Emby server error leading to manual fsck at reboot

[Shadowking 3]

Hi guys,

 

First, let me thank you for all your hard work making Emby a delight to use.
Behind this title I'm not confident it couldn't be a hardware failure, but my host is kinda new (10 months), so...

 

Some information before anything else :

uname -a

Linux NUC 4.19.0-8-amd64 #1 SMP Debian 4.19.98-1 (2020-01-26) x86_64 GNU/Linux

sudo lshw
nuc
    description: Ordinateur de Bureau
    produit: NUC8i3BEH (BOXNUC8i3BEH)
    fabriquant: Intel(R) Client Systems
    version: J72753-304
    numéro de série: [REDACTED]
    bits: 64 bits
    fonctionnalités: smbios-3.2.1 dmi-3.2.1 smp vsyscall32
    configuration: boot=normal chassis=desktop family=Intel NUC sku=BOXNUC8i3BEH uuid=76763E41-E495-9F26-AF64-1C697A024B2E

I'm using Emby through linuxserver image, docker-compose file :

version: "2"
services:

   emby:
      image: linuxserver/emby
      container_name: emby
      network_mode: bridge
      hostname: emby
      environment:
         - PUID=1000
         - PGID=1000
         - TZ=Europe/Brussels
      labels:
         - "com.ouroboros.enable=true"
      volumes:
         - "./config:/config"
         - "/mnt/nas/video/films:/data/films:ro"
         - "/mnt/nas/video/series:/data/series:ro"
         - "/mnt/nas/video/anime:/data/anime:ro"
         - "/mnt/nas/video/films_j:/data/films_j:ro"
         - "/mnt/nas/video/series_j:/data/series_j:ro"
         - "/mnt/nas/video/reportages:/data/reportages:ro"
         - "/mnt/nas/music:/data/music:ro"
         - "./transcode:/transcode"
      ports:
         - 8096:8096
      devices:
         - /dev/dri:/dev/dri
      restart: unless-stopped

What's wrong ?

System is totally a mess, docker.service not responding, most services are dead, except ssh. Only solution is to reboot, autoboot in rescue mode I'm forced to run fsck <partition> to get filesystem repaired, then I can reboot and everything is okay again.

It happenned two days ago, but also a couple of times since late summer (maybe 3-4 times).

 

Trying to troubleshoot

I noticed the entire system was faulty around 12:40 am. Went to check logs, found something on syslog :

Apr  9 07:58:30 NUC kernel: [479707.054121] EmbyServer[1251]: segfault at 4 ip 00007fa0eb3b74ba sp 00007fa00b7facc8 error 4 in libcoreclr.so[7fa0eb2b6000+2ef000]
Apr  9 07:58:30 NUC kernel: [479707.054137] Code: 6c ba 1c 00 66 44 0f 6f 0d 73 ba 1c 00 48 89 7c 24 90 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 48 89 c8 48 8b 30 48 83 e6 fe <8b> 4e 04 83 3e 00 bd 00 00 00 00 79 0a 8b 50 08 0f b7 2e 48 0f af
Apr  9 07:58:30 NUC kernel: [479707.268347] traps: EmbyServer[1310] general protection ip:7f088bd91207 sp:7fffe5e734f0 error:0 in libc-2.27.so[7f088bcfa000+1e7000]

I was not using Emby at this time.

From the logs, I can say that the entire server went into an idle/host state until this moment.

 

What tells Emby logs ?

Last logs are showing 07:21 timestamp, and next log starts logging at... 07:58:31 Seems like related.

You'll find the syslog, and the emby logs attached.

 

Worth saying

* From what I read, people experiencing seqfault are experiencing container reboots soon after they run it.
* Seems SMB related, right now I'm mounting my NAS folders as root in /etc/fstab, Emby is running with my normal admin user on the NUC, and my NAS Synology user is admin on the media folders.

* Systemfile is ext4

* System run on a 240 Gb SAMSUNG 970 EVO Plus SSD

 

Remaining at your disposal for any further information that could help me debug this.

 

Thanks !

embyserver-63722015911.txt

embyserver-63722016817.txt

syslog.1.txt

hardware_detection-63722015916.txt

Edited April 10, 2020 by Shadowking

[mastrmind11 722]

perhaps try the official docker repo and we can go from there?  https://hub.docker.com/r/emby/embyserver

[Luke 42079]

Yea that might be a good thing to try for comparison purposes.

[Shadowking 3]

I switched to the official image, I will report if anything goes wrong.

[Luke 42079]

Great, thanks!