My BGP peering with ISP is getting flapped frequently. This flapping will be there for few mins again the circuit comes up and works in a normal way.
can some one help me in finding the root cause for this isse.
Pls provide some more information, how is it connected, had this worked before or a new issue ???
we have provider mux at our premises and it is terminated with RJ 45. The problem has been occuring after a week, the ckt got commissioned..
in show ip bgp summary it is showing
sh ip bgp summary
BGP router identifier 10.X.X.X, local AS number 65433
BGP table version is 306620, main routing table version 306620
170 network entries using 20400 bytes of memory
289 path entries using 15028 bytes of memory
46/25 BGP path/bestpath attribute entries using 5704 bytes of memory
12 BGP AS-PATH entries using 288 bytes of memory
2 BGP community entries using 48 bytes of memory
63 BGP route-map cache entries using 2016 bytes of memory
0 BGP filter-list cache entries using 0 bytes of memory
BGP using 43484 total bytes of memory
Dampening enabled. 0 history paths, 0 dampened paths
BGP activity 12401/12231 prefixes, 94856/94567 paths, scan interval 60 secs
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/PfxRcd
10.12.201.58 4 65433 475235 479087 306620 0 0 7w2d 118
172.16.0.22 4 65213 422 421 306620 0 0 01:05:11 116
would like to know, what causes the BGP to flap frequently? my physical interface shows UP. will there be any configuration issue with BGP attributes?
Here is one of the cause of BGP flap
Below is the log when my BGP peering is down. Thanks for the URL you shared with me..
000678: .Sep 3 09:52:20.324 UTC: %BGP-5-ADJCHANGE: neighbor 172.16.0.22 Down BGP Notification sent
000679: .Sep 3 09:52:20.324 UTC: %BGP-3-NOTIFICATION: sent to neighbor 172.16.0.22 4/0 (hold time expired) 0 bytes
000680: .Sep 3 09:52:20.324 UTC: %BGP_SESSION-5-ADJCHANGE: neighbor 172.16.0.22 IPv4 Unicast topology base removed from session BGP Notification sent
000681: .Sep 3 09:52:49.068 UTC: %BGP-5-ADJCHANGE: neighbor 172.16.0.22 Up
You could run an IPSLA and see how often you loose connectivity to your WAN peer IP. (google it) If the frequency of the flaps keep incrementing, eventually BGP will dampen your networks and won't get advertised upstream by your ISP. If it is a physical layer issue, your Service Provider should be able to do a Layer 1 test and verify where in the path lies the issue.