Plex Docker crashing everynight - Tried Many Solutions

Plex Media Server Desktops & Laptops
, ,
21

I am also having this issue. Seems to happen at night only. Docker container logs show nothing useful. Docker container doesn’t but becomes completely unresponsive until I restart it a few times. On latest version

22

I am also having this issue, only recently (~7 days). I’m running the docker under my unraid server. I notice memory is maxed out every time plex becomes unresponsive, like a memory leak? restart of docker frees up the over 10gig of ram, and everything is good for 8-48hrs

23

I broke down the day after I posted this and started over, I kept my old AppData in case anyone ever comes up with a possible solution. So the only thing I can figure is that it was corruption inside the database that the SQLite repairs we’re unable to pick up on.

It’s been 9 days without a crash where it used to go down daily, so far so good.

I lost 3 years of watch status across 20 odd users and a lot of custom work on playlists/collections. I’m still hoping there is a solution out there.

24

I found it easier to switch to another docker container. It was easy enough to switch over and retain watch status, collections, posters, etc

1 Like
25

I’m assuming you haven’t actually read the full first post, that was one of my first moves. My problem followed me over with a new container.

26

I did the exact same thing a while ago and it eventually started happening again and the server in question is only used for Live TV & DVR… so it may just be a coincidence.

27

I also haven’t had any crashes on 1.23.5.4801 … yet … but I put PMS behind a reverse nginx proxy that is only accessible via subdomain, not IP addess so it would be much more difficult if malicious intent is involved via open remote access ports… reason is there are few methods to pull the subdomain it sits on when zone transfers are turned off in DNS.

Not saying it is malicious but I’m trying to rule it out.

28

I deleted the Docker image, but retained data folder, then reinstalled new docker. I still have a memory leak issue, is this the same problem others are having?

29

Yep, I switch from official plex docker image to linuxserver’s and I’m still having issues

30

@BigWheel any chance you can get the right people to look at this and help investigate the issue? I don’t remember who the docker “expert” is of the Plex employees/ninja’s that participate in the forums.

Thanks,
Shark2k

31

Hey @shark2k I should be able to help you out. I think the next step here is to attach lldb to the PMS process and get a backtrace of the crash when it next occurs. I can help you out with some steps of your keen to try this out?

32

Hey @chrisallen, thanks for replying.

I’m actually not experiencing this issue, I was just trying to help out @snowblind182 get some attention on this thread from an employee that would be better suited to help out. Hopefully snowblind182 or one of the other users that posted in the thread can work with you to be able to dig more into this.

Thanks again for popping into the thread though!

-Shark2k

33

Hi I’m currently experiencing the deadlock issue right at this moment, is there something I can do to capture and send you?

Edit: I couldn’t wait any longer so I had to reboot.

34

Here are some instructions to capture a back trace of the crash which will help us investigate from our side.

  1. Configure your docker container to run in privileged mode or give it CAP_PTRACE
  • On Unraid simply edit the container and check the Privileged box then save/apply
  • On other systems, you simply add --privileged=true to the docker run command --or-- --cap-add=SYS_PTRACE
  1. open an interactive bash command line in your running container
# docker exec -it <name-of-plex-container> bash
  1. Install lldb
# apt update
# apt install lldb
  1. Find the current PID for /usr/lib/plexmediaserver/Plex Media Server 
# ps aux

  1. Start and attach LLDB to the currently running PMS Process
    lldb -p

  2. Once LLDB is connected, hit C to continue running Plex Media Server.

  3. Once the crash occurs, please type bt all and then copy the entire output (it will be long) and paste it into a text file or save it as a https://gist.github.com/

  4. Share it here along with your latest Plex Media Server.log :slight_smile:

  5. You can now exit the debugger and restart the container

2 Likes
35

Add “–restart=unless-stopped --no-healthcheck” to your advanced view extra parameters. I’ve had similar issues and that seems to have fixed it for me with the offical plex docker container.

36

I have been up for 4 days now without a crash. The last Plex docker update may have been my fix. Thank you for looking into this thread @chrisallen

37

The problem is that the docker doesn’t crash, plex just becomes unresponsive with a bunch of connections stuck at “Pending Schedule - Paused”. Usually what fixes it is a full reboot of the server and a database optimization. but then it can come back minutes later or hours later.

I can’t run the “bt all” command.

(lldb) c
Process 309 resuming
(lldb) bt all
error: invalid thread
(lldb)
38

This is because the process is not paused/crashed. You can’t take backtraces from threads while they’re executing since they’d be changing out from under you :slight_smile:

You need to wait for the the process to crash then grab the backtrace bt all

39

What’s going on is Plex doesn’t crash at all as in segfault no matter how long you wait. It will deadlock and become unresponsive and sit there until a forced restart.

1 Like
40

I thought I would chime in and say I am also having problems on Unraid. It’s not nearly as frequent as the OP’s problem, but weekly or biweekly Plex just crashes. From what I can tell in the logs, Plex continues on just fine, but libraries are inaccessible, and the WebUI is also inaccessible. My logs from last night till today at 6:20pm show events regarding a network change, with very long gaps in between. The shutdown about an hour ago is me restarting the container.

Jul 24, 2021 09:10:10.952 [0x14c735fb6b38] DEBUG - NetworkInterface: received Netlink message len=84, type=RTM_DELADDR, flags=0x0
Jul 24, 2021 09:10:10.952 [0x14c735fb6b38] DEBUG - NetworkInterface: Netlink address message family=2, index=13, flags=0x0
Jul 24, 2021 09:10:10.958 [0x14c735fb6b38] DEBUG - Network change.
Jul 24, 2021 09:10:10.959 [0x14c735fb6b38] DEBUG - NetworkInterface: Notified of network changed (force=0)
Jul 24, 2021 09:10:10.967 [0x14c735fb6b38] DEBUG - Network change notification but nothing changed.
Jul 24, 2021 20:12:14.076 [0x14c735fb6b38] DEBUG - NetworkInterface: received Netlink message len=84, type=RTM_DELADDR, flags=0x0
Jul 24, 2021 20:12:14.094 [0x14c735fb6b38] DEBUG - NetworkInterface: Netlink address message family=2, index=13, flags=0x0
Jul 24, 2021 20:12:14.094 [0x14c735fb6b38] DEBUG - Network change.
Jul 24, 2021 20:12:14.104 [0x14c735fb6b38] DEBUG - NetworkInterface: Notified of network changed (force=0)
Jul 24, 2021 20:12:14.111 [0x14c735fb6b38] DEBUG - Network change notification but nothing changed.
Jul 25, 2021 07:15:04.989 [0x14c735fb6b38] DEBUG - NetworkInterface: received Netlink message len=84, type=RTM_DELADDR, flags=0x0
Jul 25, 2021 07:15:04.997 [0x14c735fb6b38] DEBUG - NetworkInterface: Netlink address message family=2, index=13, flags=0x0
Jul 25, 2021 07:15:04.997 [0x14c735fb6b38] DEBUG - Network change.
Jul 25, 2021 07:15:05.012 [0x14c735fb6b38] DEBUG - NetworkInterface: Notified of network changed (force=0)
Jul 25, 2021 07:15:05.013 [0x14c735fb6b38] DEBUG - Network change notification but nothing changed.
Jul 25, 2021 18:00:40.024 [0x14c735fb6b38] DEBUG - NetworkInterface: received Netlink message len=84, type=RTM_DELADDR, flags=0x0
Jul 25, 2021 18:00:40.027 [0x14c735fb6b38] DEBUG - NetworkInterface: Netlink address message family=2, index=13, flags=0x0
Jul 25, 2021 18:00:40.027 [0x14c735fb6b38] DEBUG - Network change.
Jul 25, 2021 18:00:40.043 [0x14c735fb6b38] DEBUG - NetworkInterface: Notified of network changed (force=0)
Jul 25, 2021 18:00:40.060 [0x14c735fb6b38] DEBUG - Network change notification but nothing changed.
Jul 25, 2021 18:20:27.086 [0x14c73819ab38] DEBUG - Shutting down with signal 15 (Terminated)
Jul 25, 2021 18:20:27.086 [0x14c73819ab38] DEBUG - Ordered to stop server.
Jul 25, 2021 18:20:27.086 [0x14c73b04ec40] DEBUG - Stopping server...
Jul 25, 2021 18:20:27.086 [0x14c73b04ec40] DEBUG - HttpServer: Stopping server.
Jul 25, 2021 18:20:28.086 [0x14c73b04ec40] WARN - Timed out waiting for server to finish.
Jul 25, 2021 18:20:28.086 [0x14c73b04ec40] DEBUG - Server stopped.
Jul 25, 2021 18:20:28.092 [0x14c73b04ec40] DEBUG - NetworkService: Stopping advertisement.
Jul 25, 2021 18:20:28.092 [0x14c73b04ec40] DEBUG - Killing plug-ins.
Jul 25, 2021 18:20:28.092 [0x14c73b04ec40] DEBUG - [com.plexapp.system] Killing.
Jul 25, 2021 18:20:28.095 [0x14c73819ab38] WARN - JobManager: Could not find job for handle 318
Jul 25, 2021 18:20:28.099 [0x14c73b04ec40] DEBUG - ChildProcessMonitor: Waiting for Plex Tuner Service to exit.
Jul 25, 2021 18:20:28.099 [0x14c73b04ec40] DEBUG - [JobRunner] Signalling job ID 370 with 2
Jul 25, 2021 18:20:28.231 [0x14c73819ab38] DEBUG - Jobs: '/usr/lib/plexmediaserver/Plex Tuner Service' exit code for process 370 is 0 (success)
Jul 25, 2021 18:20:28.231 [0x14c735ca9b38] DEBUG - [ChildProcessMonitor] Plex Tuner Service exited.
Jul 25, 2021 18:20:28.237 [0x14c73b04ec40] DEBUG - Database: Shutting down.
Jul 25, 2021 18:20:29.803 [0x14c73631cb38] DEBUG - Stopping file watcher for com.plexapp.system