Commit Memory never goes down

[john.aubrey]

Hello,
I think this might be a Windows issue, but might be Veeam too. Starting in probably November, there has been 2 processes that continuously eat commit memory, while the Working Memory stays about the same, or at least within reason. The 2 processes are Veeam.Backup.Service.exe and Veeam.Backup.Agent.ConfigurationService.exe. It seems like they eat 1-2 GB per day until there is no more memory and the server comes to a halt and I have to reboot it. Veeam seems to work fine until it runs out of memory. I don't really understand the deep down workings on how memory is allocated with respect to the OS and applications requesting more. It does have a SQL Express instance on the server, but that doesn't take much RAM and DB is only 1 GB in size.

Servers are Server 2016 with Dec 2021 patches
Veeam is running v11 (we were going to go to v11a, but wanted not to throw that into the fire)

[PetrM]

Hi John,

The described behavior is not normal and I recommend to let our support engineers to troubleshoot the issue, please paste a support case ID over here. Most probably, process profiling or process dumps analysis are needed to explain this.

Thanks!

[john.aubrey]

I opened a case earlier this week. Working through it still. Case # 05235176. I sent a bunch of logs when I opened the case. I couldn't find anything here of similar cases. I was hoping someone ran across something before and could share there work.

[john.aubrey]

Hi all,
I wanted to update this for anyone that ever comes across this. It turns out our vulnerability scanning with Tenable.IO/Nessus was causing the issue. Every time there was a scan, those 2 Veeam processes would somehow consume more RAM. On normal months, we scan once a week, and a reboot at least once a month with Windows Updates. This last month we were scanning everyday looking for Log4J stuff apparently. We also didn't apply this months Windows Updates.