From: Pavel Bykov (slidersv@gmail.com)
Date: Mon Dec 08 2008 - 07:55:34 ARST
While 2 is the standard time to consider something dead, 1 is a special case
for NSF.
#1 timer activates in case your router is NSF-AWARE and/or NSF-CAPABLE, and
neighbor is NSF-CAPABLE router, that has some sort of SSO redundancy, and
it's control plane went dead.
E.g. router R1 is NSF-AWARE and R2 is highly redundant system with SSO/NSF
(NSF-CAPABLE)
R2 Supervisor engine crashes, but dataplane is still ok. Using NSF
communication, R1 knows that it should break the neighborship, because even
though the neighbor apeears dead, it's not really. So it keeps forwarding
traffic, and waiting for R2's control plane to get up again. It waits for
that NSF #1 timer, and if it doesn't come up by then, then it breaks
neighborship. If it does, then NSF recovery takes place, everything
converges and with a bit of luck you lost almost 0 packets, even though R2's
control plane crashed
On Sun, Dec 7, 2008 at 3:11 PM, GAURAV MADAN <gauravmadan1177@gmail.com>wrote:
> Hi Group
>
> Simple query ..... As I read from doc cd ; I have a confusion in 2 CLIs :
>
> 1) timers nsf route-hold
> and
> 2) ip hold-time eigrp
>
> hold-time (3 time the hello by default) determines that after this interval
> of not hearing the hellos from neighbor ; routes from the neighbor are held
> invalid and hence removed from table
>
> timers nsf route-hold will also do the same thing ( after this time period
> ;
> the routes from inactive peer are removed )
>
> Can someone clearly stae the situation where these 2 should be used .
>
> Thnx
> Gaurav Madan.
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>
-- Pavel Bykov ---------------- Don't forget to help stopping the braindumps, use of which reduces value of your certifications. Sign the petition at http://www.stopbraindumps.com/Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Thu Jan 01 2009 - 12:53:08 ARST