Unraid version 7.2.3

I have been running an unraid system with plex, *arr stack, and sabnzbd for a few months now with no errors. Yesterday the docker image filled, and the whole system refused to give me any webgui access. I hard restart, read about increasing the docker image size. I increased it to around 150gb. It was fine throughout the day, and then It filled up last night, and it crashed this morning. This morning I hard reboot, I increased the docker image size to around 250gb. After about an hour of poking checking different menus and settings, it crashed out again and I could not access the webgui again.

So I have done a fair amount of research into the docker image issue, which I am hoping is leading to the loss of webgui access as well. The problem here is most of these tutorials/solutions require hitting the container size button in docker settings in order to see the culprit. I have tried that repeatedly and after restarts and it will not work for me. I have tried a few scripts to determine docker sizes, but those scripts also failed to show anything at all. My gut is telling me this is some sort of logging or something going into the docker file based on what I read, but I can't seem to find out which docker program.

For now it is turned off. I can turn it on for an indeterminate amount of time and interact if there are logs I need to acquire, it just eventually crashes out on me.

Thanks in advance for any help that can be offered.

Enable the syslog server and post that if it crashes again, together with fresh diagnostics after a reboot.

docker.img should not grow. The usual reason is an application writing to a path that is not mapped.

Take each of your containers one at a time and see if it causes docker.img to grow when run by itself.

So yesterday I restarted the system, and tried to cycle through different docker images to see if the used docker vdisk percent under system would change, but it seemed stuck. The total wouldn't change i I changed it, or the amount being used. It seemed almost frozen, but just the docker vdisk, other things seemed to be updating.

So this morning The docker vdisk began to fill again. I only had plex and tautulli running in docker last night, everything else was turned off. At around 3:22am EST I got the first push notification that my docker image was filling, and it proceeded to fill up until it it 100% at 5:38am. I notice looking back at the last few days, this always seems to happen early in the morning. It did not crash, it is currently running with just plex and tautulli, with a full docker vdisk.

I have attached the diagnostics.

crykflix-diagnostics-20260127-0726.zip

31 minutes ago, cryk said:

plex and tautulli

Post docker run for both

This is the docker run for plex. 

docker run
  -d
  --name='Plex-Media-Server'
  --net='host'
  --pids-limit 2048
  -e TZ="America/New_York"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="Crykflix"
  -e HOST_CONTAINERNAME="Plex-Media-Server"
  -e 'PLEX_CLAIM'='Insert Token from https://plex.tv/claim'
  -e 'PLEX_UID'='99'
  -e 'PLEX_GID'='100'
  -e 'VERSION'='latest'
  -e 'NVIDIA_VISIBLE_DEVICES'='all'
  -e 'NVIDIA_DRIVER_CAPABILITIES'='all'
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.webui='https://[IP]:[PORT:32400]/web'
  -l net.unraid.docker.icon='https://raw.githubusercontent.com/plexinc/pms-docker/master/img/plex-server.png'
  -v '/tmp/plex':'/transcode':'rw'
  -v '/mnt/user/data/':'/data':'rw'
  -v '/mnt/user/appdata/Plex-Media-Server':'/config':'rw'
  --runtime=nvidia 'plexinc/pms-docker'

c663dfab5afba5d2097d16a64fabdc288ae9136318fe9b810cbb9680016d0d74
docker: Error response from daemon: endpoint with name Plex-Media-Server already exists in network host.

The command failed.

And here is the tautulli docker run,
docker run
  -d
  --name='tautulli'
  --net='bridge'
  --pids-limit 2048
  -e TZ="America/New_York"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="Crykflix"
  -e HOST_CONTAINERNAME="tautulli"
  -e 'PUID'='99'
  -e 'PGID'='100'
  -e 'TZ'='UTC'
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.webui='http://[IP]:[PORT:8181]/'
  -l net.unraid.docker.icon='https://raw.githubusercontent.com/Tautulli/Tautulli-Unraid-Template/master/img/logo-circle.png'
  -p '8181:8181/tcp'
  -v '/mnt/user/appdata/tautulli':'/config':'rw' 'tautulli/tautulli'

3674fa01805c38f95b1e3c67b1508958dd289590b530793555e8248ced040b40

The command finished successfully!

Another note, now that I have stopped plex, it seems to refuse to start back up, and gives me an error 403. So maybe it is plex related? Everything else starts and stops fine.

Not sure why you got that error from plex docker run.

Click on its icon, remove it, and reinstall using Previous Apps on the Apps page.

Do you use the DVR capability in Plex?

No, I do not use the DVR capability.

In Plex, what is your setting for "Transcoder temporary directory"

/transcode

That should be OK then. Don't see how plex could be writing into docker.img

It only writes to /transcode, a DVR path which you don't have, and its database is in appdata.

I don't use tautulli

4 hours ago, cryk said:

At around 3:22am EST I got the first push notification that my docker image was filling

What do you have in Plex Settings for Scheduled Tasks?

A database backup from 2am to 5am to

/config/Library/Application Support/Plex Media Server/Plug-in Support/Databases

Some searching also tells me the optimize database checkmark could be causing issues? Unsure.

Edited January 27Jan 27 by cryk

Update

So I have tried to find the culprit for the docker size increase/filling but alas, nothing. The container size button works now, but it only shows a total size of 6.13 gb.

Also on the unraid dashboard, the vdisk size never updates. I have changed the size of the docker image multiple times, but it still says the same numbers it has for days. It won't change the usage amount either, whether docker is running or not. When I hover docker vdisk under system, vdisk usage, it always says out of 250gb (which I have changed since setting it that high) and it always says 4% used. I am not sure if it has stopped me from changing the vdisk size, stopped updating, or if it's just giving me a webgui bug.

I noticed the docking filling up happens around the same time as my plex database backups happen early in the morning, but for the life of me I cannot find where it is saving these. I have /data mapped which I assumed would be enough, but I don't think it is. However I cannot figure out where these files are specifically to delete them, or what I need to change.

21 minutes ago, cryk said:

I cannot find where it is saving these. I have /data mapped

You have /config mapped which is where the database is, your plex appdata

Just now, trurl said:

You have /config mapped which is where the database is, your plex appdata

So should not be in docker.img

23 minutes ago, cryk said:

The container size button works now, but it only shows a total size of 6.13 gb.

Does that change after the backup?

My container size did not change when doing database optimizations in plex. I do not know how to force a database download to the backup, so I think I need to wait for the automation.

as for the system itself, it seems to be working, except the vdisk percent has not updated in two days. I also haven't been running most things. If it's plex related or some other backup related it's happening early in the morning, but not every morning.

I suppose at this point I should run it as normal and wait for the system to crash, but I need system logs I'd imagine.

Is it safe to run mirror to flash for an entire week, or will that destroy the drive? I feel like at this point I just need to have it crash again, and I don't know how to force a crash other than letting it run as normal. I expect it would happen within a few days.

Thanks for all the help so far by the way, I appreciate it.

You can check usage of docker.img from command line:

df -h /dev/loop2

So this is weird, because I still have the notifications in my pushover of my docker being full a few nights ago. I have also set my docker size at 50gb currently, this 250gb number is the number I had set the second time it filled up, around the same time the vdisk updates got stuck on the system dashboard.
Now it says only 7.5gb, but also the changes I make apparently aren't saving.

51 minutes ago, cryk said:

I have also set my docker size at 50gb currently, this 250gb number

You have to delete and recreate

https://docs.unraid.net/unraid-os/troubleshooting/common-issues/docker-troubleshooting/

Docker image has been recreated. Seems to be a normal size at the moment. Not sure where to go from here, unless maybe all I needed to do was recreate.