Dell\EMC SC series plugin and SC replication?

[raronson@dga.org]

We install the recently released SC series plugin for Veeam B&R. We have a primary san in our main office which is a 5020 and replication site which is an SC3000 and that's the reason for my post. After running the plugin enabled backups for 6 weeks we nearly ran the replication site out of space. I discovered a number of manually created snapshots that were set to never expire. Enough that we chewed up 10 tb of storage. Its only on the remote system, the primary san shows the volumes being frozen and released correctly.

I have some suspicions about what's going on but I think it's comes down to Veeam being unaware of the remote system. That makes it impossible to clean up after the backups when a snapshot is replicated. I talked to support at both Veeam and Dell. Most of the conversations were about using storage snapshots. I had to do a lot of explaining to support, even for tier 2 support.

My solution at the moment is to use powershell to expire the snapshots on the remote system after every series of backups. Has anyone seen this before?

Thanks, Rob A.

[veremin]

Hi, Rob,

Can you tell us what exact snapshot gets replicated to remote site? I think those are temporary snapshots created during backup activity that get replicated to secondary array and stay there forever cause backup server is not aware of storage replication configured, but still worth double checking?

Thanks!

[evilaedmin]

Rob,

Your replications have their own remote snapshot expiration policies.

You should be able to modify any replication to expire based on rules.

How to modify is in the Administrators guide under “ Modify a Snapshot Profile Expiration Rule for Remote Snapshots.”

[raronson@dga.org]

Thanks all. The snapshots are not created under a snapshot policy, the veeam backup snapin creates a manual snapshot. Once the backup is completed the manually created snapshot on the primary is expired although it will hang around until it’s completely replicated on the remote system. The remote system gets the same manually created snapshot as the primary system but it never will be expired.

[evilaedmin]

Hi Rob,

Apologies for misunderstanding. In my environment the 'expired' status of a snapshot seems to replicate successfully. I manually created a "never expire" snapshot, watched it replicate. Then set it to expired. Then created a new snapshot right after to trigger On Demand progression and replication. Storage Manager reports that the remote snapshot is also marked expired with same exact timestamp as original expiration.

By the time I logged into the destination storage center's unisphere UI to check, it was already deleted.

I hope you find a good root cause why yours aren't. Is data progression otherwise working fine on your devices?

[evilaedmin]

Rob,

One more question are you using async, sync, or semi-sync ("Replicate Active Snapshot") Replication ?

[foggy]

Hi Rob, looks like the plug-in doesn't set expiration settings for the temporary snapshots correctly (or doesn't set them at all). We will verify this behavior internally and report the issue to the Dell EMC team responsible for the plug-in so they address this in the next version. Thanks for the heads up!

[raronson@dga.org]

thanks so much for the acknowledgement. I guess i should keep working on my powershell script to remove the remote snapshots in the interim. And thanks to everyone who gave me input.

[integis]

Hi, I've had the same issue here, which is pretty annoying. Storage snapshots get created with no expiration date and have to be manually cleaned up. Has anyone had an update on this? I checked for a new version of the Dell plugin, but it doesn't appear there is one yet. I have just updated the firmware on my SC5020, and was likely going to have to open a ticket with Veeam and/or Dell soon.

[foggy]

Hi, I know that Dell EMC has an internal issue open for this but haven't heard of any news in this regard recently. I've contacted the Dell EMC plug-in development team for an update.

[integis]

Thanks foggy! I am making sure my Vmware hosts are updated before I open a ticket as well.

[nillywilly]

Hi, is there an update on this? We're about to implement Veeam Storage Snapshots on the Dell SC series. Just trying to make sure if this was a bug and if it's been fixed in the newer Veeam plugin or SC firmware.

[foggy]

Unfortunately, I still haven't got any response on that issue from Dell EMC engineering. Trying to reach out again.

[dmitri-va]

Hi,
we had the same issue and at some point we just disabled the storage integration. Per Dell support in the past, and per SCOS 7.5.1 release notes the issue is supposed to be fixed, even though the description is not an exact match:
"SCOS-61101 Vendor If a Veeam backup job fails, but the failed backup job is successful when it is retried, Veeam might create a snapshot without an expiration date on the Storage Center."
https://www.dell.com/support/manuals/en ... lang=en-us

We upgraded one of our SC SANs to that version, but are now getting this error when trying to backup with storage integration enabled (plugin v1.0.210):
"7/14/2021 10:00:57 PM :: Failed to create snapshot for LUN OR-CML_ESXvol14 Details: StorageCenterError - Unable to access snapshot profile information. Information not found."
Similar error when trying to do a restore from a snapshot.
Opened a support case with Veeam, but it has been siting unanswered since Monday.

[foggy]

Hi Dmitri, might be an unrelated issue. Could you please share the case ID?

[dmitri-va]

foggy: shared