We are experiencing slow increases in memory usage on our ArcGIS Monitor VM. Once it reaches ~75% memory usage the paging file usage for that VM also slowly begins to increase as well. If this is left unattended the Monitor instance becomes unstable, crashes and requires a reboot. Once restarted the application stack restores correctly and behaves normally. The resolution to it is to ensure we reboot the Monitor VM every fortnight or so.
It is highly likely this issue is due to my infrastructure environment and/or custom extensions we are calling from the app.
Has anyone else encountered similar issues?
You need to whitelist the ArcGIS Monitor executables within the Nessus scanner so it (Nessus) doesn't interfere with ArcGIS Monitor operations. It is not uncommon for virus scanners to quarantine ArcGIS Monitor files.
Exceptions/whitelisting for ArcGIS Monitor application and related *.exe for for services may be needed. This would include:
We're also coming across the same issues with Nessus causing the ArcGIS Monitor 10.8 Services / Processes to Terminate unexpectedly.
We have tried excluding ports that Nessus scans related to Monitor:
As Tenable indicate this does happen to some applications, but the fault is with how the application handles the requests. Unfortunately this exclusion did not stop Monitor from crashing.
There are additionally many dynamic ports utilised by node.js. which we were not able to exclude.
We have been advised by IT the following relating to further Nessus exclusions:
The only workaround we've found is to configure recovery settings for these services so that they restart after failure.
Ideally these services would not fail when the server is scanned.