So far, after disabling Webhooks for my server I’ve not had the lockup. Not quite at 24hrs of uptime yet and it’s a few hours from when it normally happens, but if I can get another 12 hours out of it I’d be happy that doing that has solved the issue.
Previously running Plex for FreeBSD 1.23.2.4656, I had been experiencing mysterious occasional lockups requiring a server restart in the early morning hours. It might happen once or twice a week, but infuriatingly, it would often happen again shortly after a restart.
The problem manifests itself as clients going yellow-exclamation-mark on all the libraries in clients. When logged in via app.plex.tv Plex Web, the server goes from green lock-in-circle to red exclamation. The server is live on a real IP, static IPv4, on the Internet. There’s no firewall, no NAT involved. The host hasn’t been rebooted, and has been configured the same way for nearly a decade. The daemons are still running on the backside, and usually it is possible to connect to the IP:32400/web directly.
This morning, it kept locking up on a very quick cycle, and after several restarts, I updated to 1.25.3.5385, hoping for a quick and easy fix. No such luck. After letting it do its database migrations for quite some time, it came online, and then quickly locked up again. I spent some time, found this thread, disabled webhooks, killed an old iOS sync job, no joy.
So I sat there watching various things, and I noticed that the DVR was downloading its schedule, would get to about 98%, and then that was when everything was going to hell. This was consistent through several additional server restarts.
I removed the DVR. Things got substantially better. Perhaps not fixed entirely. I did see the client here go all yellow-exclamation for several minutes, but it then CAME BACK without a server restart. I wasn’t able to identify why.
this only issue being that once the server is in bad shape, I can’t get to the X-Plex-Token. So I just opened the url you want once I restarted and I’ll just have to reload the page I guess when it crashes again. I am right?
This is when your server is not responding but still running? Is there an error code? You’ve attempted this on browser that is on the same machine as your server?
Disabling the options didn’t fix the issue. Can’t watch a movie fully and it crashes. This is very annoying. Used to work for years without any issues.
I have uploaded the logs and let me know when a fix is available.
yes, “localhost”. There is no error, the browser is attempting to establish the connection. I can curl to get you something more definitive (my guess is that the port listener is down.)
If you still have issues on the current releases please run with debug logging enabled - restart the server after enabling debug logging and capture logs zip and upload after next occurrence
There was a crash at Mar 23, 2022 22:01:35 and I have located the crash dump
I have also located other crash reports - do you happen to have any logs saved that cover these times
Mar 18, 2022 23:08
Mar 15, 2022 23:40
Mar 12, 2022 20:02
The dump for the 23rd does not give complete backtrace and I am hoping that it may be similar to the 18th March crash which has a better dump file
Nothing specific for those time dates but I have exported all the logs going back to Feb
There was another crash overnight aswell so you should have another fresh crash Logs.zip (3.6 MB)
Thanks - I can see crashes occurred at these times
Mar 25, 2022 01:10:16
Mar 21, 2022 23:19:23
Mar 20, 2022 21:28:09
The Plex Media Server main log files cover Mar 24, 2022 19:27 to Mar 25, 2022 10:15:04
so covering the crash at Mar 25, 2022 01:10:15
The crash reporting uploads do not always succeed - even when logging it as successful
Could you in terminal session find the $TMPDIR for the Plex Media Server user account an see if you can find files named