Product guide

Router Redundancy Using XRRP
Overview of XRRP Operation
The Solution. Beginning with software release E.09.05, you can optionally
configure XRRP infinite fail-back, which blocks automatic fail-back as long
as the fail-back router continues XRRP operation with at least one of its XRRP
VLANs remaining up. In this mode, the fail-back router maintains “permanent”
primary and secondary address control. This means that recovery of the fail-
over router does not automatically result in a fail-back from its peer, and can
only occur when either a system operator uses the CLI to force fail-back or
there is a system change affecting the fail-back router. (Events that can cause
a fail-back to occur are described under “Router Operation in the Infinite Fail-
Back Mode” on page 12-14.)
Causes of Fail-Over and Fail-Back
Fail-Over. An XRRP router fail-over to its peer occurs in these instances:
■ The router loses connectivity on any XRRP VLAN (and the peer router
has maintained connectivity with all of its XRRP VLANs).
■ None of an XRRP router’s multicast advertisements are detected by the
peer router within three advertisement intervals (and the peer router has
maintained connectivity with at least one of its XRRP VLANs).
Note If a peer router is unable to support a fail-over due to its own failures, the fail-
over does not occur and routing will cease in at least some areas of the
network.
Fail-Back. Without infinite fail-back enabled, an XRRP pee r router with both
primary and secondary control in a protection domain automatically initiates
a fail-back when it detects XRRP advertisements indicating that the failed
router has recovered access to all of its XRRP VLANs.
With infinite fail-back enabled in the protection domain, a change in the state
of the failed router does not initiate a fail-back.
12-12