Issue with VEB job to tape

[AllenK]

Not sure if I should drop this here, or in the VEB forums.

I'm having an issue with one of my 2 VEB to tape jobs. The job is writing the full, uncompressed data size to tape as opposed to the compressed size. If I go in to modify the tape job and select the VEB job as source, it says full backup size is 5.4TB, which is also the size of the VEB files on disk. However, when I run the tape job, it writes the full, uncompressed 11TB.

My other VEB to Tape job isn't having this issue and is writing the compressed size. Anyone have any insight?

[veremin]

You mean a backup to tape job decompresses the data before archiving it to tapes?

[AllenK]

It looks like it's behaving that way, yes. The files on disk are 5.5 TB, the full job reports estimated size as 5.4TB. When it writes to tape, it writes 11.2TB(The uncompressed size). Just not sure what's causing it.

None of my other tape jobs(Regular Veeam files or the other VEB job, which includes about 5 servers) are exhibiting the same behavior.

[veremin]

Is it the only job that is specified as a source for a backup to tape job? Is anything specific about location where the full backup resides? Say, Windows Deduplication is used or something?

[AllenK]

It is the only job specified in the source. It resides on a server 2012 volume, but no dedupe(Had problems with files over 4TB even). The only specific thing is that it's configured with storage spaces, though it all resides on the same DAS. The same volume houses my other VEB job files, as well as the VBR job files.

[veremin]

That seems quite strange, actually. So, feel free to open a ticket with our support team. Once it's done, post the case number here, so that we can follow it.

[AllenK]

I was slow getting back on this, but here's the ticket number: 01099003.

[AllenK]

Issue has been resolved. Here's a break down of the problem:

-The VEB to disk job holds 8 restore points.
-The VEB to tape job failed initially due to a tape drive failure.
-The job was set to run a synthetic full job on sunday, and incrementals every other day
-Due to the job failure, Veeam saw that there were two sunday restore points on disk and tried to write them both to the tape job(Hence the doubling)
-I perpetuated the problem by cancelling the job when it ran over-long to meet off-site vaulting schedules

Support had me disable incremental jobs and I changed the full job to a different day. This allowed the job to write a single synthetic full to tape. Subsequent synethic fulls ran correctly and turning incrementals back on didn't cause any issues.