Netconfig caused 4006 to reboot

Unanswered Question
May 4th, 2007

Running LMS 2.6 - RME 4.0.5 on a windows platform. We have multiple 4006 swicthes running everything from 5.4(2a) to 8.4(7).

The Netconfig job was just supposed to run the command sh banner on the switches and archive the config.

Some of the 4006's were successful, some rebooted, and some partialy booted.

Has anyone ran into something similar to this before?

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Joe Clarke Fri, 05/04/2007 - 21:02

I have never heard of any such reports. Netconfig should not be doing much differently that what would happen if you logged in manually, and performed the same commands. what is does do differently is setup the session for less interactivity (i.e. disables the pager, turns off some session messages, etc.).

If you happened to have ConfigJob debugging enabled when you ran this job, the job directory would be helpful in determining when the switch lost connectivity with RME (i.e. when it reloaded).

If not, the best way to troubleshoot this would be to schedule a maintenance window, and repeat the same job with ConfigJob debugging enabled as well as while running a sniffer trace. When the switch reloads, there will be a good paper trail showing what led up to that. Opening a TAC service request with that information will get you to the right team that can troubleshoot why CatOS is crashing.

It may also be good to open up a TAC service request before doing the job to see if there are additional things that should be set up on the switch side for making the most out of the crash.

All that said, if your job is to do a show banner, you should consider using Netshow instead of Netconfig. Netconfig is geared to doing configuration changes rather than executing commands for purposes of capturing output.

Actions

This Discussion