Recently switched from unRAID docker to bare-metal Debian 9 for PMS.
Logs are full of lines like there:
Jun 23, 2018 03:43:22.795 [0x7fc5e3663700] INFO - Plex Media Server v1.13.2.5154-fd05be322 - ubuntu PC x86_64 - build: linux-ubuntu-x86_64 ubuntu - GMT -04:00
Jun 23, 2018 03:43:22.796 [0x7fc5e3663700] INFO - Linux version: 4.9.0-6-amd64 (#1 SMP Debian 4.9.88-1+deb9u1 (2018-05-07)), language: en-CA
Jun 23, 2018 03:43:22.796 [0x7fc5e3663700] INFO - Processor Intel(R) Xeon(R) CPU E5530 @ 2.40GHz
Jun 23, 2018 03:43:22.796 [0x7fc5e3663700] INFO - /usr/lib/plexmediaserver/Plex Media Server
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] ERROR - Butler: failed to start deep media analysis job
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Killing job.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Butler: Scheduling deep analysis for: Didn't You Used to Be . . . Wait . . . Don't Tell Me
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - JobRunner: exec of /usr/lib/plexmediaserver/Plex Media Scanner cancelled. In shutdown state.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] ERROR - Butler: failed to start deep media analysis job
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Killing job.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Butler: Scheduling deep analysis for: Room for Improvement
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - JobRunner: exec of /usr/lib/plexmediaserver/Plex Media Scanner cancelled. In shutdown state.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] ERROR - Butler: failed to start deep media analysis job
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Killing job.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - Butler: Scheduling deep analysis for: Alarmed by Burglars
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] DEBUG - JobRunner: exec of /usr/lib/plexmediaserver/Plex Media Scanner cancelled. In shutdown state.
Jun 23, 2018 03:43:22.793 [0x7fc5b83fe700] ERROR - Butler: failed to start deep media analysis job
Jun 23, 2018 03:43:22.794 [0x7fc5b83fe700] DEBUG - Killing job.
Jun 23, 2018 03:43:22.794 [0x7fc5b83fe700] DEBUG - Butler: Scheduling deep analysis for: Hey Dude!
Jun 23, 2018 03:43:22.794 [0x7fc5b83fe700] DEBUG - JobRunner: exec of /usr/lib/plexmediaserver/Plex Media Scanner cancelled. In shutdown state.
Jun 23, 2018 03:43:22.794 [0x7fc5b83fe700] ERROR - Butler: failed to start deep media analysis job
Jun 23, 2018 03:43:22.794 [0x7fc5b83fe700] DEBUG - Killing job.
I’ve attached logs, any help would be greatly appreciated. Not sure why it’s dying every night.
FYI - I have turned of deep analysis in the tasks…
To add to @trumpy81 post - it appears that there was a signal to shutdown the server at 03:43:02.861
Jun 23, 2018 03:43:02.861 [0x7f157a36e700] INFO - [I] Received SIGTERM: Software termination signal.
But the scheduled tasks activities continued to run and was attempting to schedule analysis jobs but these did not start because the server was marked as being shutdown
May be system logs have something around that time - there is no indication that it was a crash at 03:43:02
The server was relaunched at 03:44:47. Did you relaunch it or was it simply that the shutdown and restart process took that long (from 03:43:02 to 03:44:47) ?
I am surprised the last Plex Crash Uploader.log file was created Jun 22, 2018 08:07:38.486 - this was the time of the previous launch of Plex Media Server. At 03:44:49 on 23 June there should have been a new log created
Jun 23, 2018 03:44:49.778 [0x7f82c73ff700] DEBUG - Job running: '/usr/lib/plexmediaserver/CrashUploader' '--directory=/var/lib/plexmediaserver/Library/Application Support/Plex Media Server/Crash Reports/1.13.2.5154-fd05be322' '--version=1.13.2.5154-fd05be322' '--platform=linux-ubuntu-x86_64 (4.9.0-6-amd64 (#1 SMP Debian 4.9.88-1+deb9u1 (2018-05-07)))' '--serverUuid=190e4be137fd2dcf76610d1a3c6e93199925f922' '--userId=xxxxxx@xxxx' '--url=https://crashreport.plexapp.com'
Jun 23, 2018 03:44:49.782 [0x7f82c73ff700] DEBUG - Jobs: Starting child process with pid 17250
Jun 23, 2018 03:44:49.843 [0x7f82b83ff700] DEBUG - Jobs: '/usr/lib/plexmediaserver/CrashUploader' exit code for process 17250 is 0 (success)
Anything out of the ordinary with the filesystem setup? Are you using nfs ?
Lots of disk space, it’s a brand new install. The service I had to SIGTERM and then killall as it would not stop. As far as permissions on the mounted library drives they are CIFS / SMB and I can dreate, rename, delete files and folders, so I don’t think it’s a permissions issue.
What’s strange is, it’s behaving like this on the new installation, but it’s had the same media files for years when it was running as a docker under unraid. It didn’t seem to have issues then, so I am doubtful it’s file corruption. I did take a peek at the permissions, but, as I indicated in my earlier post I have rwx on the CIFS mounts. I can also open and play that file within Debian.
Was this at 03:43:02 on 23 June or 03:44:47 ?
We have no Plex Media Server log files pre 03:43:13
So do i understand this that there was a hang that you noticed at 03:43:02 and decided to kill the server but it did not shutdown until you did a killall at 03:44:47?
If there is a hang - try to get all the logs copied out and zipped manually as the shutdown for some reason was not being actioned
The logs would then indicate if there was a deadlock and if that is the case we get into getting diagnostics for a deadlock which would include a /connections list and process dump (kill -segv <pid>) and logs
There is no evidence of any new hang in the all logs tar you have attached
The logs for the server cover Jun 25, 2018 02:48:22 am to 08:58:38 am
No evidence of a hang. Requests were being processed with the last one coming in from wan ip 199.83.xxx.xxx at 08:58:37 am and the preceding request completed at 08:58:35.267. You captured the logs tar at 08:58:39 am. So no hung process and no queue of requests built up.
The last deep analysis completed at 05:04:59 am on Jun 23, 2018
