>Hi all , I have the below topologyR1 - R2 - R4R1 - R3 - R4The first path >operates in AS 12 and the second path operates in AS 13Now , I advertised >Loopback 0 on R4 (4.4.4.4/32) in both AS numbers , what I can see in R1 >IP routing table is only one path and in the topology table I can see two >entries but the second one which is the not selcted one shows FD >inacesible , I shut down the interface and I could learn the prefix via >the second one , activated again and as it preempted , what is the reason >behind selesting the first path ? all the links are similar
If I understand correctly, R1 learns the same prefix via two different routing sources (eigrp 12 and eigrp 13). IOS can only install the route from one routing source, so it needs to make a decision, and uses the administrative distance. If the two sources use the same distance, the behaviour is undefined, usually the first one wins. With two EIGRP processes with the same admin distance, competing for the installation, IOS selects the path received from the lower numbered AS/process ID (see http://www.cisco.com/en/US/tech/tk365/technologies_q_and_a_item09186a008012 dac4.shtml#eight2). oli P.S: I thought I heard that the selection also considers EIGRP composite metric (which is an exception for EIGRP as normally protocol metric is not considered when multiple processes with the same AD compete), but not sure about this.. _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
