[j-nsp] Event Log RPD_SCHED_SLIP M20

[Nilesh Khambal]

The message shows that the scheduler slips were caused due to a user process
taking up CPU for longer than 4 seconds. This could very well be some
task/job inside RPD or could be some other process. Was there any
configuration change done recently on the router that triggered these
messages? Are they seen after running any command? What is the role of this
router in your network? I would suggest running the "show system processes
extensive" and "show chassis routing-engine"  immediately after the message
shows up in the syslog.

You could, for short term, also enable task accounting with "set task
accounting on" from operational level and monitor the output of "show task
accounting" after the message is seen. Usually, there will be an additional
message after enabling task accounting, showing which task inside RPD (if it
is indeed RPD) is taking up the CPU. Do not forget to disable it after you
have seen a couple of incidents of these messages.

Thanks,
Nilesh.

 


On 4/9/10 3:00 AM, "Juniper" <juniper at iber-x.com> wrote:

> Hello there,
> 
> Since few days we are detecting these message in our log message in a
> Juniper M20 with the JUNOS version 7.3R1.4 :
> 
> Apr 6 06:00:15 xxxx-yyy2.abc-d.net LEV[2625]: RPD_SCHED_SLIP: 5 sec
> scheduler slip, user: 4 sec 940542 usec, system: 0 sec, 14925 usec
> Apr 6 05:58:07 xxxx-yyy2.abc-d.net LEV[2625]: RPD_SCHED_SLIP: 4 sec
> scheduler slip, user: 4 sec 75182 usec, system: 0 sec, 0 usec
> 
> Does anyone know what could be the problem?
> 
> Thanks in advance,
> 
> 
> 
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp