Server Version#: Version 1.18.4.2171
Logs attached
Seemingly random server shutdowns - not sure what’s causing it. It happens with all versions of 1.18. Reverting to 1.17.x fixes the problem (I think, there have been a few random crashes of that as well). I’ve completely uninstalled/reinstalled and rebuilt the database.
You might want to check out this post:
Thanks - I do not have an Intel Graphics adapter, nor are the crashes tied to any event that I can ascertain (i.e., it does not crash when transcoding or when any stream starts).
The closest thing I see is the TV Tuner (HDHomeRun) logs starting a shutdown on their own (I think):
Dec 22, 2019 13:04:20.427 [0312] DEBUG - Unable to lock named mutex F8EEB26D-AB23-4061-8250-130AB5285A4A: 0x80
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term. Stop all devices
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term. Shut down all device auxes
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] bda_aux_t::term
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] web_server_t::stop. Web server is stopped
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term. Stop all devices
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term. Shut down all device auxes
Dec 22, 2019 13:04:20.442 [0312] INFO - [I] broker_t::shutdown. Shutdown is completed.
Also, it is a headless computer (connecting through RDP), but the crashes happen regardless of whether I am connected or not.
You and me are having very similar issues. Take a look at this thread and see if your issue seems similar: https://forums.plex.tv/t/pms-v1-18-1-2019-crashed-2-times-in-3-days-win10
Mind you, I was having “crashing” almost daily on earlier versions, and had it mostly fixed by the 2171 release. I’m still seeing the issue occasionally, but nothing like before.
You should check to see if the PlexMediaServer.exe process is actually crashing, or actually just hanging. In my case Plex would continue to run, but would become unresponsive for up to 45 minutes at a time, eventually fixing itself without intervention. If your situation matches up with mine, then it would be great if you could submit logs and a process dump to the team as outlined in the thread I linked.
Hi Justin - mine is a bit different. The process itself just shuts down. It looks like (in the tuner log) a server shutdown is called by something. Mine definitely doesn’t hang like that, and the process is completely closed.
I’m pretty sure this is in some way related to the DVR / Tuner functionality.
Had another crash today - similarly accompanied by the same tuner issue:
Dec 22, 2019 13:04:20.427 [0312] DEBUG - Unable to lock named mutex F8EEB26D-AB23-4061-8250-130AB5285A4A: 0x80
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term. Stop all devices
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] device_manager_t::term. Shut down all device auxes
Dec 22, 2019 13:04:20.428 [0312] INFO - [I] bda_aux_t::term
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] web_server_t::stop. Web server is stopped
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term. Stop all devices
Dec 22, 2019 13:04:20.441 [0312] INFO - [I] device_manager_t::term. Shut down all device auxes
Dec 22, 2019 13:04:20.442 [0312] INFO - [I] broker_t::shutdown. Shutdown is completed.
In your shoes I would downgrade from this beta release to the previous (now stable) release (v1.18.3).
Thanks. I’ve tried every 1.18.x release with no success. The last 1.17 release is mostly stable, but hangs every 3rd to 4th day with similar logs.
Any update from the Plex team? Were my logs attached properly?
Bit of an update - two crashes in the past 3 days which, when restarted, could not see the mapped network drive. I had to restart the computer to get it to show up again in Plex (Windows could see/access it perfectly).
Plex is not running as a service.
I’m having the same problem and ended up having to revert to 1.18.4.2164. Still was having Plex shut down when I would try to use hardware transcoding. Disabled and enabled hardware transcoding, now its working again with hardware transcoding. I don’t think they fixed the hardware transcoding issue.
I don’t think my issue is related to hardware transcoding, since it happened both on an AMD system and with it turned on and off. The crashes were happening with no clients streaming either.
I did a fresh install on a completely different computer with a fresh installation of Windows 10 as well. This computer has a GTX1050ti in it and it’s an Intel chipset. I got a similar error, though I’m traveling and wasn’t able to get logs. Also, it’s been up for a week without crashing this time.
I’m beginning to think it’s an issue with the database, though I did a complete reinstall.
This topic has been open for nearly three weeks (admittedly over the holidays) - will someone from the @Plex team respond?
Still no response?
Have you disabled hardware acceleration in transcoder server settings?
There are no logs attached that I can see
I would like if possible two sets of logs - one on latest 1.17.x that does not crash and on one on 1.18.x that crashes - logs collected after doing same action on both tests
I’ve been having a very similar issue with random crashes about once or twice a day. It just happened again this morning. I have attached my logs for reference. I’m running version 1.18.5.2309. I just tried shutting off hardware acceleration, so I’ll see if that makes a difference.
Thanks for the response - to your two questions:
One other note - I migrated the exact settings / data to another server as mentioned above, and I have experienced no crashes in a few weeks of operation.
Not a crash in the true definition (process crash) but a deadlock that rendered the server unusable after a while - leading to 503 errors
Jan 31, 2020 11:23:07.796 [0096] DEBUG - Completed: [64.136.236.218:57867] 503 GET /media/providers (4101 live) TLS GZIP 0ms 451 bytes
Jan 31, 2020 11:23:07.978 [2120] DEBUG - Completed: [127.0.0.1:62344] 503 GET /media/providers (4100 live) TLS GZIP 0ms 451 bytes
This shows how many requests got queued (4101 live)
Deadlock investigations are complex and require a process dump and also list of connections - so if this happens again - please get the connections list copied to a text file and then take a process dump of Plex Media Server.exe process and then capture the logs
See this post for diagnostics for deadlocks
PM’d you a set of logs from 1.18.x (back from December). Running 1.17.x now, will upload when I have a set of logs showing a crash.
No crashes in 2 weeks on 1.17.x (but no connections to the server either).
@sa2000 I sent the crash logs on 1.18.x, any luck?
