My problems began, or were first noticed, when the Preclearing of a new drive failed during the post read of the first run. I had successfully run 2 other preclear operations the week before. All 3 Preclear operations were run on Seagate 10TB drives. When I got the Post read failed error, I started to have issues accessing the unRAID web GUI. I searched the Forums and came across posts suggesting a memory test be run. I downloaded the current free version of memtesst86 and ran 4 passes of all test with 0 errors detected. I do have ECC memory so ran the tests with ECC enabled. Mover would not move anything prior to the memtest and upon reboot following the memtest the Docker service fails to start and the VMs that were installed are no longer there.

 

Currently running version 6.10.3.

 

Issues in the order as I remember them

 

1. Preclear failed during post read

2. Unable to access unRAID web GUI

3. Reboot and Mover will not move anything

4. Boot into Memtest, tests run without errors

5. Reboot and Docker service fails to start and VM's are gone.

 

Attached are the Diagnostics and a syslog from before and after the memtest/reboot. I am kind of stuck at this point. Thanks to all who take the time to help.

syslog after reboot.txt syslog before reboot.txt blackhole-diagnostics-20221114-0714.zip blackhole-diagnostics-20221114-1503.zip

Cache pool is completely full, you must move/delete some data, then better to recreate the docker image.

Can you tell why the mover isn't working?  Or why the web GUI became unresponsive? 

Edited November 15, 20223 yr by elco1965
Added second question

Which shares are you trying to move?

I somehow had set the TV share to use cache. I have changed this to no and I'm currently moving some files using the file browser.

 

I'm still confused as to why the VMs have disappeared, or why the web GUI stopped responding. Does a full cache cause this?

 

Thanks for your reply.

Assuming the libvirt image is stored on cache it will cause VM issues, GUI should still respond.

9 hours ago, JorgeB said:

Cache pool is completely full, you must move/delete some data, then better to recreate the docker image.

 

 

After cleaning up the cache pool and a reboot, the VMs and Docker Containers have returned. Thank you for the help.

After moving some files the cache started to fill up again. Turns out the Unifi-Control Docker was filling the cache with a logs in the cache/appdata//unifi-controller/data/db/journal. The last time I look at the calculation of this folder it was 1.28TB. I have no idea why. I deleted and recreated the Docker img, reinstalled all docker containers, and so far things seem to be back in order. 

 

JorgeB, Thanks again for pointing me in the right direction.