Shared memory connection has been forcibly closed by peer

[NorthGuard]

I have a new install of B&R 11 on a dedication device ( i7-8700T with 16GB RAM and SSD Drive ).

It backs up a physical server via agent to a SOBR - the local repo is a USB-attached drive and then the capacity repo is Wasabi S3.

The initial backup to local storage is fine and has no issues, but the copy to Wasabi consistently fails after roughly 7 minutes and 20-30 seconds of running:
9/9/2022 2:10:34 PM :: Shared memory connection has been forcibly closed by peer.

I have found this in the event logs:
Faulting application name: VeeamAgent.exe, version: 11.0.1.1261, time stamp: 0x621ec21c
Faulting module name: ntdll.dll, version: 10.0.19041.1806, time stamp: 0x1000a5b9
Exception code: 0xc0000005
Fault offset: 0x0000000000063416
Faulting process id: 0x3a80
Faulting application start time: 0x01d8c47662992ffe
Faulting application path: C:\Program Files (x86)\Veeam\Backup Transport\x64\VeeamAgent.exe
Faulting module path: C:\WINDOWS\SYSTEM32\ntdll.dll
Report Id: 4e22003b-5963-4f8f-8683-2c30d1fab927
Faulting package full name:
Faulting package-relative application ID:

I have an open support case ( ticket 05618883 ), but they are telling me the computer is underpowered for what I am trying to do and that I need more CPU and more RAM. However, when I look at the performance statistics while tiering is running, CPU and RAM never get anywhere close to be at their max.

I have already installed the latest Windows feature update and patches... removed and reinstalled Veeam... removed AV.... and it still crashes at the same point.

For what it's worth, I have an identical device at another location that is not crashing.

I'm stumped and don't feel like I am getting very far with support.

Has anyone seen this before?

[PetrM]

Hello,

The lack of resources on the server could be one of the possible reasons for this misbehavior but I think it would be better to ask your support engineer to dig a bit deeper into the issue and examine performance statistics. Also, please keep in mind that you can escalate a support case if you're not satisfied with how it's being handled. However, as far as I see the case was opened yesterday and most probably our engineers need to have more time for research.

Thanks!