stgrule actiontype=copy for the tsm for virtual machine restore from tape copy

[fredric]

Ok, the problem is that the tape library is hooked up to the primary server (different sites). Today I only replicate "production data". In that case I would move the library and buy more disks to the target server.

Today I have only one node/container-stgpool for the entire VMware environment, so both production and other servers are stored at the same node name/stgpool. I recently realized that this causing problem for filespaces that not are replicated. The data in those filespaces are not expiring as it should according the policy because the node is enabled for replicating. So I'm planing to separate the production from the rest (as a start).

 

[dietmar]

i would not recommend stgrule copy on filespace level anyway . This is to difficult to handle/maintain.

If you do not protect the pool local to tape AND do not copy/protect all node/pools to the target - how are u able to repair a damaged pool ?

From the Backup Admin point :

* Keep enough free resouces in your environment, and keep it simple and strait forward.
* Backup all Systems to Disk and Tape , there will always be a missconfigration/use of a System ( Testsystem is a Production one ) wrong VM Tag and so on.
* Even if the customer rule says 14 days is enough , keep 21-30.
* Prepare in DR that Prod System are available fist, if wrong Tag then yes it is there but have to wait.

if you not consider this, backup admin job could be a pain.

 

[fredric]

Yes, it can be a pain sometimes I just work as a backup admin part of my time, I mainly spend my hours on aix, storage(file), databases.

I'm going to set up protection for the pool locally, that's the reason I want to separate the data into several pools (prod, test etc.)
Today we only protect all servers on the primary backup server. In case of a disaster, we will only have the production environment protected (primary backup environment unavailable). But I agree that we should aim to also protect the rest. If/when I start with local protection of the container-stgpools, all servers will be protected (disk+tape).
Today we keep the backups for months on disk.

Thanks for your input, it's valuable information for me (and hopefully for others).