Archive Update job frozen

Unanswered Question
Aug 29th, 2007

The archive update job has frozen since 14/8/07 - attempts to stop it result in error suggesting that it is in the wrong state; attempts to delete it result in an error saying that the job is still running. How can I remove the job?



Attachment: 
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (1 ratings)
Loading.
jreekers Wed, 08/29/2007 - 17:30

one quick thing you can try is to restart the ConfigMgmtServer process.


You can do this by going to

$NMSROOT/bin

and issue the commands:


pdterm ConfigMgmtServer

pdexec ConfigMgmtServer


The running jobs will get stopped. Now try using job cli remove the job.


If that doesn't work, could you please provide more information such as:

1. Platform this resides on (Windows or Solaris)

2. Version of RME you're running

3. cut-and-paste of the error messages you receive when trying to delete the job.


Thanks,

-J

Joe Clarke Wed, 08/29/2007 - 19:04

The safest thing to do is restart dmgtd. That should forcibly stop the job. Once dmgtd comes back, you should be able to delete the Failed instance.

marshajr Wed, 08/29/2007 - 20:22

Tried stopping dmgtd and restarting it - however the job still appears in the RME Jobs, and cannot be removed. Note:the job summary shows that the job has completed with a completion date and time, however the RME Job list doesn't show any completion time.

Joe Clarke Wed, 08/29/2007 - 20:33

Okay, the final thing you can try is to run the following command:


NMSROOT/bin/cwjava -cw NMSROOT com.cisco.nm.cmf.jrm.jobcli


(Note: NMSROOT is the path into which CiscoWorks was installed. By default this is C:\PROGRA~1\CSCOpx on Windows and /opt/CSCOpx on Solaris.)


Once the JRM shell initializes, run the following commands:


cancel 1027

delete 1027

exit


The job 1027, and all of its instances should be gone. You can then go to RME > Admin > Config Mgmt > Collection Settings to reschedule the System Collection Job.

marshajr Wed, 08/29/2007 - 21:30

Attempt to run the command results in the message "Unable to launch JVM". Is this a conflict in versions of Java on the server, as there are multiple versions on the Server?

marshajr Thu, 08/30/2007 - 14:42

Have resolved the issue, as follows:

On the server entered pdshow and noticed that the process 1027 was listed as 'never started'. It obviously had hung up at some stage and not responded to daemon and system restarts. Issued a pdexec for the process to restart it, and this appears to have cleared the job.


Thanks for your assistance.

Actions

This Discussion