Apr 2 18:28:25 aaa kernel: [26784954.650653] Plex Media Serv[7136]: segfault at 0 ip 000000000102a01f sp 00007ff6ba7fb790 error 4 in Plex Media Server[944000+da1000]
Repeatedly, most players do not work. Empty crash logs.
I rolled back to 1.18.2.2058 and it’s working again. Had to set the VERSION environment variable so it wouldn’t autoupdate on start. Definitely some incompatibility between docker and the images some time in the last ~month. On a slightly older version, it still crashed but I got a non-empty crashdump.
I have the exact same problem with newest PMS version 1.19.1.2645-ccb6eb67e running the official plexinc/pms-docker image.
Rolled back to a one week old backup and changed image to use PMS version 1.18.9.2578-513b381af and everything works again. If I then try to update to newest PMS image again i will see the Segmentation fault in the docker container logs again and the server will keep restarting and instantly crash over and over again.
This is happening to me as well on both 1.19.1.2645-ccb6eb67e and 1.18.9.2578-513b381af using the cloudb0x image. It appears to happen way more often whenever transcoding is needed. Turning off hardware transcoding didn’t change the incidence level. Transcoding will work but it will be super unreliable (transcoder won’t start or it will but the server crashes at some point so any transcoded streams die).
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/37734580-9223-050d-6a735d9e-6261e615.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/13fcd15d-da04-65f9-4d715eea-0de5d5b6.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/0170da01-81a0-178e-75ed9995-0528830f.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/26d15828-f00f-bb3f-42cc2ce7-1adb55ea.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/306dbf88-f53a-5bac-3db81c30-41b29136.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/270c628b-24c6-a829-4fad0476-182f1375.dmp
Segmentation fault
Starting Plex Media Server.
Error starting framework core for /config/Library/Application Support/Plex Media Server/Plug-ins/Sub-Zero.bundle.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/4c4d8f5f-a7de-e120-5c6f6b08-2ddafbfb.dmp
Segmentation fault
Starting Plex Media Server.
****** PLEX MEDIA SERVER CRASHED, CRASH REPORT WRITTEN: /config/Library/Application Support/Plex Media Server/Crash Reports/1.18.9.2578-513b381af/PLEX MEDIA SERVER/281e3ce9-d179-2a96-6e7b939c-1cb7b89f.dmp
Segmentation fault
Starting Plex Media Server.
I will note that this is happening on these two bare metal server configs:
Processor Intel® Xeon® CPU E3-1275 v6 @ 3.80GHz
Processor Intel® Core™ i7-7700 CPU @ 3.60GHz
But it’s not happening on a GCP cloud instance with this configuration (using the same cloudb0x image):
Processor AMD EPYC 7B12
EDIT: not sure if I should call it so quickly and jinx it but it looks like v1.18.2.2058-e67a4e892 fixes it. I’ve been stressing it by running multiple streams (mix of direct play and transcoded) and it has not crashed.
EDIT 2: it still seems to crash every once in a while but not with segmentation fault.
Did not work for me. Thought it might be memory corruption related so I moved everything to the Xeon mentioned above (ECC memory) and the problem persisted.
Still happening with latest version. It crashes at the same point on every start, but I have not been able to figure out why. I tried various things (Turning off settings, changing the library, etc) with no luck.
Anyone know what format the crash dumps are in? I’ll take a look. I haven’t looked to see what the code looks like at the crash address manually, trying to avoid that.
Also if anyone knows how to attach strace to the Plex processes inside the container that might help.
May 22, 2020 12:21:46.091 [0x7f9808ff9700] DEBUG - Sync: synchronizing 1 sync lists to generators
May 22, 2020 12:21:46.092 [0x7f9808ff9700] DEBUG - BPQ: generator 5875 is in failed state
May 22, 2020 12:21:46.092 [0x7f9808ff9700] DEBUG - Sync: sync list 26342549, sync item 38716212 is in failed state
May 22, 2020 12:21:46.092 [0x7f9808ff9700] DEBUG - BPQ: generator 5876 is in failed state
May 22, 2020 12:21:46.092 [0x7f9808ff9700] DEBUG - Sync: sync list 26342549, sync item 38725675 is in failed state
May 22, 2020 12:21:46.093 [0x7f9808ff9700] DEBUG - Sync: synchronized 1 sync list(s) to 2 generators
May 22, 2020 12:21:46.093 [0x7f9808ff9700] INFO - Sync: downloaded 1 sync list(s) with 2 sync items(s): 0 new, 0 updated, 0 deleted
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - BPQ: garbageCollect - sync - forced - 10
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: starting Sync garbage collection: 10, forced
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: garbage collecting filesystem
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: garbage collected filesystem
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: scanning for items to reprocess in sync list 26342549
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: scanned for items to reprocess in sync list 26342549
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: finished Sync garbage collection: 10
May 22, 2020 12:21:46.093 [0x7f97867fc700] DEBUG - Sync: updating status for sync list 26342549, sync item 38716212, generator 5875
May 22, 2020 12:21:46.094 [0x7f9808ff9700] DEBUG - HTTP requesting GET https://plex.tv/api/v2/features?X-Plex-Token=xxxxxxxxxxxxxxxxxxxx
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - LPE: unknown item 29510.
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - Sync: failed to generate query for path library://e4d024fe-f3f9-4098-af35-5ec7e765810c/directory/%2Flibrary%2Fmetadata%2F29510%2Fchildren in sync item 38716212
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - Sync: empty sync set, unable to update status for sync list 26342549, sync item 38716212
May 22, 2020 12:21:46.095 [0x7f97867fc700] DEBUG - Sync: updating status for sync list 26342549, sync item 38725675, generator 5876
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - LPE: unknown item 29515.
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - Sync: failed to generate query for path library://e4d024fe-f3f9-4098-af35-5ec7e765810c/item/%2Flibrary%2Fmetadata%2F29515 in sync item 38725675
May 22, 2020 12:21:46.095 [0x7f97867fc700] ERROR - Sync: empty sync set, unable to update status for sync list 26342549, sync item 38725675
May 22, 2020 12:21:46.184 [0x7f97ed7fa700] DEBUG - HTTP 200 response from GET https://plex.tv/media/providers?X-Plex-Token=xxxxxxxxxxxxxxxxxxxx
May 22, 2020 12:21:46.185 [0x7f97ed7fa700] DEBUG - HTTP requesting GET https://plex.tv/media/providers?X-Plex-Token=xxxxxxxxxxxxxxxxxxxx
May 22, 2020 12:21:46.260 [0x7f97d2ffd700] DEBUG - HTTP 200 response from GET https://plex.tv/api/v2/features?X-Plex-Token=xxxxxxxxxxxxxxxxxxxx
May 22, 2020 12:21:46.262 [0x7f980affd700] DEBUG - DVR: Starting up.
May 22, 2020 12:21:46.353 [0x7f980affd700] DEBUG - Job running: ‘/usr/lib/plexmediaserver/CrashUploader’
Has anyone had issues with the live TV portion of plex causing their computer to crash? I don’t even need to have a stream going so I don’t know if some sort of guide update or something is causing the crash. I know it’s the Live TV portion that’s causing this b/c if I delete the “Live TV & DVR” section the computer runs fine without freezing. Another really weird thing is if I format my computer and install plex including the live TV portion it’ll run for weeks on end without crashing. The 1st time I restart the computer post installing plex the computer crashes every time within 5 minutes of booting up. I’ve attached my log and database filesPlex Media Server Databases_2020-06-03_14-01-16.zip (1.2 MB) Plex Media Server Logs_2020-06-03_14-01-13.zip (3.5 MB)
And is Crash Reporting enabled in Plex Media Server settings (advanced) ?
Yes
Please could you also include server logs zip with debug logging not disabled for period up to time of crash
Debug logging is enabled. I have downgraded the user to the version recommended in this thread (1.18.2) and the crashes appear to have stopped as others have reported.
I’m attaching the logs here. I have pulled the crash trace, but am unable to upload a .dmp file – the reporter appears to have uploaded the dump successfully, however. The latest two restarts are on on 1.18.2; however, the logs for 1.19.5 are in PMS.{2,3,4,5}.log. These logs are all relatively small and it appears these 4 restarts all occured within 10 minutes of each other.
This specific crash is a 1.19.5 regression and not related to any previous posts in this forum thread. A fix has been produced
It relates to having access to a shared server that has a domain specified as Server Access URL. The crash I looked at for 10th July at 17:30 was due to this 1.19.5 bug when trying to communicate with shared server Perplexed
I do have a build with the fix that is being tried by one user - but there is no easy way for me to give it to you to use in a Docker environment
Remove that shared server and this would get round the problem - unless you have other shared servers with domain names
Thanks for the follow-up and quick response. Glad to know it’s a simple regression.
No worries about the test build since downgrading has fixed this in the meantime. Otherwise, I’ll get out of this thread since the issues are different. However, I’m guessing that I should just be able to keep an eye on the PMS changelogs for the fix?
