Sizing memory for Veeam server for SureBackup use

[DI_MAB]

All,
I have an HP DL370g6 with 8GB of DDR3 memory running Veeam. I have an application group of 10VMs that run in a SureBackup Job. As the VMs boot, it seems that the Veeam server memory gradually uses 7.95 of the 8GB of memory and then Veeam crashes (I'm assuming this is due to the memory issue). Is there a recommendation from Veeam published anywhere that is best practice for sizing memory appropriately for the Veeam server for SureBackup jobs? I'd like to avoid breaking the VMs into different Application Groups, as I want to test functionality between then and use it as a sandbox in the future.

What about SureBackup is so memory intensive on the Veeam server if all the VMs are booting on the ESXi host? Is this behavior normal? I've done some searching and can't find anything.

Any insight or experience would be appreciated.

[ThomasMc]

Read this post and setup in a similar fashion, not sure if its recommended to dump all the VMs into the same app group as I see a lot of that happening but its no good for small environments

[ThomasMc]

My apologies as my answer has nothing to do with you OP although not having so many VMs in you app group might help, do all 10 VMs need to be on at the same time?

[foggy]

Mark, I would say this behavior is not normal. We've tried to reproduce the SureBackup job starting the Application Group containing 10 VMs on the server with 2GB RAM and it started successfully leaving about 20% of memory free. Though our test lab setup could be different, I would suggest to collect logs and open a support case to investigate.

[abigot@serians.fr]

hello;

I've a similar problem with Surebackup :
Veeam6 is installed on a HP DL370g7 with 8GB (with Vcenter5, Vconverter and Trendmicro Officescan client).
While Surebackup Jobs, memory also gradually uses ~7.95 of the 8GB.
I created an Application group with only one DC and one Exchange 2010 : I had to increase timeout up to 1000 seconds for Exchange to start successfully
(with lowest timeouts, I've got error 10061 while runing mail script)

Increase application groups timeouts make U-air operations too long...
Is it normal to wait for more than 40 minutes for the 2 VM to start successfull in the VirtualLab ?
Is memory on Veeam server not enough ? (of course ESX for the Vlab is well sized and not overloaded)
It seems that VBR VirtualLab uses lot of memory to present deduplicated & compressed VMs to ESX.?

regards,

[foggy]

Arnaud, this is not normal, VMs start time should be comparable to one in production. I would also recommend opening a support case and have our staff investigate what is causing the delay with the logs files.

[creativeTechie]

I have the same issue. Did you get a resolution on this?

Thanks,

@creativeTechie

[GB@CFS]

We are also having a similar behaviour, running v6.1 with even just one Surebackup job. Server had 6GB and was sitting at about 5.95GB used. Bumped up to 8GB RAM and now memory usage is about 7.95GB.

Any resolutions for this?

[Vitaliy S.]

Unfortunately, none of the posters above has mentioned his support ticket number, so I cannot update this topic with the resolution. Please contact our support team for assistance.

[Bigpants2k]

Was there a fix for this?

I am running v7. with three VM's, the first two start fine, both DC's, but when exchange boots, it fails the power on test (even though I can log into the OS via console vSphere client). I have even un-ticked the two options, about ping etc and let the application start for 1000 seconds. But still no joy.

The whole process of powering them up takes ages.

If need be I can create a support call and post back here?

Thanks

[Vitaliy S.]

Allan,

Can you please clarify what fix you're asking for? Your issue seems to be unrelated to the OPs post. If you have any issues with SureBackup jobs, it would be more efficient to investigate it with technical support team directly.

Thanks!

[Bigpants2k]

Sorry, its just bugging me.

Ill raise a support request.