Hung jobs (ready to write, idle) Oracle shuts down.

[BigLouSCSA9]

Has anyone experienced a circumstance where savesets are waiting to run, tapes get mounted and are ready to write, the job seems to start but then fades down to a crawl (sometimes it never even starts) until finally the jobs just stop. It doesn't 'end' or 'fail', it just sits there forever and the drive shows: "Ready to write, idle."

This causes Oracle to panic and shut down to protect itself.

Any ideas how to attack this kind of issue? Any comments/tips would be greatly appreciated.

 

[wallace88]

What version of NetWorker are you using?

Are you performing automatic cloning with your backups?

 

[joeunixadmin]

Hi,

We experience this problem from time to time and when it's not due to a network problem, I can usually get things going again by killing the nsrmmd (but not the nsrmmdbd process) processes. THis will force NetWorker to take a look at what is supposed to be happening.

Hope this helps!

-Joe

 

[silvanaivana]

I had this problem also (W2K server) and after that I realised some serious tape device failure, so I had to replace device (cannot eject, cannot mount, ...)
----it was also a storage node...

Now I change my backup configuration to Autochanger storage node so I'm not sure is there any other cause. Hopefully not!

 

[BigLouSCSA9]

The issues have cleared up.. thanks to all of you for your assistance.

 

[wallace88]

It would be helpful if you told us what was the cause and solution... so that others would know what to do?

 

[MasK]

Hi.
I'm getting this same behavior. Jobs get hung. BigLou: how did yo resolve this ?

 

[fazil1]

Hello,
Networker cannot backup Oracle Hot.
Stop the services and backup or use rmon and do the backup.
or
use rmon to create oracle backup localy then backup the backup directory, i.e dump.
or
use oracle module form legato.
Configure oracle clients timeouts and group timeouts.