I suspect that this is due to a bug that has been identified in the Unraid shutdown sequence where the location containing the required binaries gets unmounted too early.
If so this is fixed for the 6.12.4 release which I think is probably getting close now. I would check when that is available if the problem still occurs or there is something else going on.
This suggests you have their working sets mapped to a location that is not an actual physical one so ends up being in RAM which is why they get lost on reboot.
There may be a permissions issue, or it may be the one that docker is currently using.
I would try the Dynamix File Manager plugin in preference to Krusader as that would not be affected by a permissions issue.
You are looking for a docker container that is growing in size as it runs. Probably the commonest is Plex transcoding to a location that is internal to its container, whereas ideally that location should be mapped to an external location on the Unraid host.
The Container Size button on the Docker tab can help with this.
You have to decide which one is the most current (probably the one on the cache, and then you can delete the other one.
Note that the high utilization of docker.img is unrelated to this issue - it tends to mean that the way you have plex configured it is writing to a location internal to the docker.img file rather than to an external mapped location.
You can use the process documented here in the online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page to sort out your drives.
Not sure to be honest!
One thing to remember is that UD devices cannot participate in user shares whereas pools can. Whether that will affect your workflow I am not sure.
it is possible the mover restriction might disappear in the 6.13 release, but until we actually get visibility of that release that is just a guess.
You might want to try enabling the syslog server so we can get a syslog that survives a reboot.
Posting your current diagnostics might allow us to see if you got an unclean shutdown or not. If you did not then this would suggest something actively shutdown your server.
The network interface seems to be continually going up and down. Have you tried rebooting the router.
i notice that you seem to have Jumbo frames enabled. In my experience these tend to cause more problems than they are worth and provide negligible performance gain.
Single drive pools can be any supported file system. Multi-drive pools have to be ZFS or btrfs. Btrfs provides maximum flexibility in expanding pools, particularily with drives of mixed sizes, whereas ZFS normally provides maximum performance.
1). Using a dtive for cache is about improving writing (not reading) speed but is completely optional.
2). The OS is not ‘installed’ in the traditional sense. It is always loaded off the flash drive at boot time to run from RAM.
Have you tried the steps for troubleshooting Unclean shutdown that are covered here in the online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page.
You should not assume that anyone from the Unraid team sees every post on this forum as it is primarily used for community support. To get a formal answer I would expect you need to email Unraid support.