I'll take a stab at this as it gives a sense of deja vu.

Is it always asking for the same old clone tapes or the same old backup
tapes??   For me this was the case.. whenevever I went to clone it would
ask for an old backup tape from over a year ago.. always.

I originally thought it was tape location problem but the inventory was
fine.  I then started suspecting the clientid, ssid and mediadb.  The reason
for this is the problem started right after we added a new client to the
server..  I have no explanation as to why this may have cause it though.

Looking at my nsrclone log (from the scripted cloning) I saw errors about
an invalid ssid or ssid not found (not that sure about the exact error).
Using mminfo to search for that ssid, it turns up on the piece of old media
that Networker kept asking for!

For myself, that piece of media was no longer offsite and should not have
been so I deleted that volume to rid myself of that ssid.  I think if you
can determine what ssid is being looked for you can delete just that ssid.

So it may be mediab corruption/confusion but I recall trying all the normal
mediadb procedures without success for this one.  Only deleting the ssid
worked.  If the volume being called out or ssid is a valid one that you
want to keep I'm not sure what to do.

With any suspicion of mediadb corruption a call to Legato is
warranted (but be prepared for a long case on this one).


Robert Maiello
Thomson Healthcare

On Thu, 27 Mar 2003 08:08:43 -0500, Mcandrews, Carleen M NAVSAFECEN
<carleen.mcandrews AT NAVY DOT MIL> wrote:

>        We are using Solstice Backup 6.1.2 Build.340. We are using it for
back-up/recovery of Informix Databases and Solaris o/s file systems. Last
fall we upgraded our tape library to Sun StorEdge L9 autoloaders.  We then
started to share the default clone pool for cloning both the database
savesets and the o/s  filesystems (prior to that we had separate pools
defined for each).
>
>        All was working well until the begining of March. On the database
side, we have always had cloning to automatically follow the production
back-up once a week. On the o/s side, our filesystems are cloned with a
shell script, utilizing the nsrclone command.
>
>        Recently the clones seem to be "confused'. Backups are still
working fine. But the cloning afterwards keeps calling for old, irrelevant
volumes. Either it's asking for volumes with old savesets that it wants to
clone, or it prompts for old clone tapes to be mounted when there are clone
tapes already mounted and ready to be written to.
>
>What are we overlooking here? Why would cloning suddenly stop working
correctly? Any ideas?
>
>
>Carleen McAndrews
>Database Administrator
>Naval Safety Center
>757-444-3520  X(7050)
>mailto:carleen.mcandrews AT navy DOT mil
>
>
>--
>Note: To sign off this list, send a "signoff networker" command via email
>to listserv AT listmail.temple DOT edu or visit the list's Web site at
>http://listmail.temple.edu/archives/networker.html where you can
>also view and post messages to the list.
>=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=