endpoint address; standby; virtual-circuit-id number; }
For a list of hierarchy levels at which you can include this statement, see the statement summary for this statement.
The backup-neighbor statement includes the following
configuration options:
community—Specifies the community for the backup neighbor.psn-tunnel-endpoint—Specifies the endpoint address for the packet switched network (PSN) tunnel on the remote PE router. The PSN tunnel endpoint address is the destination address for the LSP on the remote PE router.standby—Configures the pseudowire to the specified backup neighbor as the standby. When you configure this statement, traffic flows over both the active and standby pseudowires to the CE device. The CE device drops the traffic from the standby pseudowire, unless the active pseudowire fails. If the active pseudowire fails, the CE device automatically switches to the standby pseudowire.virtual-circuit-id—Uniquely identifies the primary and standby Layer 2 circuits. This option is configurable for Layer 2 circuits only.
Configuring the Switchover Delay for the Pseudowires
To configure the time the router waits before switching traffic
from the failed primary pseudowire to a backup pseudowire, include
the switchover-delay statement:
switchover-delay milliseconds;
For a list of hierarchy levels at which you can include this statement, see the statement summary for this statement.
Configuring a Revert Time for the Redundant Pseudowire
You can specify a revert time for redundant Layer 2 circuit and VPLS pseudowires. When you have configured redundant pseudowires for Layer 2 circuits or VPLS, traffic is switched to the backup pseudowire in the event that the primary pseudowire fails. If you configure a revert time, when the configured time expires traffic is reverted back to the primary pseudowire, assuming the primary pseudowire has been restored.
To configure a revert time for redundant pseudowires, specify
the time in seconds using the revert-time statement:
revert-time (Protocols Layer 2 Circuits) seconds maximum seconds;
With the maximum option, specify a maximum reversion
interval to add after the revert-time delay. If a revert-time
delay is defined but a maximum timer is not defined, VCs are restored
upon the revert-timer's expiration.
To reduce as much as possible the amount of traffic discarded, and potential data-path asymmetries observed during primary-to-backup transition periods, you can use this restoration timer. This restoration timer is activated when the backup path is performing as active, and then the primary path is restored. The goal is to avoid moving traffic back to the primary path right away, to make sure that the control plane's related tasks (such as IGP, LDP, RSVP, and internal BGP) have enough time to complete their updating cycle.
By enabling a gradual return of traffic to the primary path, you can ensure that the relatively-slow control-plane processing and updating does not have a negative impact on the restoration process.
The maximum option extends the revert timer’s
functionality to provide a jittered interval over which a certain
number of circuits can be transitioned back to the primary path. By
making use of this maximum value, you can define a time interval during
which circuits are expected to switch over. As a consequence, circuits’
effective transitions are scattered during restoration periods.
When making use of revert-time x maximum y statement, you can ensure that the corresponding
circuit that is active is moved to the primary path within a time-slot
(t1) such as that: x <= t1 <= y. In other words, by activating
this statement, you can ensure the following:
VCs stay in the backup path for at least x seconds after the primary path comes back up.
VCs are moved back to the primary path before y seconds have elapsed.
y maximum value = x maximum value * 2 = 1200 seconds.
The ideal values for x and y will are conditioned to internal aspects of your network. For this reason, there are no default values for these settings. If no revert-time is set, the default behavior is non-revertive. That is, circuits are not returned to the primary path upon restoration. They are kept on the backup path.
For a list of hierarchy levels at which you can include this statement, see the statement summary for this statement.