Backup From Storage Snapshots - IBM

[crackocain]

Hi everyone.

We planned to use "Backup from Storage Snapshot" and "Primary Snapshot Orchestration" for quick restore. These two features will be used together.
1 hour RTO for Primary Snapshot Orchestration
1 day RTO for "Backup from Storage Snapshot"

IBM Storwize use copy on write mechanism. So if delete a snapshot in the snapshot tree other snapshots are going offline.

I know backup from storage snapshot create another snapshot and deletes it after the backup is complete. Is it affect Primary Snapshot Orchestration snapshots?

We plan backup 400 vm, create and delete too many "Backup from Storage Snapshot" requests. Are there any Veeam customer who use it this way and have no problems?

[Andreas Neufert]

There is no issue in IBM deleting any snapshot from the middle of a snapshot chain. So no issue with our processing there.

[crackocain]

Hi Andreas

I delete only one snapshot from Veeam console for empty space and other snapshots are going offline. That's why I'm afraid actually

[Andreas Neufert]

What do you mean by "goind offline" can you please explain? What does IBM support say about this?

[crackocain]

I use Storwize 8.1.3.6 firmware and Veeam 9.5.4.2866



When i try to delete middle snapshot only delete flashcopy mapping. Not execute volume delete command. (rmvdisk)



Storage snapshot volume offline.



And storage snapshot can not deleted.



After 10 minutes i try to force delete



After that another volume are going offline.



Deleting a snapshot in the middle makes the previous ones offline(maybe flashcopy mapping is deleted). Latest snapshots are good state (online)

[Andreas Neufert]

I have at the moment no such system to test, but I think you have another issue. If I remember right we create the volume just temporary from the snapshot for rescan and then remove it. If the from snapshot cloned volume is present then you can not delete the snapshot.

So the question is why the volume is still there and not automatically deleted after our storage scan (where we detect what VMs are on the volume).

Please open a support case and upload logs there. Let them know this Forum entry exist with details and as well share the ticket number here so that we can have a look at the support case as well.

[crackocain]

Case opened and logs are uploaded Case #03896955

Thank you Andreas...

[Andreas Neufert]

Thanks, I have added some internal comments to the case.

[gmajestix]

Not sure but maybe it is related to volume protection feature?

https://www.ibm.com/support/knowledgece ... lumes.html

[Andreas Neufert]

Indeed it is related to it.
Our delete process wait until the IBM volume protection feature allow us to delete it. This is why it takes "forever". Usually 30min is I think the default.
When you force delete the volume manually it has your described side effect.

[crackocain]

Yeah maybe but it suppose to delete smoothly. Because snapshot deletes are flawless when primary storage integration jobs. When i try to delete Storage Infrastructure tab this is happen. Because of Flashcopy IBM Storwize can not give permission to delete. For example primary storage integration deletes the snapshots volume with force removal.

[Andreas Neufert]

Please open a support ticket and upload our logs from that timeframe. Are you sure that those 2 are not the ones that you have force deleted manually as described above?

[crackocain]

Yeah those commands sent by Veeam. Ticket opened and engineer is connected. I'm uploaded new log and log level is 6.

[Andreas Neufert]

Can you please share the ticket number here so that we can comment there?