Mediaset creation issue and excessive tape movement

[Shunx239]

Hi everybody!

Preface: Veeam is a great backup product, however it is grown up as specialized one and lacks many enetrprise features.
Every time I miss one, I get frustrated, especially if no easy workaround is available.

Definitively the way jobs are permitted to append data to tape provides no flexibility. It has to be decided on the target mediapool with few options for mediaset rollover: every time, never or periodically but at least once a week.

I created a backup to tape job that saves full backups on separate media pool, virtual fulls are scheduled once a month. The ful backups fit nicely in one tape and I don't like them to be appended so I configured the target media pool for creating mediaset every time.

I know the backup to tape job evaluates every time whether any full backup have to be saved to tape. It tells me:

Code: Select all

 10/02/2017 21:12:57 :: Processing full backup of Afrodite Backup: nothing to backup   

But for some reason it unloads the incremental tape from the drive, loads free tape and makes it new mediaset in the full media pool, writes nothing, loads again incremental tape and starts writing data.

All that tape movement remained unnoticed in the log as long as the target media pool had been configured to reuse the same mediaset, but now it ends up with a tape wasted on the every job run since the next time the new tape is allocated (and no data written).

I have been able to mitigate the issue setting the target media pool for creating mediaset once a week, but this has definitely to be fixed.

Nobody faced any similar beforehand?

Support Case # 02066701

BR
Nikolay

[lyapkost]

Hi, Nikolay.

This is a known issue in 9.5 GA and it is fixed in Update 1, so please install it.

[Shunx239]

Yes, indeed, already pointed to this by the support, installed updates and got it fixed.

Curious enough, the <nothing to backup> lines don't show anymore in the proggress windows.

I know, the tape scheduler has been reinvented in 9.5, feeling it to be changed again in update1.

In the 9.5GA the tape jobs started waiting for the tape availablity, but there was no real queue: if multiple jobs needed it the first one in the alphabetical order won.

Not sure whether it has been changed also in update1, observed by chance in GA.

Thank you for attention.