Backup Job cancels mysteriously - out of space?

[mcse924]

I have a backup job that ends/cancels itself at about 62GB of data. I'm using 80GB compressed DLT for media. Also, that last session that gets cut off by the job is only 10GB, so the total size of the job is 72GB, still plenty of room left on the media.

I have other jobs that run over 70gb with no issues. I have tried tweaking some things here and there, but still the job dies just over 60gb.

H-E-L-P!

Carpe diem, procrastination is the thief of time...

 

[speshalyst]

HI,

What is being backed up when the job dies? Any Databases? Any file pattern?

Cheers
Speshalyst

So it Shall be Written!
So it Shall be Done!!

 

[mcse924]

Actually, the final session begins and then asks to load the next tape (Sequence 2). Again, I don't know why as there is plenty of room left for these final sessions. The last server being backed up is not a database server specifically.

There is an MSDE databse on here but it is not large at all. It's an asset DB less than 100MB. Also I stop MSDE before the job begins.

Carpe diem, procrastination is the thief of time...

 

[davidmichel]

Full tape that's all.

The tape is 40/80 = 40gb native uncompressed & 80gb that the marketing heads decided would be a good number to use. Anything past 40gb is compressed and so the tape could fill up. So it comes down to a poor compression ratio, most likely due to the type of data.

 

[mcse924]

DM,

I use hardware compression. Would it matter at all if I used Arcserve's software compression?

Carpe diem, procrastination is the thief of time...

 

[davidmichel]

doubt it.

Change the job to no timeout for the span and then swap tapes when you get it.