[c-nsp] BFD bug in IOS SXF6

[Rodney Dunn]

Guys,

Can you please try any BFD test on 12.2(33)SRA latest code and not SXF?

There were some large changes to BFD and scheduling to improve this
and they will not be back ported to 12.2(18)SXF.

Rodney




On Fri, Jan 12, 2007 at 01:07:43PM +0100, airflow wrote:
> Byrne, Cameron wrote:
> 
> Hi all, I'm new to this mailinglist.
> 
> > Long story short, BFD does not work well in SXF6 because of CPU
> > scheduler issues. I have a TAC case, but it is not going well. 
> > Does anyone have  timer settings that they feel are stable?  I am
> 
> I also had massive problems with this feature and also had a SR open regarding this. The outcome was unsatisfying, as there is just no value which will not trigger a "false positive" under every circumstance. The engineer said that you have to experiment in the live network or set up a lab which very much resembles your live network in every aspect and test there. It is also recommended to configure "process-max-time 20". The default value for that is 200msecs, which already covers 2-3 multipliers of BFD at 100msecs.
> 
> In the end we ended up with disabling BFD. If you decide to use it, I would recommend to use conservative timing (300msecs intervals).
> 
> Regards,
> Florian
> 
> -- 
> Homepage: http://fp.ath.cx/                         PGP-keyID: C9FEDBA2
> _______________________________________________
> 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/