From: Darren Pedley (darren@dnsl.com)
Date: Thu Feb 09 2006 - 10:56:36 GMT-3
Hi,
I have a network setup like so:
bb1 -------- r1
| |
| |
bb2 ---r2----r3
I have ospf running as my igp between r1,2,3
I have a confed_id of 1 and sub-as of 64512 running on r1
I have a confed_id or 1 and sub-as of 64513 running on r2,3
I have ibgp peering peering between r2 and r3.
I have ebgp peering (within the confed) between r3 and e1.
I have ebgp peering bettwen bb1 - r1 and bb2 - r2.
bb1 and bb2 are advertising the route 1.1.1.0.
Currently,
r2 and r3 have the entry 1.1.1.0 in the bgp table with two paths
r1 only has one entry from bb1
Now after playing for a while its not consistant, depending on order the routers r1 and r2 come up (I think) is where the troublesome bgp table will be. i.e. it was r2 before I wiped my configs and started again.
Now on R1 when the bgp session is reset the entry appears for approx. 30sec's, then I get the following:
"BGP(0): 2.2.2.2 rcv UPDATE about 1.1.1.0/24 -- withdrawn"
My understanding is r2 send a withdraw msg for some reason?
Is this a "feature" or is there something I'm missing?
Thanks all
Darren
This archive was generated by hypermail 2.1.4 : Wed Mar 01 2006 - 11:28:17 GMT-3