Does anyone have an idea about the following problem:
I've got a 7206 VXR (12.0.12S SP) connected over Ethernet to a 3640 (12.0.11 SP) with an iBGP session.
The 3640 receives 5 routes from the 7206, every minute when the BGP table is scanned the routes
are removed, the next minute they are back again.
Oct 8 18:31:37.064 MEST: BGP: xxx.162.23.237 sending KEEPALIVE
Oct 8 18:31:37.064 MEST: BGP: xxx.162.23.237 KEEPALIVE rcvd
Oct 8 18:31:57.204 MEST: BGP: scanning routing tables
Oct 8 18:31:58.164 MEST: BGP: nettable_walker xxx.23.160.0/19 calling revise_route
Oct 8 18:31:58.164 MEST: BGP: revise route installing xxx..23.160.0/19 -> xxx.23.164.254
Oct 8 18:31:58.164 MEST: BGP: xxx.162.23.237 computing updates, neighbor version 546, table version 547, starting at 0.
0.0.0
Oct 8 18:31:58.164 MEST: BGP: xxx.162.23.237 update run completed, ran for 0ms, neighbor version 546, start version 547
, throttled to 547, check point net 0.0.0.0
Oct 8 18:32:37.080 MEST: BGP: xxx.162.23.237 sending KEEPALIVE
Oct 8 18:32:37.080 MEST: BGP: xxx.162.23.237 KEEPALIVE rcvd
Oct 8 18:32:57.213 MEST: BGP: scanning routing tables
Oct 8 18:32:57.217 MEST: BGP: no valid path for xxx.23.160.0/19
Oct 8 18:32:58.181 MEST: BGP: nettable_walker xxx.23.160.0/19 no best path
Oct 8 18:32:58.181 MEST: BGP: xxx.162.23.237 computing updates, neighbor version 547, table version 548, starting at 0.
0.0.0
Oct 8 18:32:58.181 MEST: BGP: xxx.162.23.237 update run completed, ran for 0ms, neighbor version 547, start version 548
, throttled to 548, check point net 0.0.0.0
None of the routes originate from either routers, they all come from other parts of the network and there is no flapping
on the rest of the network. This seems to be local to this iBGP session. I've checked internal routes (they don't move), the status
of the interfaces (ok) and tried running the session both from the ip address of the ethernet interface and from the loopback
(didn't change a thing)
Any ideas?
Thomas
This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:18 EDT