PMS crash happening once a week

Plex Media Server
, ,
1

Server Version
Plex Media Server Logs_2023-02-20_13-11-40.zip (3.3 MB)
: 1.31.0.6654
Running in docker (linuxserver container) on unraid 6.11.5.

Using iGPU for hardware transcoding.

Hi,

I noticed a crash yesterday around 20:55. A restart of the container is required to get access to PMS again.

It seems to happen randomly around once a week.

Thank you for the help !

2

Quick look, only thing I noticed although I’m no expert is the database reporting long waits, eg:

Feb 19, 2023 21:04:11.395 [0x151a8c41fb38] WARN - [Req#795e] Waited one whole second for a busy database.

Anything in your unRAID syslog when it crashes?

3

I see nothing at the time PMS crashed.

Feb 19 15:00:16  crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 16:00:16  crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 17:00:16 crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 18:00:16  crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 19:00:16  crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 20:00:16 crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 21:00:16 crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 22:00:16  crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
Feb 19 23:00:16 crond[1691]: exit status 1 from user root /usr/local/sbin/mover &> /dev/null
4

Hmm weird, I run the official PMS docker under unRAID myself.

I presume you excluded lines showing the PMS container being restarted ?

5

No I just removed my server name.

6

Hmm, normally I see lines like this :

Feb 20 15:36:47 Tower kernel: docker0: port 6(veth011ab1b) entered blocking state
Feb 20 15:36:47 Tower kernel: docker0: port 6(veth011ab1b) entered disabled state
Feb 20 15:36:47 Tower kernel: device veth011ab1b entered promiscuous mode
Feb 20 15:36:47 Tower kernel: docker0: port 6(veth011ab1b) entered blocking state
Feb 20 15:36:47 Tower kernel: docker0: port 6(veth011ab1b) entered forwarding state
Feb 20 15:36:47 Tower kernel: eth0: renamed from vethcaff287
Feb 20 15:36:48 Tower kernel: IPv6: ADDRCONF(NETDEV_CHANGE): veth011ab1b: link becomes ready
7

So what should I try next ?

8

Sorry, not sure, you need help from Plex support I think

9

Thanks, let’s see if someone can come up with a suggestion.

10

Turn DEBUG logging back on. We can’t see anything that PMS is doing without it.

DEBUG logging will give at least a hint of why something is happening.

To improve the capture window,

  1. Stop PMS
  2. Add LogNumFiles="10" to Preferences.xml right before the closing /> on the last line.
  3. Restart PMS to make it effective.

This increases capture from 5 rollovers (50 MB) to 10 (100 MB) which will compress to about 6-7 MB ZIP file for the forums

When next happens, immediately pull logs ZIP file and attach

11

Just a quick update: I have not been able to catch a crash yet.

Also PMS is now on version 1.31.1.6733

12

Crashes never happened again after some networking changes.