Master thinks there is no more media available

[Machiavel]

Hi all!

Small / medium recurrent problem.

My libraries (2x ESL 9326D) indicates that there is 20 medias available, but my Windows 2003 Netbackup 6.5 server indicates that there are no medias available so my back-ups fails.

I ran the discovery script, it indicates no frozen medias, and no medias leff.

I don't know wich information you might need to help me out with this one and I am a little clueless as this problem come and go.

Thanks!!

[Omar_Villa]

is this media on the scratch pool? is the pool configured under your policies the correct one? and what error are you getting?

regards

[Machiavel]

Thanks for your interest in our question Omar:

Yes the media is in the scratch pool.

The pool is configured under the right policies, according to our knowledge and the the installation / optimisation guide we read.

The error message is:

 unable to allocate new media for backup, storage unit has none available(96)

When we scan on the media server with:

 /usr/openv/volmgr/bin/vmquery -h gir-pgesin0z -w -rn 2 | awk '{print $12}' | grep Scratch | wc

It sees 19 cassettes this morning, but our master sees nothing available.

This happens when we fall under 20 cassettes.

[Andy_Welburn]

Can you confirm your 'scratch' pool IS actually a Scratch Pool & not just some other pool called Scratch?

vmpool -h <hostname> -list_scratch

On the same lines, you haven't got another pool have you with 'Scratch' in its name that your command is picking up?

e.g if you have a pool called "This_is_not_the_Scratch_Pool " and it has 20 tapes in it then when you run the command:

/usr/openv/volmgr/bin/vmquery -h gir-pgesin0z -w -rn 2 | awk '{print $12}' | grep Scratch | wc

it will pick these up which you are then confusing for the real thing?

Can you not check on the GUI that the scratch tapes are in the library?

[Andy_Welburn]

Additionally check this out (posted yesterday on another thread:(

Exit Status Code 96: Unable to allocate new media for backup, storage unit has none available.

May give you a few other things to check for.

[Machiavel]

Thanks for helpin' out!

 vmpool -h <hostname> -listscratch (note that my master is a windows box)

As per what we configured: 1 scratch pool named Scratch.

I ran  vmquery -h pgesin0z -w -rn 2 >> c:\pool and checked manually if any of our pool name contained Scratch, negative.

So it is not confusing the real scratch pool and another with scratch in it.

We checked manually, went to the robot and counted, the last 2 times it happened, they were there all 12 and 19 cassettes.

We are still investigating any configuration problem on the windows master.

Regards,

Peter

Message Edited by Machiavel on 08-06-2008 06:41 AM

[Patrick_Whelan_]

One thing I've run accross many times. You inventory the library and for one reason or another either bar code rules are wrong or the inventory doesn't use them so the tapes are assigned as a different type. E.g. HCART when they should be HCART3 or something similar. Just a thought.

[Machiavel]

Thanks for your thoughts, currently investigating on that side

Another thing that might help you guys think of something I missed:

I have two libraries, let's say "A" falls under 20 cassettes available in it, it won't fall to the "B" one even if I have 100 available in it. Even if our policies indicates so.

Is there a setting we are missing ?

Thanks again!

Message Edited by Machiavel on 08-06-2008 07:05 AM

[Patrick_Whelan_]

You don't mention which version, but, unless you're using a very old version, take a look at storage groups.