[c-nsp] Troubleshooting High CPU

[e ninja]

*Response inline in italics...*

On Mon, Aug 31, 2009 at 9:45 PM, David Warner
<davidwarner1975 at yahoo.com.au>wrote:

> Hi - Thanks for the feedback.
>
> Yes - same hardware (3845), sw, config and traffic. Just deployed as
> active/standby for HSRP. The chief suspect is the port channel between the
> ESWs but would like to get the info to confirm whether its spanning tree or
> another culprit.
>
> Im unable to get any useful info out the box as when it occurs the CPU
> spikes and hits VTY so unable to get any commands back. Is there anyway to
> reserve a minimum amount of CPU to VTY so we can at least issue the commands
> while the problem is occuring?
>


*Yes. Use scheduler interval 1000 global config command. This will schedule
low priority processes to run every 1000 milliseconds (1s) and provide you
time to run some commands, even when the CPU is at 100%.*

-Eninja



> ------------------------------
> *From:* Eninja <eninja at gmail.com>
> *To:* David Warner <davidwarner1975 at yahoo.com.au>
> *Cc:* "cisco-nsp at puck.nether.net" <cisco-nsp at puck.nether.net>; Eninja <
> eninja at gmail.com>
> *Sent:* Tuesday, 1 September, 2009 12:11:07 PM
> *Subject:* Re: [c-nsp] Troubleshooting High CPU
>
> What platform is the device? Are the primary and standby devices the same -
> platform, SW, config, traffic?
>
> During the spike, is the CPU consumed at interrupt or process level?
>
> Grab and send over sh proc cpu, sh align, sh int stat, sh log (if device is
> too unresponsive during failover, have onsite personnel pull out
> traffic-laden cables one by one until device is responsive, grab captures
> before reinserting cable/s).
>
> -Eninja
> PS. Disable console logging
>
>
> On Aug 31, 2009, at 10:46 PM, David Warner <davidwarner1975 at yahoo.com.au>
> wrote:
>
> > Hi All,
> >
> > Just wondering if I can get some advice. We have two routers in a HSRP
> active/standby pair. If ones reloads then the CPU on the other hits 100% and
> crashes. The only way we can recover this is for a field engineer to pull
> all the Ethernet cables out and then, when system is calm, repatch them. We
> access the device via VTY so struggling to actually get any commands in to
> troubleshoot as per the Cisco 'Troubleshooting High CPU Utilization' doc as
> the CPU spikes .
> >
> > Any advice on best way to progress this?
> >
> > Regards, David
> >
> >
> >
> >
> __________________________________________________________________________________
>
>
>
> > Find local businesses and services in your area with Yahoo!7 Local.
> > Get started: http://local.yahoo.com.au
> > _______________________________________________
> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
> > https://puck.nether..net/mailman/listinfo/cisco-nsp<https://puck.nether.net/mailman/listinfo/cisco-nsp>
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
>
>
> ------------------------------
> Find local businesses and services in your area with Yahoo!7 Local. Get
> started<http://au.rd.yahoo.com/search/local/mailtagline/*http://local.yahoo.com.au>
> .
>