Networker

Re: [Networker] Unloading drives that are not loaded

2006-01-26 10:53:31
Subject: Re: [Networker] Unloading drives that are not loaded
From: "Thomas, Adam (NIH/NIMH) [E]" <adamt AT NIH DOT GOV>
To: NETWORKER AT LISTSERV.TEMPLE DOT EDU
Date: Thu, 26 Jan 2006 10:50:27 -0500
> It's because networker keep track of mounted volumes.
> At restart, networker will restart/recreate nsrmmds and re-read
> volume labels for the drives he know to be mounted.
> 
> Before shutting down networker, use "nsrjb -uvs <server>"
> or "nsrjb -uv" issued respectively from the storage node
> handling the robotic arm or from the backup server if it is
> operating robot to unmount all devices.
> 
> This will prevent behaviour you are seeing.

Unfortunately I'm usually in this circumstance when the unload command
has already failed or I'm recovering from a scsi error or some other
circumstance where the networker server is not performing as I would
like.

I'm looking for a command that I can run after the fact that tells
networker: "forget whatever you thought was going on with the tape
drives and the library and start over" 

Is there such a command? nsrjb -H doesn't seem to work for this on my
setup.

-Adam


-----Original Message-----
From: Thomas, Adam (NIH/NIMH) [E] 
Sent: Thursday, January 26, 2006 10:26 AM
To: Legato NetWorker discussion
Subject: Unloading drives that are not loaded


My networker install seems to get confused about whether there is a tape
in the drive.

My situation, which I've experienced many times before, is that for one
reason or another I've had to force a shutdown of the networker daemon.
Then I manually empty the drives and restart the daemon. When it comes
back up, nwadmin doesn't show any tapes in the drives. However if I run
"nsrjb -H" or "nsrjb -I", networker tries to unload one or more of the
drives. This fails since there is no tape in the drive, but I have to
sit through 10 minutes or so of:
media info: unload retry for jukebox 'tls5433' failed - will retry
again"
media info: unload retry for jukebox 'tls5433' sleeping 30 seconds"

Is there anyway I can avoid this? Or at least is there a way I can
decrease the number of times it retries unloading a tape that is not
there?

This is Networker 7.2 on Redhat Enterprise Linux 3 server. The library
is a scsi-atached qualstar TLS5433 with 2 SAIT tape drives.


Thanks,
-Adam Thomas

To sign off this list, send email to listserv AT listserv.temple DOT edu and 
type "signoff networker" in the
body of the email. Please write to networker-request AT listserv.temple DOT edu 
if you have any problems
wit this list. You can access the archives at 
http://listserv.temple.edu/archives/networker.html or
via RSS at http://listserv.temple.edu/cgi-bin/wa?RSS&L=NETWORKER