Best way to go when increasing repo disk size?

[TommyB]

Hi,

the free space of my backup repository is running low and I already have ordered new disks to double the total capacity.

I have a secondary NAS storage with sufficient free capacity, I thought to do the following:

- Stop all veeam services and set them to manual start
- copy the old repo to the NAS system
- remove the old disks
- install the new disks and move the old repo form the NAS back to the original location
- re-enable all Veeam services

Is this the recommended way to go?

regards
Thomas

[Vitaliy S.]

Yes, but don't forget to re-scan your new repository to populate Veeam backup console with the existing backup chain or you don't want to run your backup jobs until you add some more space to old backup repository?

[TommyB]

..or you don't want to run your backup jobs until you add some more space to old backup repository?

Good question - I will probably decide it after I know how much time it took to move the 2.3 TB of data to the NAS.

The old repository (or to be precise, the four 750GB disks) will be swapped with four 2TB drives - so it will be gone.

Because the old intel servers raid controller can only create volumes with a maximum of 2 TB I will have to use Windows 2008 dynamic disk feature to create a 8 TB volume that spans 4 x 2 TB physical volumes.

Currently tape backup is still running so I think I can start the procedure on Wednesday.

[Vitaliy S.]

Got it, once you move your backup files to the new box, I would still recommend doing incremental backups (should not take much space). If you proceed with this advice, then you will need to create a repository on your NAS device and do a rescan procedure, so that backup jobs could be mapped to these files.

[TommyB]

Got it, once you move your backup files to the new box, I would still recommend doing incremental backups (should not take much space). If you proceed with this advice, then you will need to create a repository on your NAS device and do a rescan procedure, so that backup jobs could be mapped to these files.

Copy process started - estimated time is 12 hours because the local disk system is so slow (60mb/s) - unfortunately there currently is no budget for a faster backup machine.

So I will not have any nightly backup today

I will create a new repository tomorrow morning and map the backup jobs to it to be able to run a backup then.

[Vitaliy S.]

You still can do a full backup of most critical VMs (just in case) and tomorrow delete it.

[TommyB]

Importing the temp repository worked like a charm, incrementals are running... next step is changing hard disks and moving the repo back to its original location.

One thing I missed was the configuration backup - but there is no option to "map backups" for that.

So I got an error message that the system could not create the folder for the config backup as I already removed the data partition. But that's not critical anyways.

[Vitaliy S.]

Did you import or rescanned the repository? In both cases mapping option should be available. Thanks for keeping us updated.

[TommyB]

Did you import or rescanned the repository? In both cases mapping option should be available. Thanks for keeping us updated.

I did both, but in the config backup dialog there was no "map backup" link.

Also, after moving everything back in place this morning (incl. re-scan of the repo) the tape backup started to do a full backup including the incremental chain of Wednesday & Thursday - but this already happened on Monday.

I expected the backup only to send the incrementals from Wednesday and this morning to tape?!

[veremin]

Vitaliy seems to have talked about common backup jobs, not configuration backup one.

Anyway, for tape issue it would be better to create a separate topic in corresponding subforum in order not to derail the existing thread with several different questions.

Thanks.

[TommyB]

Vitaliy seems to have talked about common backup jobs, not configuration backup one.

Ah - ok... for common backup jobs, everything was just fine - mapping worked in both occasions.

There just seems to be no way to map the config backup.

[veremin]

If needed, you can leave previous configuration backup at hand (separate folder or something) for a while in case restore is needed and decommission it when time comes.

[TommyB]

Thanks guys for the great support here - I wished more IT companies would monitor and respond on their forums in such an active way.

[veremin]

You're welcome. We just stand for one of our core values.