[RFE] Snapshot retention parameter on destination

[fjcloud]

Describe the feature you'd like to have. It can be a good thing to have a snapshot retention on destination.

What is the value to the end user? (why is it a priority?) in case that you have a corruption it will replicate it on the only snapshot available.

How will we know we have a good solution? (acceptance criteria) Add a parameter of number of snapshot retention you want on destination.

Additional context I have a big issue on my cluster1 that generate corruption on volumes, it take more than 1 hour to see it, unfortunately volsync has replicated corruption in my cluster2 and I have only one snapshot available on cluster2.

[JohnStrunk]

For handling corruption, have you considered SnapScheduler? (SnapScheduler would be on the primary cluster in this case)

[fjcloud]

So after we can transfer snapshot with VolSync on a destination cluster ?

[JohnStrunk]

No, VolSync only handles PVCs currently. The two could be used in a complementary way though. SnapScheduler would provide local snapshots to protect against application corruption, allowing quick recovery on the local cluster. VolSync would protect against cluster/namespace-level disasters by mirroring the PVC to a remote cluster.

I'm not suggesting that there's no role for a remote snapshot retention schedule, but the above may get you the protection that you need today.

[fjcloud]

Yes thanks for suggestion, will definitely take a look to SnapScheduler!