The utter tedium of adding individual backup jobs

[OWTech]

I'm sure this has been requested before but why can you not create individual jobs from an entire vCenter inventory in one go? After creating 160 jobs you start to get a bit hacked off. I don't want to have one job for all the vm's on a host or a cluster. If there's a way to do this already I apologise in advance and please educate me.

[foggy]

James, job cloning functionality available in Enterprise Manager could help a bit. Also I believe you can achieve what you want with the help of Powershell.

[dellock6]

Instead of creating a per-VM job, you can choose different alternatives:
- one job per Resource Pool, easy if you have multi-tenancy like in vCloud Director, where 1 resource pool = 1 customer
- one job per Datastore, this has the advantage of taking care of free space on the datastore, in regards of taking snapshots while backup runs.

Both ways,whenever you have a new VM in the resource pool or datastore (but you can also choose clusters, or VM folders...) it will be added to the backup set.

Luca.

[tsightler]

If you are creating one VM per job, that's not really the way the product was designed, it was designed to have multiple VMs in a job. However, if you like that method, then PowerShell is definitely the answer.

[OWTech]

Many thanks for your suggestions and knowledge. As it's the test environments I'm using reverse incrementals with weekly fulls so I don't have enough of a window to the destination disk resource to have the weekly fulls done by resource pool, datastore or host. The job cloning and powershell sounds good and i'll give that a go. Many thanks again to all who took the time.

[dellock6]

Why you want to use weekly full if you are using reverse incremental? In this method the last backup is always a full after the injection of the new data.

[OWTech]

Luca, that answers another of my questions. I thought I had seen somewhere that it was prudent to do a full every so often with reverse incremental in case one of the restore points got corrupted.

[dellock6]

Even if it's reverse, it always an incremental, so the chain is correct if every element is not corrupted. So yes having once in a while a full basically places a stop/restart in the chain, but I personally ever used this schema, since chances are you always have to restore from last backups and not from the oldest. Maximum chain I've used was 60 days, longer than that we suggested customer to peek a full once every month and store it offsite, and delete older reverse incrementals.