Snapshot timed out on SQL server cluster during 1st replication

[TDG]

Hi you lovely people. Happy Friday

I was setting up a new replication job and snapshot timed out on MS SQL server cluster VMs during replication. Other VM's on the same batch replicated without any issues this includes non clustered SQL servers.
Note this is the first time I was running this job. Maybe I should try this on a weekend when it's really not busy ?

The failed both VM's are the primary SQL node running Win 2019 DC, 4 vCPU, 128GB RAM and 1.25TB in size and running SQL server enterprise

The both VM's became un-responsive during this time approx. 20-25 mins
Error: Failed to create VM snapshot. Error: CreateSnapshot failed, vmRef vm-1072, timeout 1800000, snName VEEAM BACKUP TEMPORARY SNAPSHOT, snDescription Please do not delete this snapshot. It is being used by Veeam Backup., memory False, quiesce False
Error: Operation timed out.
Processing finished with errors at 28/05/2025 23:08:55

But the same failed SQL cluster is on daily Veeam job (different time) to wasabi for many months backing up without any issues

I have a Pure storage FC array maybe going forward setup storage snapshot backup/replication with Veeam ?

Please advise

[PetrM]

Hello,

I would start by researching the root cause: why does snapshot creation not work? I think you can test manual snapshot creation directly in vSphere, because Veeam simply acts as the requestor, it does not create or delete snapshots itself. You may also open a support ticket and share the case ID with us.

Regarding the idea of using storage snapshots, it is always a good way to protect workloads as it reduces the impact of backup and replication activities on the production environment, and the VMware snapshot exists for only a very short period of time. However, a VMware snapshot is still required.

Thanks!

[TDG]

Thanks PetrM

Case # 07718031

Like I said the same VM on Wasabi backup job works fine
May be some sort out SQL maintenance plan was running o the VM or a file is locked but I'm not sure

I have a remote session with Veeam support next week so I can show them

So if we use the storage snapshots you saying we still need a VMW snapshot for a short period of time ?
But this is much better compared to what I do now ? i.e. less impact on production VMs

I do have 10Gb p2p link just for Veeam replication/backup traffic from DC to HQ so B/W is not an issues

Or I need to put a cheaper Pure at HQ and do asynchronous replication using the above link so I can bring up VM's at HQ in case of a disaster