Used to happen a lot. Then stopped. Now, last week or two seems to crash randomly once a day or so. I don’t see anything in the windows logs standing out.
How can I get to the bottom of this? How to create log and/or where to upload? This is getting frustrating…
I need the logs immediately after a restart following a crash. Also please turn off verbose logging leaving just debug logging on. The log files filled up within 4 minutes
The logs were captured at 21:35. The server was restarted at 20:54
I presume the crash happened just before 20:54
The oldest log time i have is 21:31 - 37 minutes after the crash
After disabling verbose logging please restart the server to get fresh logs created
Then immediately after next crash and restart, get the logs
Sounds good. Disabled the auto-restart app, unchecked verbose, enabled debug, and restarted the server. Next time it crashes I will manually reboot the plexmediaserver.exe and instantly download the logs for upload. I am on most current install too, version 1.16.5.1488. Thanks sa2000
The Plex Media Server process appears to have run out of memory with failure to allocate memory just before crashing
Aug 20, 2019 21:27:33.865 [36320] ERROR - Thread: Uncaught exception running async task which was spawned by thread 38148: bad allocation
Can you tell if there is an empty dmp file within this path
C:\Users\Matt\AppData\Local\Plex Media Server\Crash Reports\1.16.5.1488-deeb86e7f\PLEX MEDIA SERVER
When there are memory issues - then Plex Media Server may also fail to get a crash report created
The logs show there were issues much before that with build up of requests - .
If the crash is predictable, we can try and get diagnostics before it happens
One way to check for build up of requests is to get the list of current connections from the server
Substitute the server token for the xxxxxxxxxxxxx
If you get back hundreds or thousands of connections listed then that would be indicative of a problem and we can then get the logs and a process dump.
The logs show there were about 300 connections at the beginning of the log and over 600 when we crashed
copy the responses of the connections requests to text files and save noting down the time = use filenames to give times when they were done