Server Version#: 1.14.1.5488
Player Version#: n/a
OS is Linux Mint 19
A little while ago, my media server hung. Keyboard was not responding and no display but I was able to SSH into it. In the syslog, I found this:
Jan 20 17:48:34 server-pc kernel: [128822.203002] [drm] GPU HANG: ecode 9:2:0xcab6fff5, in Plex Transcoder [4973], reason: Hang on vcs0, action: reset
To my untrained eyes, it looks like the Plex transcoder might have brought my system down?
Sadly I can’t help.
You have DEBUG logging disabled.
DEBUG logging would have allowed me to see what it was doing at the time of failure.
If if happens again, DEBUG logs will show us.
Such a hang is often the kernel driver or hardware itself in these cases
@ChuckPa No worries. I think this is only the 2nd, MAYBE 3rd time this has happened in the span of a year or so. If I enable debug logging, will my drive eventually fill up or does it overwrite the log files after 24 hours or after they get so big?
I think I read about a way to push all the logs to the syslog but I’d rather not do that.
Log files are always “Rollover”"
It will retain up to the number specified.
It will then FIFO action through them.
If you push PMS logs to SYSLOG, I won’t be able to support you. Engineering made that decision for us.
GPU hang again. Debug mode was enabled ever since my last note about this. Logs attached.
Feb 9 10:03:17 server-pc kernel: [ 9106.740016] [drm] GPU HANG: ecode 9:2:0xa8dfbffd, in Plex Transcoder [26155], reason: Hang on vcs0, action: reset
Feb 9 10:03:17 server-pc kernel: [ 9106.740072] i915 0000:00:02.0: Resetting vcs0 after gpu hang
Feb 9 10:03:25 server-pc kernel: [ 9114.732118] i915 0000:00:02.0: Resetting vcs0 after gpu hang
Feb 9 10:03:33 server-pc kernel: [ 9122.731930] i915 0000:00:02.0: Resetting vcs0 after gpu hang
Feb 9 10:03:41 server-pc kernel: [ 9130.731798] i915 0000:00:02.0: Resetting vcs0 after gpu hang
Feb 9 10:03:49 server-pc kernel: [ 9138.731539] i915 0000:00:02.0: Resetting vcs0 after gpu hang
Unfortunately, by the time you pulled the logs, the oldest entry is:
Feb 09, 2019 15:02:02.863 [0x7fa951120700] INFO - Plex Media Server v1.14.1.5488-cc260c476 - ubuntu PC x86_64 - build: linux-ubuntu-x86_64 ubuntu - GMT -05:00
Feb 09, 2019 15:02:02.863 [0x7fa951120700] INFO - Linux version: 4.15.0-45-generic (#48-Ubuntu SMP Tue Jan 29 16:28:13 UTC 2019), language: en-US
Feb 09, 2019 15:02:02.863 [0x7fa951120700] INFO - Processor Intel(R) Core(TM) i7-7700K CPU @ 4.20GHz
Feb 09, 2019 15:02:02.863 [0x7fa951120700] INFO - /usr/lib/plexmediaserver/Plex Media Server
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160326_143639.jpg"
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160327_161521.jpg"
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160327_180356.jpg"
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160327_180419.jpg"
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160327_180425.jpg"
Feb 09, 2019 15:02:02.863 [0x7fa91eff9700] DEBUG - Adding file for scanner: "/mnt/md0/Pictures/2016/March/20160327_180431.jpg"
All those photos caused it to go through all 5 rollover logs in seconds.
@ChuckPa Well heck… ok, I’ll try and catch it earlier next time.
Or don’t add so many photos all at once. Each individual photo was listed. Look for yourself
@ChuckPa Actually, i wasn’t adding any photos. My picture library is fairly static. Once every few months, I’ll offload my digital camera to my server and I haven’t done that in a long time so I’m curious why the logs indicated i was adding photos which I wasn’t.
I don’t know.
Should it be needed, we can add LogNumFiles=“15” to your Preference.xml file temporarily (PMS Stopped) to increase log retention to 15 files.
@ChuckPa Did I catch it this time?
[91934.168060] [drm] GPU HANG: ecode 9:2:0xa8dfbffd, in Plex Transcoder [8914], reason: Hang on vcs0, action: reset
[91934.168073] i915 0000:00:02.0: Resetting vcs0 after gpu hang
[91940.155417] i915 0000:00:02.0: Resetting vcs0 after gpu hang
[91948.151251] i915 0000:00:02.0: Resetting vcs0 after gpu hang
[91956.119129] i915 0000:00:02.0: Resetting vcs0 after gpu hang
[91964.150963] i915 0000:00:02.0: Resetting vcs0 after gpu hang
yes. and that is a loud & clear Kernel bug
The i915 driver in the kernel is hanging.
Because that is compiled into the kernel, there is nothing I can do…
Submit that to the distro for fixing.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
