One particular page in our old monolith .NET Framework 4.8 webapp gets the connection_abandoned_by_reqQeue error thrown to the HTTPERROR log and immediately after IIS crashes and 5011 is logged, which kicks out all of our users. We migrated this app to Azure in December and it's been running fine till 7/24. We did patch on 7/21 with KB5062557, KB5062068, servicing stack 10.0.17763.7557, and dotnet 8.0.18.0 (not for this app), but have created a server from a backup that didn't have those applied but we still get the connection_abandoned_by_reqQeue/5011 IIS crashes.
We were pretty blind to the 5011/blanket event at first and saw it hit during dotnet 8 patching for other apps, so we tried repairing that which did nothing.
We've disabled EPO temporarily, no help. Never saw anything in trellix logs anyway, just a straw grasp.
We also increased the pagefile size to best practice settings, which did nothing.
We've both manually hunted for events lining up with timing and queried everything in event viewer with posh for anything with our pool name, and nothing lines up as a precursor clue.
We've added another server to the pool to alleviate traffic, but we weren't seeing high cpu or mem consumption anyway. Memory consumption was close enough to the default settings for the page file, so that's why that was attempted, but did not resolve. We've used debug diagnostics to create dump files but found no good evidence... We can't recreate this in development or stage.. Help me Obi Wan! (and thanks for any guidance)
