Veeam for Microsoft 365 memory usage

[sogapex]

Support case ID = 05618374
version 6.1.0.254 P20220825
Server = physical, 32GB of ram

Since the last update which hapened wednesday 7 september 2022 (proposed while entering Veeam Backup for microsoft 365), we can't run our "365" backup job anymore.

Little historic :(the jobs are runing by night)

wednesday, I installed the update (7:38), and then during the morning, I lost connection with the Veeam server (ping ok, but no RDP session and black screen in local) = I had to hard shutdown the server (12:45)
I didn't realize it could be Veeam for microsoft365 at this time.

Thursday morning = I get an email telling me the backup job for Microsfot365 failed.
There is a problem with the "jet" database
I opened a suport ticket and get in touch with the support team => we repaired the database file and I launched the job again @16:00
As soon as the job was running, the memory consumption rised to 100% (I didn't note the "idle" memory at this time)
And quickly, I lost control on the server

Again, I had to hard shutdown the server @18:34

I disable the job, and all other jobs run well this night (Veeam B&R and the other 365 job)
Now, I am a little stuck. I check the mmeory usage this morning, and it sits at more than 80% while the server is idle (no jobs running)

When I check the task manager or ressource manager, I can't see any process "eating" all this memory.
We can clearly see that the sum of all memory consumption displayed is far from approching 32GB

I am a little stuck here

[sogapex]

screenshot after a restart of the server (9% memory usage)


screenshot after several minutes (Veeam services running = 78% memory usage)

[sogapex]

so, we repaired again the database file (repository.adb) with "eseutil.exe" given by the support team.
The job is currently running, and the memory is only at 24%.

Eventually : I can't say it was or wasn't the update which was problematic. Maybe it is a coincidence, and the database file was already corrupted before the update.
But clearly, we can see that in normal conditions, there is no way to max out the 32GB of memory while backupping Microsoft 365 (in our case = less than 100 users, without onedrive nor sharepoint : only exchange online)

thanks to Alecs for his patience and support

[Polina]

Hi sogapex,

Thank you for the heads-up, and I'm glad that the situation is now resolved. I'll check your case details, and if needed push it to our QA team for additional research. If the issue proves to be systematic, we'll fix it with the next product updates.