[c-nsp] slow convergence for full bgp table on a Cisco 7613/SUP720-3BXL
Emanuel Popa
emanuel.popa at gmail.com
Tue Mar 13 09:49:58 EST 2007
Here are some snapshots:
br01.frankfurt#sh proc cpu sorted | e 0.00
CPU utilization for five seconds: 52%/6%; one minute: 16%; five minutes: 16%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
326 1014963840 6595633 153885 37.71% 7.03% 7.11% 0 BGP Scanner
br01.frankfurt#sh proc cpu sorted | e 0.00
CPU utilization for five seconds: 78%/4%; one minute: 21%; five minutes: 17%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
326 1014966764 6595647 153885 57.11% 11.04% 7.94% 0 BGP Scanner
br01.frankfurt#sh proc cpu sorted | e 0.00
CPU utilization for five seconds: 78%/4%; one minute: 21%; five minutes: 17%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
326 1014966764 6595647 153885 57.11% 11.04% 7.94% 0 BGP Scanner
br01.frankfurt#sh proc cpu sorted | e 0.00
CPU utilization for five seconds: 5%/3%; one minute: 19%; five minutes: 16%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
123 4034002923513388775 0 1.51% 1.89% 1.92% 0 IP Input
324 63436592 119933896 528 0.15% 0.30% 0.42% 0 BGP Router
these are consequent outputs at no longer then a couple of seconds...
bye,
emanuel
On 3/13/07, Gert Doering <gert at greenie.muc.de> wrote:
> Hi,
>
> On Tue, Mar 13, 2007 at 04:20:59PM +0200, Emanuel Popa wrote:
> > I have attached an image with the cpu load graphs.
>
> Actually I didn't want to see pretty pictures, but wanted to see *which*
> process is causing the CPU load...
>
> This could be "IP input" (baaaad), "BGP transport" (hints at packet
> loss), "BGP Router" (hmmm), or something else which is drowning out your
> CPU so much that there is no time left for actual BGP update processing...
>
> gert
> --
> USENET is *not* the non-clickable part of WWW!
> //www.muc.de/~gert/
> Gert Doering - Munich, Germany gert at greenie.muc.de
> fax: +49-89-35655025 gert at net.informatik.tu-muenchen.de
>
More information about the cisco-nsp
mailing list