03-20-2017 01:13 AM - edited 03-08-2019 09:49 AM
one of out router has frequent flapping issue, see below log:
Feb 17 01:18:57.855 UTC: %BGP-5-ADJCHANGE: neighbor 192.168.1.1 Up
Feb 17 02:00:57.513 UTC: %BGP-3-NOTIFICATION: sent to neighbor 192.168.1.1 4/0 (hold time expired) 0 bytes
Feb 17 02:00:57.513 UTC: %BGP-5-NBR_RESET: Neighbor 192.168.1.1 reset (BGP Not ification sent)
Feb 17 02:00:57.518 UTC: %BGP-5-ADJCHANGE: neighbor 192.168.1.1 Down BGP Notif ication sent
Feb 17 02:00:57.518 UTC: %BGP_SESSION-5-ADJCHANGE: neighbor 192.168.1.1 IPv4 U nicast topology base removed from session BGP Notification sent
Feb 17 11:01:08.504 UTC: %BGP-5-ADJCHANGE: neighbor 192.168.1.1 Up
Feb 18 03:20:38.940 UTC: %BGP-3-NOTIFICATION: sent to neighbor 192.168.1.1 4/0 (hold time expired) 0 bytes
Feb 18 03:20:38.941 UTC: %BGP-5-NBR_RESET: Neighbor 192.168.1.1 reset (BGP Not ification sent)
Feb 18 03:20:38.945 UTC: %BGP-5-ADJCHANGE: neighbor 192.168.1.1 Down BGP Notif ication sent
Feb 18 03:20:38.945 UTC: %BGP_SESSION-5-ADJCHANGE: neighbor 192.168.1.1 IPv4 U nicast topology base removed from session BGP Notification sent
Feb 18 03:42:48.727 UTC: %BGP-5-ADJCHANGE: neighbor 192.168.1.1 Up
but we already has bgp dampening configuration, in the past four years, we don't have such issue, recently two days, we have such issue, in this case, how to fix the issue by ourselves, any action we should take? this is within our core network, nothing related to service provider
03-20-2017 01:22 AM
Whats the physical interface show where it's dropping is there any errors ?
03-20-2017 03:11 AM
No physical interface down
03-20-2017 03:17 AM
Not that it was down the error count I was asking , if the interface is receiving errors , crcs , inputs etc it will cause it to flap as it could be dropping the TCP packets or hello packets, the physical interface can still be up it does not have to fail for BGP to fail
also if there is recursive routing in place that will also cause constant flapping , debug ip routing will show you in logs if there is recursive going on
run an ip sla on the link as well make sure its stable , check MTU has not changed on either side of link
03-21-2017 08:17 PM
Hi Mark, thanks, one more information, the bgp flap is happened in DMVPN stub, 192.168.1.1 is the hub, in stub side I check ip nhrp, the tunnel is created mny month ago with correct NBMA address.
hub and spoke connected by BGP only.
stub is IOS-XE, so no command to show dmvpn, why this bgp flap again and again?
03-22-2017 04:22 AM
show dmvpn works in ios-xe , I have IWAN running with DMVPN on ASRs and 4331s so the command should work for you
Really need to see some outputs here from both spoke side , interface outputs and show ip bgp neighbour output , I presume the config is ok if it was working fine before and nothing has changed and this only started
There could be multiple reasons its happening , flaky circuits , dodgy sfp , bad mux on ISP side etc , need to troublshoot the outputs and what it is showing to be able to identify the issue
is the connection over mpls or dsl between the hub and spoke
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide