Command Line Interface
1142 1553-KDU 137 365 Uen D 2006-06-16
receives the list of RP candidates from the BSR also elects an active RP
for each group range using the same election process.
• The election process for each group is based on the following criteria:
• Find all RPs with the most specific group range.
• Select those with the highest priority (lowest priority value).
• Compute hash value based on the group address, RP address, and
hash mask included in the bootstrap messages.
• If there is a tie, use the candidate RP with the highest IP address.
• This distributed election process provides faster convergence and
minimal disruption when an RP fails. It also serves to provide load
balancing by distributing groups across multiple RPs. Moreover, when
an RP fails, the responsible RPs are re-elected on each router, and the
groups automatically distributed to the remaining RPs.
• To improve failover recovery, it is advisable to select at least two core
routers in diverse locations, each to serve as both a candidate BSR and
candidate RP. It is also preferable to set up one of these routers as both
the primary BSR and RP.
To Next Page
Loading...
