[c-nsp] Catching OSPF hello problems

[Rodney Dunn]

Not a good scalable one directly for OSPF.

What you could do instead though is setup a parallel IP SLA probe
running between the two boxes and trigger/alarm when you lose
X consecutive probes. It's just an IP packet.

Rodney

On Fri, Sep 29, 2006 at 04:22:07PM +0100, Jee Kay wrote:
> I currently have a not-entirely-right metro ethernet solution that
> seems to go unidirectional for a few seconds every so often. We've had
> to change our OSPF timers across these links to hello 1, dead 9 to
> stop OSPF flapping when this occurs.
> 
> (Un)fortunately, this means we no longer get big red alarmbells going
> off when the line does its pseudo-flap thing. We do however get logs
> that looks very much like the following:
> 
> Sep 26 07:20:20.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on
> FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:20.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:20.937 CDT: OSPF: Rcv hello from 216.210.122.174 area 0
> from FastEthernet2/1 x.y.z.a
> Sep 26 07:20:20.937 CDT: OSPF: End of hello processing
> Sep 26 07:20:21.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on
> FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:21.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:21.945 CDT: OSPF: Rcv hello from 216.210.122.174 area 0
> from FastEthernet2/1 x.y.z.a
> Sep 26 07:20:21.945 CDT: OSPF: End of hello processing
> Sep 26 07:20:22.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on
> FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:22.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:23.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on
> FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:23.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:24.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on
> FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:24.717 CDT: OSPF: End of hello processing
> 
> (I'm guessing gmail is going to wrap that to unreadable death, so have
> also attached it)
> 
> Basically what has happened here is that for 3 seconds we're sending
> hellos, but not receiving any. After this, the hellos start being
> received again and everyone is once again happy.
> 
> What I'm looking for is some way to watch for this happening (send 2
> or 3 consecutive hellos on the same interface without receiving one
> back) and alert on it.
> 
> Can anyone think of a relatively easy way to do this?
> 
> Thank,
> Ras

> Sep 26 07:20:20.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:20.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:20.937 CDT: OSPF: Rcv hello from 216.210.122.174 area 0 from FastEthernet2/1 x.y.z.a
> Sep 26 07:20:20.937 CDT: OSPF: End of hello processing
> Sep 26 07:20:21.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:21.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:21.945 CDT: OSPF: Rcv hello from 216.210.122.174 area 0 from FastEthernet2/1 x.y.z.a
> Sep 26 07:20:21.945 CDT: OSPF: End of hello processing
> Sep 26 07:20:22.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:22.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:23.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:23.717 CDT: OSPF: End of hello processing
> Sep 26 07:20:24.701 CDT: OSPF: Send hello to 224.0.0.5 area 0 on FastEthernet2/1 from x.y.z.b
> Sep 26 07:20:24.717 CDT: OSPF: End of hello processing

> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/