BackUp Job Capacity Issue

[Jaime_Neal]

We have a Dell server running Windows Server 2003 (not SP1). We had a backup job that was backing up around 256GB onto one tape, until there was a problem with the drive (Dell 110T SDLT). Dell shipped a replacement and then the backup job would get to 210GB and alert for a new tape.

I've update the Veritas BackUp Exec software from 9.0 to 9.1, swapped the new Dell 110T with a known working Compaq SDLT320 drive, tried different tapes, tried a couple of things I have found here on the forums and still no joy - it still takes two tapes to backup what was previously on 1 without any data increase.

As an additional note - I have run a test run on teh job and it completes sucessfully.

Please help!Message was edited by:
Jaime Neal

[A_van_der_Boom]

It could be that the new tapeunit has a different firmware version, causing it to compress less the your previous unit. You might try to upgrade it. (contact DELL for this)

Regards
A van der Boom

[Jaime_Neal]

Hi A van der Boom,

Thanks for posting the comment on firmware, however we tried getting the latest firmware for it, and that failed.

The Compaq unit that is now sitting on there worked correctly on the server it was previously on, but this issue appears when it is plugged into the current server, which makes me think it might be something to do with Veritas itself...

Thanks,
Jay

[A_van_der_Boom]

The only sugestion I have is to try different device drivers, try Veritas and the manufactures driver and see if this makes a difference.

Regards
A van der Boom

[Jaime_Neal]

I've tried both the Veritas device drivers and Dell's - both the revision that was on the old unit and the latest ones, and the HP unit I have done the same but still no change.

It's very frustrating!

[A_van_der_Boom]

The last sugesstion I have is to use a brand new tape or format an existing one and see what happens.

Regards
A van der Boom

[Jaime_Neal]

Unfortunately I have tried a full format of an existing tape, and two new tapes (from different manufacturers). Nothing helped!

[Thomas_Bianco]

i wonder if this is related to any of these:

http://forums.veritas.com/discussions/thread.jspa?threadID=54814
http://forums.veritas.com/discussions/thread.jspa?threadID=56794

un-compression ratios seem to be a new problem in 9.1 SP4

[Jaime_Neal]

Thanks for the links Thomas. It certainly looks like an issue with the newer release. I'm going to try changing tonights back-up to use software-only compression to see if that yields favourable results, other than that - it's back to the drawing board!Message was edited by:
Jaime Neal

[Jaime_Neal]

I tried changing from hardware to software compression and we-re still running across two tapes. It's prompting for the swap at the same point (210GB), which may or may not be useful information...

[Jaime_Neal]

I ran the BackUp Exec Diagnostics on our problematic server and one that we have no problems on and I couldn't find any differences that would point to the problem we are experiencing.

[Sharvari_Deshmu]

Hello,

If hardware data compression is used, then the compression is done by the tape drive.

VERITAS Backup Exec (tm) only sends instruction to the tape drive to do compression. If data is not compressing (either because hardware compression on the tape drive is disabled or because the tape drive does not support it),

Also please see the technote:

Backups fail to get double the amount of the native capacity recorded to a media, even when hardware compression is being utilized.
http://support.veritas.com/docs/199542



Thanks,

NOTE : If we do not receive your reply within two business days, this post would be marked assumed answered and would be moved to answered questions pool.

[Jaime_Neal]

Hi Sharvari,

I understand everything that you have said, but putting the tape drive onto another server to do its back-up job works fine, then switching it back - we get the same problem again.

Moving the back-up job to the other server works fine, so it points to a local configuration to me.

It is not possible to run this back-up from the other server permanenetly, so we need a fix if anyone has any other ideas?