0

We use a NetApp FAS 6-series filer with an application that spans multiple volumes. For backup purposes I would like to create a consistent snapshot that spans these volumes at the same point in time (or at least with an extremely low delta); additionally, we'd like to to use SnapMirror to replace the production environment to test volumes.

The problem is in creating a consistent snapshot/SnapMirror, since these commands are not transactional and do not take multiple parameters. I tried scripting consecutive "snap create" or "snapmirror resync" commands via SSH, but there's always a 0.5-2 second difference between each snapshot. It's currently "good enough", but I'm seriously concerned about the consistency impact with increased load (we're currently in pre-production).

Has anyone managed to create a consistent snapshot that spans several volumes? How did you pull it off?

1 Answer 1

1

Do the snapshots need to be triggered manually? You might be able to schedule them to start simultaneously via Data Ontap's administrator interface if that's acceptable in your scenario. There are hardware platform specific limitations on the number of simultaneous transfers that a given filer model can support. These are detailed in the Data ONTAP Data Protection Online Backup and Recovery Guide.

1
  • That's an interesting idea - hadn't thought of that - but unfortunately yes, these snapshots need to be triggered manually (since we migrate production data to a test environment, all data generated after the snapshot is deleted on the next resync. We have to consider this every time we sync, hence the manual trigger). You did give me the idea to rewrite the scheduler to, say, 10 minutes hence, wait for the process to complete and then disable it. I wonder how hard this would be to script... Mar 12, 2010 at 0:43

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .