How long before OSPF route cleared?

[Colin Higgins]

I have a router that uses OSPF to determine routes to a central location. This same router also has a static route with a higher administrative distance on it, pointed to a backup router.

ip route 172.25.19.0 255.255.255.0 192.168.105.2 250

in case the primary WAN link goes down.

I just saw the primary router lose its OSPF peer for about 1 second, maybe a little less. The router did not redirect to the backup. The peer came back up (apparently something happened in the cloud, because interfaces never went down), and everything went back to normal.

So the question is: how long does that peer and the route (172.25.19.0) need to be down before the primary router switches over the static route?

[Jon Marshall · Accepted solution]

Colin

It depends on the link eg

ethernet and point to point have a hello timer of 10 seconds and dead timer of 40 seconds. So if a hello has not been seen within the dead timer interval the OSPF neighborship is declared down

non broadcast network timers are 30 and 120 seconds and same rules apply.

So a 1 second blip is not going to affect this at all ie. the OSPF router may miss one hello if it happened to be sent at that exact moment but it isn't going to miss the next one.

If you want this to be a lot quicker and your IOS/platform supports it you can look at OPSF fast hellos but for a 1 second blip on a WAN link i don't think you would want to do this because it would flap between the backup and primary links -

http://www.cisco.com/c/en/us/td/docs/ios/12_0s/feature/guide/fasthelo.html

Jon

[Colin Higgins]

Perfect Jon

thanks!

[Jon Marshall]

Colin

Just to clarify i was talking about when the interface doesn't actually go down.

Jon

[Joseph W. Doherty]

Disclaimer

The   Author of this posting offers the information contained within this   posting without consideration and with the reader's understanding that   there's no implied or expressed suitability or fitness for any purpose.   Information provided is for informational purposes only and should not   be construed as rendering professional advice of any kind. Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In   no event shall Author be liable for any damages whatsoever (including,   without limitation, damages for loss of use, data or profit) arising  out  of the use or inability to use the posting's information even if  Author  has been advised of the possibility of such damage.

Posting

As Jon notes, OSPF relies on its hellos to timeout, if the link is up, but the neighbor isn't reachable.  (Physical link down should take out the OSPF route, on that device, within milliseconds.)

Default timeouts can be rather long, like up to 40 seconds.  (A long time to black hole traffic.)

Standard timers can be reduced, and some platforms support subsecond OSPF hellos (fast hellos), as noted by Jon.  Some platforms also support subseconds hellos riding on BFD.

If your running something like VoIP across such a link, would recommend you do look at reducing dead time for a neighbor.

Concerning Jon mention of a flapping interface, some platforms support IP dampening.

Also, Cisco's OSPF often has some knobs concerning how quickly a link changed is advertised to the rest of the topology.