| 摘要 |
Graceful/hitless restart features from a routing/signaling protocol may be utilized along with synchronization between cluster members in order to provide routing capabilities in a clustering environment. During normal operation, the active cluster member may operate the routing protocol(s) and communicate with the neighboring components using the cluster's address. As the active member learns the routes to the destinations, the routing data may be propagated to the stand-by cluster members across an internal communication mechanism. Configuration information for the routing component may also be propagated to the stand-by cluster members. Upon failure of the active routing component, the standby routing component may be started up such that neighboring routing components located outside of said cluster do not recalculate network topology based on said failure. This allows for failover to occur without affecting neighboring components and unduly burdening the system. ® KIPO & WIPO 2007
|
