[c-nsp] BFD on 12.2(33)

alaerte.vidali at nsn.com alaerte.vidali at nsn.com
Mon May 21 13:12:30 EDT 2007


Pascal,

Did you measure and could share the recovery time you got using 250ms on
BFD timers?

Tks,
Alaerte


========================================================================
==============================

ascal Gloor pascal.gloor at spale.com
<mailto:cisco-nsp%40puck.nether.net?Subject=%5Bc-nsp%5D%20BFD%20on%2012.
2%2833%29&In-Reply-To=84FB94BF572F554792F1BD88268F0FA916D22F%40NGT-BGO-E
X-01.nextgentel.corp> 
Wed May 16 05:27:53 EDT 2007 
*	Previous message: [c-nsp] BFD on 12.2(33)  <040574.html> 
*	Next message: [c-nsp] BFD on 12.2(33)  <040584.html> 
*	Messages sorted by: [ date ] <date.html>  [ thread ]
<thread.html>  [ subject ] <subject.html>  [ author ] <author.html>  

Hi Geir,

>> Does anyone have any experience with BFD on 12.2(33)SR?
>> We have troubles with BFD in a mixed 12.2(18)SXF7 and 12.2(33)SRB  
>> environment.
>> BFD will occasionally toggle our 10G interfaces  when we do "write  
>> mem" or shut/no shut, and I'd like to verify that the problem is  
>> caused  by the routers running 12.2(18)SXF7.  If so I can upgrade  
>> all routers to 12.2(33)SRB. Otherwise I'll have to consider  
>> running ospf fast hello.

The "shut/no shut/wr" bug is already known. CSCsc96524[1]. The bug id  
status is 'terminated/closed", however it doesnt seem to be fixed.  
Pretty funny that you sent this today, we were just testing SXF/SRB  
BFD compatibility.

So far, the only workaround we have found is to set the BDF timers to  
250ms. We're running 250ms BFD on SXF since over a year without any  
issue.

As far as I remember 150ms was flapping, 200ms not, so we took 250ms  
to have some margin.

If this doesnt meet your requirements, I urge you to ask cisco to  
reopen this Bug since its absolutly not fixed.



Regards,
Pascal



More information about the cisco-nsp mailing list