mirror of
https://github.com/jellyfin/jellyfin.git
synced 2026-05-04 18:09:12 +03:00
docker wipe out config folder #600
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @icf20 on GitHub (Apr 20, 2019).
1h ago there was a container update pushed for some odd reason this update deleted everything on the /config folder and the users+ library+plugins had to be reconfigured
@sparky8251 commented on GitHub (Apr 20, 2019):
We recently pushed a 10.3.0 container in prep for the official 10.3.0 release. Guess you have autoupdate set in your docker management system.
How did you setup the docker container originally?
What was the complete command used/what ports, volumes, and options were set in the GUI?
No data should be being deleted.
@evertonstz commented on GitHub (Apr 20, 2019):
I use ouroboros to pull new images, I didn't lost plugins and the data at /config seems to be fine, but after the update jellyfin refuses to work, seems to be related with ffprobe:
ERROR(S): Option 'ffprobe' is unknown.full log: https://pastebin.com/Vt4wNurt
EDIT: nevermind, just saw this was fixed already, I didn't had the latest image.
@icf20 commented on GitHub (Apr 20, 2019):
yep i have watchtower set to update
i dont follow you, what do you mean ? the folder setup?
@sparky8251 commented on GitHub (Apr 20, 2019):
Kind of? I'm asking for info on if you used volume or if you used the host filesystem and how you had it configured.
I'm kind of curious if you had volumes set so the update created new volumes making it appear that the config folder was wiped out.
@icf20 commented on GitHub (Apr 20, 2019):
is was all like the doc guide nothing fancy
disk1_cache_/blabla/cache/jellyfin for /cache
zfspool/blabla/config/jellyfin for /config
zfspool/blabla/multimedia for /media
@sparky8251 commented on GitHub (Apr 20, 2019):
Looks right to me. Not too shocking since docker is easy to setup right.
Hmm... I wonder if @nvllsvm pushed a defective container that did something before pushing a corrected one?
We have had no other reports of a deletion of config data inside docker or otherwise so its hard to say what happened here.
I just hope you had backups so this is just a minor inconvenience for you :(
@evertonstz commented on GitHub (Apr 20, 2019):
@sparky8251 I'm pretty sure he pushed at least two containers for the new version, the first one I got automatically with ouroboros and the second one I pulled manually right after I reported my problems here. The fist push he did wasn't even starting here, but my config files were not touched.
@andrewrabert commented on GitHub (Apr 24, 2019):
@evertonstz Please share the output of this command. Do you know which image ID was the culprit?
@evertonstz commented on GitHub (Apr 24, 2019):
@nvllsvm I sadly had to remake my server two days ago since I had a hardware failure, but since Ouroboros do notify me on Gotify I have the last notification it sent me about Jellyfin:
jellyfin updated from 11e6303e05 to fa07cdd18e@sparky8251 commented on GitHub (Jun 25, 2019):
This appears resolved to me. Closing.