Docker containers not starting.

[shadowlord]

I am getting an Execution Error / Bad parameter message when trying to start most (but not all) of my Docker containers.  I have tried deleting my docker image directory and rebooting but still this message persists.  One very simple Docker container still starts (Wordle), so I know it's not problem with the Docker service itself.

 

 

server03-diagnostics-20221218-1157.zip

[Squid]

19 minutes ago, shadowlord said:

Execution Error

The docker run command will tell you what's wrong.  Post it back after following the linked directions

[shadowlord]

Here is the output of the docker run command:

 

docker run
  -d
  --name='Plex-Media-Server'
  --net='host'
  -e TZ="America/Los_Angeles"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="server03"
  -e HOST_CONTAINERNAME="Plex-Media-Server"
  -e 'PLEX_CLAIM'='claim-8FxwszqEtTrpLjRtrC7T'
  -e 'PLEX_UID'='99'
  -e 'PLEX_GID'='100'
  -e 'VERSION'='latest'
  -e 'NVIDIA_DRIVER_CAPABILITIES'='all'
  -e 'NVIDIA_VISIBLE_DEVICES'='GPU-62db3608-fe97-769e-a59d-cffd6cae6b9c'
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.webui='http://[IP]:[PORT:32400]/web'
  -l net.unraid.docker.icon='https://raw.githubusercontent.com/plexinc/pms-docker/master/img/plex-server.png'
  -v '/mnt/user/media/plex/transcode/':'/transcode':'rw'
  -v '/mnt/user/media/plex/':'/data':'rw'
  -v '/mnt/user/media':'/mediastore':'rw'
  -v '/mnt/user/photos':'/photos':'rw'
  -v '/mnt/user/nextcloud':'/nextcloud':'rw'
  -v '/mnt/user/appdata/plex/config/':'/config':'rw'
  --runtime=nvidia 'plexinc/pms-docker'
11c7b2f39062478df23ec8644d3dd3bee6fede111f21913a718590eb37b80006
docker: Error response from daemon: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: exec: "/init": stat /init: no such file or directory: unknown.

The command failed.

 

[Squid]

Did you say you recreated the docker image already?  That would be my go to for the solution...

[shadowlord]

28 minutes ago, Squid said:

Did you say you recreated the docker image already?  That would be my go to for the solution...

 

Yes, that is correct.  I am no longer using a single docker.img file, so I deleted the docker folder (using the unraid web GUI) and re-installed the images.

Edited December 18, 2022 by shadowlord

[Squid]

DOCKER_IMAGE_FILE="/mnt/user/docker/"

 

# Share exists on cache, disk1
shareUseCache="yes"

 

When using a folder, you definitely do not want mover to move it around.  This folder should be 100% a cache only folder.   Except that right now you'll have to recreate it again and set it as cache only and then reinstall all the apps

[shadowlord]

DOCKER_IMAGE_FILE="/mnt/user/docker/"

 

# Share exists on cache, disk1shareUseCache="yes"

 
When using a folder, you definitely do not want mover to move it around.  This folder should be 100% a cache only folder.   Except that right now you'll have to recreate it again and set it as cache only and then reinstall all the apps

I understand what you're saying. But I'm not exactly sure how to carry out your instructions.

The added complication is that one of my cache SSDs may be dying (increasing number of SMART reallocated sectors) and I'm offsite right now.

I have remote access to the unraid server via wireguard.

Sent from my Pixel 7 using Tapatalk

[trurl]

3 hours ago, Squid said:

recreate it again and set it as cache only and then reinstall all the apps

25 minutes ago, shadowlord said:

not exactly sure how to carry out your instructions.

https://wiki.unraid.net/Manual/Docker_Management#Re-Create_the_Docker_image_file

https://wiki.unraid.net/Manual/Docker_Management#Re-Installing_Docker_Applications

 

26 minutes ago, shadowlord said:

one of my cache SSDs may be dying

You have 2 different sized disks in your cache pool. Which btrfs raid mode are you using?

 

Default raid1 would give a mirror with capacity equal to the smallest. To get combined total capacity with different sized disks you should be using single mode.

 

 

 

 

[trurl]

35 minutes ago, shadowlord said:

I'm offsite right now.

Best approach would be to disable Docker and VM Manager in Settings until you get your pool fixed.

[shadowlord]

You have 2 different sized disks in your cache pool. Which btrfs raid mode are you using?
 
Default raid1 would give a mirror with capacity equal to the smallest. To get combined total capacity with different sized disks you should be using single mode.
 
 
 
 

I am using btrfs raid 1 mode. The unraid GUI shows a total capacity equal to the sum of the 2 SSDs (120gb + 250gb) and in the past, the cache been filled well beyond 250gb.

Anyway, I would like to get this running, even if I don't use a cache drive. Before I had my cache pool, I just ran the Docker images from my array. Is it possible to do this again?

[trurl]

31 minutes ago, shadowlord said:

btrfs raid 1 mode

with the drives you currently have assigned to that pool, you get a mirror of only 120G

https://carfax.org.uk/btrfs-usage/?c=2&slo=1&shi=1&p=0&dg=1&d=240&d=120

 

34 minutes ago, shadowlord said:

Before I had my cache pool, I just ran the Docker images from my array. Is it possible to do this again?

Yes, but Dockers/VMs will have performance impacted by slower array, especially any writes to appdata, domains, system shares if these are on the array. Also, array disks can't spin down since these files are always open.

[trurl]

16 minutes ago, trurl said:

appdata, domains, system shares

all of these shares currently have files on cache and on the array. You also have several cache:no shares with files on cache. Mover ignores cache:no shares.

 

Nothing can move open files. Disable Docker and VM Manager in Settings. Set all these shares to cache:yes, run Mover so it can try to move them to the array. Mover won't replace files so if any are already on the array you will have to decide which to keep. You can use Dynamix File Manager plugin to help with this.

 

After everything is off cache, you can reduce it to the one good disk, might be a good idea to reformat it, then you can set appdata, domains, system shares back to cache:prefer and move them back to cache. Might be best to not cache any other shares until you get things working well and see how much capacity is left for caching.

 

Then you can follow the links I gave above to get dockers recreated.

[shadowlord]

Thank you.  

 

I followed the steps in your last post, and I have been able to get my Docker images up and running again. 

 

Marking your post as the solution.