UnRaid 6.8.3 - Docker Fails to Start

[Tiger770]

I have recently been having some issues with my cache disks (where docker lives) dropping the 2nd drive in the BTRFS pool.  A stop / start of the array and a reseat of the disk typically fixes it. 

 

Last night I saw that the entire cache pool was listed as missing.  No big deal.  I stopped the array and gave it a reboot to repoll the disks.  On start up I assigned Cache 1 and Cache 2 to their appropriate spots - but the icons indicated a Blue Square next to them, rather than the green circle.   

 

Ever since starting the array in this config, the Docker service now fails to start.  I'm at a loss as to how to correct this issue.  Any assistance would be greatly appreciated.  Attached is the diagnostic file from my system.

mediaserver-diagnostics-20200726-1411.zip

[JonathanM]

Try unassigning both cache drives, start the array, stop the array, reassign both drives, start the array.

[Tiger770]

2 hours ago, jonathanm said:

Try unassigning both cache drives, start the array, stop the array, reassign both drives, start the array.

Just gave that a shot.  No luck.  Docker service still fails to start.

Edited July 26, 2020 by Tiger770

[JonathanM]

1 hour ago, Tiger770 said:

Just gave that a shot.  No luck.  Docker service still fails to start.

But does the cache pool mount properly, or does it still show unmountable?

[Tiger770]

1 hour ago, jonathanm said:

But does the cache pool mount properly, or does it still show unmountable?

Mounts properly. Docker service still fails to start. 

[JorgeB]

Docker image is corrupt, delete and recreate, also good to check this for better pool monitoring.

[Tiger770]

6 hours ago, johnnie.black said:

Docker image is corrupt, delete and recreate, also good to check this for better pool monitoring.

I came to this conclusion about 14 hours ago as well.  I started by offloading everything from the cache drive to the array shares with mover. When that didn’t work and services still hung I deleted the docker image and the services started normally. All of my containers are super pissed and I am missing configuration data and logs inside apps like plex and pihole even after updating the container paths to point to the user share path for configs rather than the direct path to /mnt/cache/appdata (or system doe some paths).

 

honestly I’m just glad to have it in a partially corrected state. I may end up having to delete and reinstall a few containers as plex and pihole were the only ones that came back and allowed me to edit settings. The rest “appear” but the system doesn’t believe they were setup with the docker templates (they were).  
 

Getting there slowly but surely. 

[trurl]

Your docker templates are on flash. The Previous Apps feature on the Apps page will reinstall any of your dockers exactly as they were.