In the past we always used Veeam SureBackup to test the majority of our backups, but as we have a new SAN for Veeam, we thought it was a good idea to use the old SAN and some ESX servers to do a full "manual" recovery test of our entire production environment. Also the perfect opportunity to update/rewrite our disaster recovery plan.
Anyway, while restoring the large number of VM's I found 2 annoying things are happening:
1) Veeam does not honor Veeam proxy disk settings(or I have a wrong understanding of how this should work - so have not created a support case yet). Lets say I setup 2 proxy's(virtual appliance) per ESX host for restores. For performance purposes(so that restores are at least split between RAID groups on my SAN) I configure proxy "A" with a manual selection of LUN's/Datastores, and proxy "B" with another manual selection of LUN's . Veeam does not seem to care about this setting at all, and just uses proxy B to restore VM disks to the LUN's/datastores that are not in his selection list(and still uses hot-add). Is this a bug, or do I not understand this feature correctly?
2) I would like to priorities backups over restores. This because restores are taking a long time to complete and my repository of course has a maximum number of jobs. I know there are settings for maximum "jobs per repository" and maximum "jobs per proxy"(which does not work very well because it's using more/other proxy's then it should - see problem 1). I this "the" way to throttle jobs, or is there some other setting to prioritize specific jobs? That would be very helpful in this case, as the backup jobs are now trailing behind a large number of VM restore jobs.
Something else: Veeam Availability Orchestrator BETA. Seems great, I want it.
Thank you in advance!
--Mark