Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Westi on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

(96) Unable to allocate new media for backup, storage unit has none av

Status
Not open for further replies.

jon4crm

MIS
Feb 26, 2004
71
NG
I just installed netbackup 5.1 on my environment. I have a DLT and 4mm cartridge standalone drives on my server. I have on my backup domain 1 unix server and 3 win clients. I have tested backup of the installation individually on each client and I was able to backup to the drives.

I have created some policies and scheduled it to run but they displayed error 96, unable to allocate new media for backup, storage unit has none available. The unix server policy backed up but the clients gave the error 96.

When I ran same policies to backup to disk, they were successful. Reason being that disk storage policy attribute doesnt have storage unit. I believe my problem has something to do with storage unit, media or volume pool.

When I ran /usr/openv/netbackup/bin/goodies/available_media, I found out that all the media are active. There is no one available for backup. Probably thats where the error is coming from.

My question is, how can I make media available for backup if this is the problem. I have put new media into the drive but I still got same error. How can I configure media and have them available for backup? How was the other media active and when I put new media, I didnt know how and where to configure it available. On subsequent scheduled backups, how do I make new media (tapes) available instead of active.

Guys I need you help. Thanks in advance.
 
Well there are potenially a lot of issues that can get you into this state. You need to do some basic checks. First are the tapes set to the same type as the tape drives. Tapes which are active will have been allocated to a media pool are the jobs which are now trying to run set up to use the those pools. THe best way is to have blank empty tapes in a scratch pool to which the jobs can help themselves the the tapes they need. New tapes should not be active. You setup the tapes when they are added via the inventory roboot interface on the GUI.
 
I have 2 standalone drives. One Dlt and the other is a 4mm tape. Right now am trying to backup only to the DLT tape and the 3 media are 2DLT and 1 4mm media. How they got allocated and active, I dont know.

I have before now the default volume pools: DataStore, NetBackup and None; and in NetBackup I saw the 3 media there. How they got there I dont know. Are you suggesting I should create a scratch pool? How will I assign new tapes (media) to the pool and you know I have 2 standalone drives on the server. No robots. How do I assign media (tapes) to the scratch pool? Also how do I remove those active media from the setup?
 
Right first ignore the active tapes they are not going to help you. You must set up some tape rules so that there is a policy (not job) for the tape lables. You should then be able to place blank tapes in the dlt drive. When the backup job starts it will look at all the avalible tape drives and if one has a useable tape it will write a lable on it and use it for the backup. But you must have set up the storage unit with the drive and asigned that to the backup job 'policy'. As you have no library which can assign and load tapes it's self then a scratch pool will not help.
 
After you follow some of lenski's advice for the intitial setup. You then have to also keep in mind that if the backup you are using is 1week retention and the media in the drive was previously backed up at a three week retention that media is set to three weeks. Netbackup by default will not allow multiple retentions on media and I would not recommend changing this. Just make sure that you have the same retention in the policy as the tape in the drive. Also make sure the pool is the same. Example: I created a pool called "Unix" that I send most of my backups to. The netbackup pool I believe is for catalog backup tapes and tapes that Veritas does not know what to do with.
 
Guys,

If I got what you are saying. You are suggesting I should probably create a policy with say 1week retention period. Assign say my Policy storage unit which is the DLT drive, have a unique Policy volume pool say "Windows". I did all these. The problem could be that I have different policies with different retention periods going to same media.

Right now, I will create new media rules and make sure I have same retention period to the tape.

However I did inserted into my DLT tape drive a new tape but the backup still had same problem.

I will make sure these rules are harmolised and tried again. I will start allover again. I will:
create new storage unit.
create new policies with probably same retention periods
recreate the standalone drives and probably
create new media rules

I hope this will work. Whatelse do I need to do to have a successful backup?
 
i have the same error too!
Unable to allocate new media for backup, storage unit has none av

its been very fustrating, i tried to list media contents of the tape and got this error: " cannot read media header, may not be netbackup media or is corrupted
 
Well beleve what it tells you, the tape you have loaded is not acceptible to NB for some reason. Try a clean new blank tape. Load your tape and then try the backup. Check the comments above for the set up.
 
it was a clean new blank tape, only used for 1 or 2 times of NBU backup. The tape is not faulty, but where could be wrong?
 
Kendas,
From the above post you mentioned that you used the tape one or two times for a NBU backup. By NBU do you mean the catalog backup?? If yes than that would be why you are having problems accessing that tape. I believe you would have to deassign and delete that tape from the volmgr and then re-inventory so the tape goes to the none(or scratch) pool and is added as new media. Then you should be able to use this tape. Since the NBDB is silent during it's own backup there is no expiration date assigned and it cannot be overwritten except by another NBDB backup run. Please correct me if I am mistaken on any of this guys. I have never actually tested this. Just from stuff I've read.
 
the tapes are not used for catalog backup.
they are used for daily backup.
btw, how to deassign and delete that tape from the volmg ?
 
A tape which has been used for 1 or 2 times is not a new clean tape. You need to take a new one out of the packet and use that. Then if that fails you have a real problem not just an issue with one tape.
 
To deassign a tape you first go into
/opt/openv/volmgr/bin/vmquery -m (tapeid)
This will display info on the particular tape.
Get the volume pool # and the status i.e. 0x0
Then perform
/opt/openv/netbackup/bin/vmquery -deassignbyid (tape #) (pool #) (status 0x0)

Then you should be able to delete the volume from the media screen in the GUI and then re-inventory and it will be added as new media. P.S. You may want to also run the /opt/openv/netbackup/bin/admincmd/bpexpdate command to expire the images on the tape.
bpexpdate -m (tape #) -d 0 -host (media server name)
Hope this helps!
 
Hi Guys,

Still on the error(96), I have got Unix clients to backup to the storage unix (tape) but the windows clients are not. They keep having error 96. Does it mean Unix and Windows backup can't go to same storage unit. The total unix size is about 10GB and the storage unit can accommodate 20GB. Why is it so. This is a new tape of 20GB. Even manual backup to the same tape also gave same error from both unix and windows boxes but I dont think the storage is acually full.

Please any help will be highly appreciated.
 
Unix and Windows can go to the same storage unit, but of course not at the same time under the same policy. Nor I think to the same tape. If your unix clients are working but your windows ones are not then stop the unix backups for the moment and just try and get the windows ones working. The art in fixing problems is breaking them down to the smallest unit possible and then building until it fails. So try one policy for one client onto one clean blank tape.

Remember the error 96 is 'I don't have a tape to match your requirements' not 'I don't have any tapes'. So for example if you had one tape and 2 jobs which have differant retention periods then you would get 96 because NB will not mix retentions on tapes as a default.
 
how can i clear the retention of all the tapes that is in the library so that the backup can run smoothly without fail?
 
whats the difference between using vmquery -deassignbyid and bpexpdate command?
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top