Saveset failed

[atomik]

Hi all,

Since a couple of days we have a strange problem with one of our servers. One of the savesets generates errors in the deamon.log file:

* <server>:G:\ has been inactive for 35 minutes since Fri Apr 26 04:25:27 2002.
* <server>:G:\ is being abandoned by nsrexec.
04/26/02 05:02:56 nsrd: runq: NSR group WINNTPOOL exited with return code 1

Despite of this, when I look at the volume save set info for the used tapes, the flag 'c' tells me that the saveset is completely contained by that particular tape. It's also possible to do a restore from a file in that saveset.
So it seems nothing is wrong at all?

Does anyone know what can cause this behaviour?

Remco

 

[MoshiachNow]

Hi,

1.Check rmt errors in the system log.

2.Possible media index corruption .Try greping for &quot;WISS&quot; errors in the daemon.log.

If found - try first running &quot;nsrck -L6&quot; , &quot;nsrck -mv&quot;.

3.If the above does not help - try the following procedure to recreate the media DB index:

The procedure of controlled corruption:

1) Run mminfo to write the records into readable format. For example:
# mminfo > file or copy the mm into mm.old.
2) Media database compression:
a) Shut down the daemons.
b) Remove /nsr/mm/.cmprssd (if it?s there) and cmprssd.(and this file it will recreate again after you run nsrim -X (step5))
c) Restart the daemons.
3) Media database scavenging:
a) Shut down the daemons.
b) Cause a &quot;controlled corruption&quot; by removing /nsr/mm/mmvolume/{*_i*, ss, vol}.

The only file in you should see in the mmvolume directory when done are:
ss.0, client.0, clients, vol.0, VolHdr ==>client.0 is for 6.x only
c) Remove all the files under the directory /nsr/tmp
d) Restart the daemons.
4) Run the following command:
# nsrck -F (nsrck –L6 for Networker6)
5) When step (4) completes, run:
# nsrim -X
&quot;Long live king Moshiach !&quot;

 

[thunga12]

/ save: SYSTEM error, no such file or directory
/ save: Cannot open save session with saptdb01