Skip to main content
  • EN

Last week I had to troubleshoot a backup job that was using a brand new proxy and repository server. No other job was using these components. Nonetheless just 4 VMs were processed at the same time. All other VMs showed status “Pending”. As additional information, “Resource not ready: Active snapshots limit reached for datastore” was shown too in active job log.

After a short investigation we found out, all VMs were located on just one VMFS datastore. Fortunately there is a rather simple solution for this limitation. You can create a registry key on VBR server to increase the limit of active snapshots per datastore:

HKEY_LOCAL_MACHINE\SOFTWARE\Veeam\Veeam Backup and Replication\MaxSnapshotsPerDatastore 

For more details see this forum posts:

https://forums.veeam.com/vmware-vsphere-f24/active-snapshots-limit-reached-for-datastore-t19711.html

Nice detail: The change enters into force immediately – even the job is currently running!

Keep in mind, this is a global setting!

 

 

Just please make sure there is sufficient free space on the datastore to keep changes made by VMs during backup window.

Very useful 😎

And good advise, @haslund 👍🏼

Just please make sure there is sufficient free space on the datastore to keep changes made by VMs during backup window.

Absolutely! Also keep in mind, your volume can provide necessary performance!

Good to know!

Thanks for share dude...

Nice reminder! But make sure your production storage has the performance to handle it.

I recently saw a customer lose production after setting this value...

On sweet. This is a nice tweak to get performance/backup windows where the need to be! Of course @haslund  with the great advice. Thanks for sharing this @vNote42 !

Thanks for sharing @vNote42!

 

What I’m about to suggest doesn’t work when customer has exactly 1 datastore and no space for anymore, but sometimes it’s just a lack of knowledge rather than design choice:

It is still possible to get VMware to save snapshots to a different datastore, so as a bandaid, you could create new datastores purely for snapshots and reconfigure VMs that might need certain minimum performance requirements etc to use these.

 

This isn’t a one size fits all solution, but can come in handy.

Thanks for sharing @vNote42!

 

What I’m about to suggest doesn’t work when customer has exactly 1 datastore and no space for anymore, but sometimes it’s just a lack of knowledge rather than design choice:

It is still possible to get VMware to save snapshots to a different datastore, so as a bandaid, you could create new datastores purely for snapshots and reconfigure VMs that might need certain minimum performance requirements etc to use these.

 

This isn’t a one size fits all solution, but can come in handy.

Thanks for you input! Do you know Veeam counts the Snapshots really for the defined Snapshot-datastore? In my experience this settings is not very common, despite it was the default in some long-time-ago-versions.

That’s a good question that I’ve not come across the limit & had to mitigate with this! I also don’t think I have enough resources in my lab to create such a constraint.

 

I have seen from some notes that it’s output into the log what the current counter is, so I might be able to get away with 2x VMs, put them on same datastore but have one create its snapshots on another datastore and see what Veeam states.

As it sounds like Veeam is calculating this information itself, it might not be a workaround for such a scenario… My original intent when posting was helping people avoid filling their datastores due to too many snapshots, but I’ll need to see if this does bypass the snapshot limit…

To be confirmed!

This is a good tweak for sure as I have used it before but always made sure of space and the array can handle it. 

Just to get sure: In general I would not recommend to change this setting! Normally VMs are located on more datastores so this limit should not really matter.

Just to get sure: In general I would not recommend to change this setting! Normally VMs are located on more datastores so this limit should not really matter.

Yes especially if you are using Clusters and Storage DRS. This may not be a good idea in that case then as I have seen it cause issues for sure. 😜

Just to get sure: In general I would not recommend to change this setting! Normally VMs are located on more datastores so this limit should not really matter.

With vSAN you have one big datastore often….

Comment


Terms & Conditions