04-28-2020 03:52 PM
Hello,
We have recently mentioned that on our core production routers we are getting constant, almost exactly every 10 min the following OSPF-5-ADJCHG:
0337: .Apr 28 20:26:54.812: OSPF-5-ADJCHG: Process 7000, Nbr 188.143.143.12 on Vlan25 from LOADING to FULL, Loading Done
And about 10 min later:
0338: .Apr 28 20:35:34.599: %OSPF-5-ADJCHG: Process 7000, Nbr 188.143.143.12 on Vlan25 from FULL to DOWN, Neighbor Down: Dead timer expired
We have mentioned that couple of months ago the CPU on the core routers spiked from 10% to about 20% and this is when we started seeing those ADJ changes but no changes were made on either device or in the network at all.
The CPU process is showing that most resources are taken by "IP Input" and "BGP Router"
CPU utilization for five seconds: 15%/7%; one minute: 16%; five minutes: 16%
The CPU spike is the only inconsistency that we see correlates with the time when we started seeing %OSPF-5-ADJCHG.
Could you please advise if there is direction that we should look at to alleviate this issues?
Thank you!
04-28-2020 04:54 PM
@kvitug wrote:
0338: .Apr 28 20:35:34.599: %OSPF-5-ADJCHG: Process 7000, Nbr 188.143.143.12 on Vlan25 from FULL to DOWN, Neighbor Down: Dead timer expired
Check this side of the network and confirm the WAN link isn't flapping.
NOTE: Some people like to disable link logging (i. e. when a link goes down/up the event won't show up in the logs). If it is disabled, enable it.
04-29-2020 09:42 AM
04-29-2020 07:04 AM
Ping the neighbor device cont. and check if the unicast reach-ability fails.
04-29-2020 01:04 PM - edited 04-29-2020 01:05 PM
Hello
sounds like a possible interface/neighbour flap due possible spf recalculations
router ospf xx
ospf log-adjacency-changes
sh ip ospf | in SPF
sh log | in OSPF
debug ip ospf monitor
04-29-2020 02:19 PM
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