Retention policy doesn't work anymore!

[phil75]

Hi everybody!,
we have successfully installed by our customer veeam to backup virtual machine hosted on hyper-v server. We had created a custom backup job that run everynight with 4 virtual machine to backup, and we a retentin policy to apply in order to consolidate and freeing space on an externa iscsi disk. Veem is working without any problems from januar 2012. Till past days, everything was fine, but in the past two days we had receveid a notification message in which we get a waring message : low disk space on disk e: (iscsi disk configured to stored backups). after browsing the directory we found there are some old files that wasn't consolidated and was not deleted as we aspected: we had configured a retention policy of 14 days, but in the store we can find file from 9 december. is there a way to force a consolidation? thanks

[Vitaliy S.]

Hi Filippo,

Can you please tell me what backup mode are you using? Forward or Reversed incremental?

Thanks!

[phil75]

Reversed incremental

[phil75]

sorry, we do backup using forward incremental

[Vitaliy S.]

Ah...that's what I've thought. How many full backup (VBK) files you have on your target repository? Pretty sure the answer will be - only one. Please take a look at the existing discussion for further details and explanation > Forward Incremental Job - Gone wrong!!!

As regards possible ways to resolve this, then please check out this thread for available options: Backup share running out of space

P.S. once you perform all steps that I have mentioned above, please make sure you schedule full backups at least once a week / two weeks in order to avoid situations like that in the future.

Hope this helps!

[phil75]

Hi Vitaliy!,
thanks for your quick reply!... so I've checked on my repository and I've found there are 3 full backups (3 vbk files). so i've read the thread you suggested me:

"Vitaliy S. wrote:
Since you're using forward incremental backup mode, there is no easy way to remove part of the restore points as all of them depend on the parent VBK file the backup job created on the first run.

You have two options to proceed with:
1. manually remove VBK + dependent VIB files from the disk (the entire chain)
2. or manually delete the most recent VIBs files starting from the newest one. In this case you will be able to use the oldest restore points in case of disaster."


so, if I understand I must:

1) move all files before last full backups on 30-12-12 in temp folder outside the backups directory;
2) run a "Rescan Repository"

Please let me know if this procedure is correct... and do you know if the future could happen anymore?
thanks

[Vitaliy S.]

Not exactly. There is no need to perform repository re-scan operation since this doesn't prevent your backup job from running. Just move or delete the files you no longer need, and start your backup job with new full schedule and updated retention policy.

[phil75]

HI,Vitaly.
I did remove manually the old full backup from disk. After two weeks the problems happenend again. What shall I do?

[foggy]

Filippo, what problem are you referring to exactly? I would note, that having several full backup files (VBK) is normal for forward incremental mode. Please review this topic for better understanding of how it works.