Mailing List Archive: Cisco: NSP

15.2(2)S issues on 7600/SUP720

Index | Next | Previous | View Threaded

jason at lixfeld

Apr 25, 2012, 11:58 AM

Post #1 of 3 (313 views)
Permalink

15.2(2)S issues on 7600/SUP720

I'm testing 15.2(2)S on a couple of 7600/SUP720 combos and I've experienced some odd behaviour and I'm curious about whether or not anyone else has seen this. It's been really hard to reproduce, but it does happen once in a while, so I'm hoping similar experiences might give me some TAC ammo.

Basically, at certain points in time, anything config related seems to hang for an indefinite period of of time before it will spit out the expected output:

show run
show run | [i | e | s | etc...]
write t

Sometimes it hangs for minutes, sometimes hours. Other times it's instantaneous, as expected. It doesn't seem to be CPU related and it doesn't seem to throw any logging messages.

Anyone seen anything similar in their testing?
_______________________________________________
cisco-nsp mailing list cisco-nsp [at] puck
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/

jeff-kell at utc

Apr 25, 2012, 12:26 PM

Post #2 of 3 (288 views)
Permalink

Re: 15.2(2)S issues on 7600/SUP720 [In reply to]

I've seen this on stacked 3750s when doing "anything" related to configurations, it
would hang up hard enough to timeout EIGRP hellos from neighbors (and vice versa)
causing general havoc everywhere.

The workaround was to include "parser config cache interface".

During the "hangs" if you could get another CLI session up the show proc cpu sort list
would show the stack configuration replication process eating up the resources. That
wouldn't really apply to 6500/7600 but just in case, thought I'd throw that one in there
(it's a harmless configuration addition as far as I'm aware).

Jeff

On 4/25/2012 2:58 PM, Jason Lixfeld wrote:
> I'm testing 15.2(2)S on a couple of 7600/SUP720 combos and I've experienced some odd behaviour and I'm curious about whether or not anyone else has seen this. It's been really hard to reproduce, but it does happen once in a while, so I'm hoping similar experiences might give me some TAC ammo.
>
> Basically, at certain points in time, anything config related seems to hang for an indefinite period of of time before it will spit out the expected output:
>
> show run
> show run | [i | e | s | etc...]
> write t
>
> Sometimes it hangs for minutes, sometimes hours. Other times it's instantaneous, as expected. It doesn't seem to be CPU related and it doesn't seem to throw any logging messages.
>
> Anyone seen anything similar in their testing?
> _______________________________________________
> cisco-nsp mailing list cisco-nsp [at] puck
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>

_______________________________________________
cisco-nsp mailing list cisco-nsp [at] puck
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/

alumbis at gmail

Apr 25, 2012, 5:11 PM

Post #3 of 3 (276 views)
Permalink

Re: 15.2(2)S issues on 7600/SUP720 [In reply to]

After the problem do you see spikes in the CPU history? I believe most show
commands run as a low priority process that can yield frequently, so it's
possible something else is going on and the symptom is the slow terminal
response time.

On Wed, Apr 25, 2012 at 2:58 PM, Jason Lixfeld <jason [at] lixfeld> wrote:

> I'm testing 15.2(2)S on a couple of 7600/SUP720 combos and I've
> experienced some odd behaviour and I'm curious about whether or not anyone
> else has seen this. It's been really hard to reproduce, but it does
> happen once in a while, so I'm hoping similar experiences might give me
> some TAC ammo.
>
> Basically, at certain points in time, anything config related seems to
> hang for an indefinite period of of time before it will spit out the
> expected output:
>
> show run
> show run | [i | e | s | etc...]
> write t
>
> Sometimes it hangs for minutes, sometimes hours. Other times it's
> instantaneous, as expected. It doesn't seem to be CPU related and it
> doesn't seem to throw any logging messages.
>
> Anyone seen anything similar in their testing?
> _______________________________________________
> cisco-nsp mailing list cisco-nsp [at] puck
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/
>
_______________________________________________
cisco-nsp mailing list cisco-nsp [at] puck
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads