Server Version#: 1.15.1.710-ece95b3a1 on Ubuntu 18.04.2
PMS has been crashing every over day like clockwork, I have included the Crash Reports below. This has been happening since I upgraded to the new 1.15.1.x betas. Nothing has changed on my VM since then.
Crash Reports:
1.15.1.710-ece95b3a1.zip (92.5 KB)
The Plex Media Server.log that associates with this is gone, but I will post an update when it happens again. Can anything be told from the dmp file?
May I see the ZIP DEBUG logs?
The debug logs for Plex Media Server.(\d\.)?log are only showing timestamps from today, so I’ll have to wait for it to happen again, grab the logs and restart PMS. I’ll post them after that. It’s been pretty consistent, so I bet it will happen tomorrow evening.
Settings - Server - Troubleshooting - Download Logs
-or-
sudo tar cfz /tmp/Logs.tar.gz "/var/lib/plexmediaserver/Library/Application Support/Plex Media server/Logs"
Both give me what I need
Ok, here you go
Your server is really active, isn’t it?
It only goes back in time until 04.53.29 01.Mar.2019
Next you need to restart it, would you grab the logs immediately please?
This guarantees I will see errors before they roll off the end of the buffer.
I do see you are attempting to use the nVidia codecs but I see no actual transcoder activity.
Mar 01, 2019 05:41:55.115 [0x7f2b3b7fe700] DEBUG - Codecs: testing hevc_vaapi (encoder)
Mar 01, 2019 05:41:55.115 [0x7f2b3b7fe700] DEBUG - Codecs: hardware transcoding: testing API vaapi
Mar 01, 2019 05:41:55.116 [0x7f2b3b7fe700] ERROR - [FFMPEG] - libva: va_getDriverName() failed with unknown libva error,driver_name=(null)
Mar 01, 2019 05:41:55.116 [0x7f2b3b7fe700] ERROR - [FFMPEG] - Failed to initialise VAAPI connection: -1 (unknown libva error).
Mar 01, 2019 05:41:55.116 [0x7f2b3b7fe700] DEBUG - Codecs: hardware transcoding: opening hw device failed - probably not supported by this system, error: Input/output error
Mar 01, 2019 05:41:55.117 [0x7f2b3b7fe700] DEBUG - Codecs: hardware transcoding: testing API vaapi
Mar 01, 2019 05:41:55.117 [0x7f2b3b7fe700] ERROR - [FFMPEG] - libva: va_getDriverName() failed with unknown libva error,driver_name=(null)
Mar 01, 2019 05:41:55.118 [0x7f2b3b7fe700] ERROR - [FFMPEG] - Failed to initialise VAAPI connection: -1 (unknown libva error).
Mar 01, 2019 05:41:55.118 [0x7f2b3b7fe700] DEBUG - Codecs: hardware transcoding: opening hw device failed - probably not supported by this system, error: Input/output error
Mar 01, 2019 05:41:55.118 [0x7f2b3b7fe700] DEBUG - Scaled up video bitrate to 10396Kbps based on 1.500000x fudge factor.
Mar 01, 2019 05:41:55.118 [0x7f2b3b7fe700] DEBUG - MDE: Selected protocol hls; container: mpegts
Mar 01, 2019 05:41:55.119 [0x7f2b3b7fe700] DEBUG - MDE: analyzing media item 145960
Mar 01, 2019 05:41:55.119 [0x7f2b3b7fe700] DEBUG - MDE: E8 - War Games: no direct play video profile exists for http/mkv/h264
Mar 01, 2019 05:41:55.119 [0x7f2b3b7fe700] DEBUG - MDE: E8 - War Games: no direct play video profile exists for http/mkv/h264/ac3
Mar 01, 2019 05:41:55.120 [0x7f2b3b7fe700] DEBUG - MDE: Cannot direct stream audio stream due to codec ac3 when profile only allows aac
Has this been giving you any trouble?
- S01E08 - War Games HDTV-1080p.mkv
Mar 01, 2019 05:41:57.042 [0x7f2baf7fe700] DEBUG - TPU: hardware transcoding: zero-copy support not present
Mar 01, 2019 05:41:57.042 [0x7f2baf7fe700] DEBUG - TPU: hardware transcoding: final decoder: , final encoder: nvenc
Mar 01, 2019 05:41:57.043 [0x7f2c0e531700] DEBUG - Auth: authenticated user 944233 as rsacca1
Mar 01, 2019 05:41:57.043 [0x7f2c0e531700] DEBUG - Request: [96.236.39.98:61541 (WAN)] GET /trans
Ha yes, I have a big family I share with.
I’ll check that video file in a bit.
For the logs I’ll get them the next time it crashes. I was tired last night when I copied the crash report and didn’t think to copy over the logs before restarting.
Extra notes: My Ubuntu VM has a Nvidia P2000 being passed thru. I can attempt to disable hardware transcoding after the next crash and see what happens.
That file appears to play just fine. I’ll post an update with logs and dmp file when PMS crash happens again. Thanks for your help so far.
Just as expected a crash again last night, here are the logs. PMS server is unreachable but the systemd service status shows it as running. Also there was nothing in ‘Crash Reports’ for this crash. Weird. I restarted PMS via systemd and everything came back.
logs.zip (5.8 MB)
Thanks. I will look in a bit but catching it unresponsive is important.
Looking at it, it looks like it’s hanging up while transcoding.
Is this your observation as well?
If so, I see the last thing it does is to launch an nvenc nVidia encode. It runs for a while then, working fine then I see your logs end. I don’t actually see a restart.
Am I missing something?
That looks to be what is happening, like you said nothing in the logs is reporting a failure but PMS is not responding. The server doesn’t load in any app I use (web / android) until I restart it.
I am an idiot, I grabbed the logs before I restarted it. I re-read your message and you wanted me to grab them asap right after the restart. Sorry about that. I’ll get those the next time it happens.
For what it’s worth I just grabbed the logs again, there hasn’t been much activity so maybe it has what we want.
logs_new.zip (5.4 MB)
Thanks but no joy. Basic transcoder activity chews the logs up very quickly.
I’ll get it right next time, sorry!
No rush. I’ll be here. 
I got a crash sooner than expected. I restarted PMS, waited for systemd to return, counted to 10 and grabbed the logs.
logs_01.zip (5.0 MB)
Hopefully I did it right this time
Happened last night 
My monitoring isn’t showing high CPU , RAM or Disk usage
Plex Media Server Logs_2019-03-09_08-55-51.zip (5.4 MB)
Thanks for the logs.
I do see your minidump got uploaded to Plex. Hopefully it will show them where it failed. I do not see any hard errors in the log.
Thanks @ChuckPa since the crash happens overnight I’ll just have a cronjob restart Plex every night at midnight for awhile.
Two more crashes in the past day
Plex Media Server Logs_2019-03-11_18-37-49.zip (5.9 MB)
This one I turned off Hardware Acceleration
Plex Media Server Logs_2019-03-12_10-25-22.zip (5.6 MB)
Edit: Maybe the update to PMS today will solve my issue
