Hey guys, weird issue right here.

I had set up my dockers and all was well for like 200 days.

My unraid updated and asked me to restart so I did. After restarting I noticed my transmission queue was empty.. no worries I thought, it's just transmission.

But sonarr, couchpotato, plex all of them had been reset...

 

The settings look fine to me, I did notice the ui has completely renewed however.

 

 

I have my config directories set to /mnt/user/appdata/dockerName, all of the directories are still there but the docker containers refuse to use the old configs.

I checked the permissions and they seem fine too, plex and sonarr are both chmodded with 777 apparantly and those don't work either. The date on them is a bit weird though, it's all from today / yesterday.

 

drwxr-xr-x 1 nobody users      20 Dec 14 18:07 Backups/
drwxr-xr-x 1 nobody users    4096 Dec 14 23:25 MediaCover/
drwxrwxrwx 1 nobody users      32 Dec 14 18:49 PlexMediaServer/
drwxr-xr-x 1 nobody users      34 Dec 14 18:07 UpdateLogs/
-rw-r--r-- 1 nobody users     419 Dec 14 23:48 config.xml
drwxr-xr-x 1 nobody users      36 Dec 14 19:03 couchpotato/
drwxr-xr-x 1 nobody users      90 Dec 15 00:03 headphones/
drwxr-xr-x 1 nobody users      24 Dec 14 18:06 logs/
-rw-r--r-- 1 nobody users  180224 Dec 14 23:03 logs.db
-rw-r--r-- 1 nobody users   32768 Dec 15 00:03 logs.db-shm
-rw-r--r-- 1 nobody users 1054320 Dec 15 00:03 logs.db-wal
drwxr-xr-x 1 nobody users     117 Dec 14 19:03 newtransmission/
-rw-r--r-- 1 nobody users 8498176 Dec 15 00:03 nzbdrone.db
-rw-r--r-- 1 nobody users   32768 Dec 15 00:05 nzbdrone.db-shm
-rw-r--r-- 1 nobody users   37760 Dec 15 00:05 nzbdrone.db-wal
-rw-r--r-- 1 nobody users       3 Dec 14 19:03 nzbdrone.pid
drwxrwxrwx 1 nobody users     139 Dec 14 18:03 sonarr/
drwxr-xr-x 1 nobody users      19 Dec 14 18:07 xdg/

 

Anyone experienced something similar?

What version of Unraid have you upgraded from and to?

 

Sounds suspiciously like the old /mnt/cache/ issue where people specified /mnt/cache/ and then stuff got moved to the array.  Although that has no longer been an issue for a while with the advent of V6.2

 

Might be worth just checking whats set under docker/settings and flick on advanced and post a screenshot.

 

But at the moment I'm at a loss to explain it.

Well.... idk what version I came from but I'm on 6.2.4 now:

 

Here's my plex docker config:

 

The old version looked like this:

 

btw, I did not use /mnt/cache and emptied my cache disk with the mover before attempting the update

Probably a good idea to place transmission and sonarr into separate subdirectories.  Some containers will recursively chmod permission on their appdata folder. 

 

Only thing I can think is try stopping and starting the docker service. Settings => docker then disable then re-enable it.  Whilst you're there grab a screenshot with the advanced tab on..

They should be in separate folders, that is a mistake after trying to get them set up again Will fix this tonight.

I will grab a screenshot of the docker settings as soon as I'm home (in about 8 hours), misunderstood you yesternight

 

 

Post diagnostics.

Meh, these settings look fine too:

 

 

diagnostics attached

cerberus-diagnostics-20161215-1736.zip

First thing I would do would be to set your downloads directory to be cache-only.  Right now its moving to the array nightly which *should* work ok, but its an abnormality that most people don't do.

 

Secondly, I'm curious why you have your docker.img file on disk2, but you have a cache drive.  The performance with the image on an array device is terrible compared to the cache drive, and you don't need to worry about keeping the image on a redundant device since its so easily recreated.

First thing I would do would be to set your downloads directory to be cache-only

 

I'd need a bigger drive for that

 

Secondly, I'm curious why you have your docker.img file on disk2, but you have a cache drive.  The performance with the image on an array device is terrible compared to the cache drive, and you don't need to worry about keeping the image on a redundant device since its so easily recreated.

 

The cache drive was added when I replaced my OS ssd in my main pc

I'm going to be deleting the docker.img and recreate it on my ssd as soon as this problem has an answer or after the weekend (can do the work on a sunday, cause I'll be adding drives anyways)

 

First thing I would do would be to set your downloads directory to be cache-only

 

I'd need a bigger drive for that

 

Then TBH I'd set the d/l share to be use cache:no  and see what happens.  The act of moving might be messing things up (although its not supposed to, but its an easy enough test to make.)

I don't see the correlation between my download share for 1 docker messing up the config files for all dockers tbh.

 

It is however reassuring that the three of you haven't found anything stupid as of yet and that this case is probably something which doesn't happen often. I haven't been able to reproduce the error either.

 

So what I'll do is the following:

[*]I'll uninstall all dockers

[*]I'll clean appdata by throwing all of it away

[*]I'll upgrade the cache drive to a 512gb model and put my download share and my docker image on there.

[*]I'll set up the dockers the way I like and figure out a backup solution for the config directories.