I am the administrator of a single server SharePoint 2019 on-premise farm with a separate server running the SQL side of SharePoint. I built this farm. I have domain admin permissions and local admin permissions.
The problem is that fabric_counters_X_000001.blg files (located in C:\ProgramData\Microsoft\SF\Log\PerformanceCountersBinary), query_traces_X_X_1.etl files (located in C:\ProgramData\Microsoft\SF\Log\QueryTraces) and fabric_traces_X_X_1.etl, fabricdeployer-X.trace, FabricSetup-X.trace, lease_traces_X_X_1.etl files (located in C:\ProgramData\Microsoft\SF\Log\Traces) are being uncontrollably created. (The Xs all represent a series of identifying numbers.) Within a month or two my SharePoint server's C drive runs out of space because of these files.
I have figured out that in Performance Monitor I have a User Defined Data Collector Set called FabricCounters. Within this data collector set I have a FabricCounters Performance Counter. Also, under Performance Monitor, I have Event Trace Sessions called FabricTraces, FabricLeaseLayerTraces, FabricAppInfoTraces, and FabricQueryTraces. Finally, I have a task in Task Scheduler under Task Scheduler (Local) -> Task Scheduler Library -> Microsoft -> Windows -> PLA called FabricCounters.
So having all of this figured out I should be able to manage these logs, right? NOPE!!! I have opened the Task Scheduler "as admin" disabled and deleted the FabricCounters task and after a couple of minutes it auto-magically comes right back and immediately starts running! I have opened Performance Monitor "as admin", deleted the FabricCounters performance counter, and data collector set. Both, eventually re-appear exactly as they were created. I have tried stopping the data collector set, editing the performance counter log modes, and manually starting the data collector set. The FabricCounters data collector set eventually auto-restarts itself and my edits are no longer there. When I try to stop, edit the log mode, and start the event trace sessions, I get an error message "When attempting to start the Data Collector Set the following system error occurred: The parameter is incorrect." The event trace session eventually restarts itself and my edits are gone.
I have gone into Powershell (as administrator) and attempted to use logman to change, disable, and delete the FabricCounters data collector set. Same result.
I have been searching the 'net for days researching Microsoft AppFabric 1.1 for Windows Server and Microsoft Service Fabric to find hints about what I could change in their config files that might be controlling these data collector sets and event trace sessions. I have found similar situations (No, I'm not using Azure and no the articles about these files building up in Lync servers do not help), but nothing applicable to my situation.
Is anyone out there familiar with Microsoft AppFabric for Windows Server and Microsoft Service Fabric specifically for on-premise SharePoint 2019 server that can help me tame these counters and sessions and the uncontrollable log files they create?
