HELP!!! I have been running Unraid for over a year now, I am still green. I have not been on the server in several days. Tonight I realised it was acting slow and cache is full so I tried a manual move and it did not move so I tried a reboot because now the docker service now will not start. I did manage to get a full set of logs from Fix Common Problems. Please any help would be greatly appreciated. logs are attached.

unraid-diagnostics-20220315-2151.zip

@Stephen67

No obvious reason your post belonged in Fix Common Problems thread so I have split it into its own topic here.

 

Your system share has files on the array instead of all on cache.

 

Your log space is completely full, not clear why since your syslogs don't seem that large. What do you get from the command line with this?

du -h -d 1 /var/log

 

After you get those results and post them, reboot to clear your logspace then post new diagnostics.

As you noted, your cache is completely full.

 

You have a cache-prefer share, anonymized as 'D--a'. What is the purpose of that share?

 

Go to User Shares, click the Compute All button at the bottom, wait for the results. If you don't get complete results after several minutes refresh the page. Then post a screenshot.

 

I apologize trurl, I was getting late last night, I was panicking. When I  ran common fix, I thought there was a url that said post to Fix Common Problems thread. Again a nubie mistake in panic mode. 

anonymized as 'D--a'. What is the purpose of that share? I have no clue, I think I was following a spaceinvader video when I did it.

 

I think the cache drive might be dead. Last night I told it to move immediately and nothing happed. 

 

A few minutes ago, I was watching on how to replace a cache drive.  One step was to do a Compute All, which I did and it froze. I did a refresh and 3 shares updated... I waited 15 minutes and nothing. Should I change all the shares cache pool to yes? including virtual system and windows, or leave them. 

 

The out of cache does that explain why docker service fails and one VM will not start? I am in a panic.

 

Thanks and I am trying the compute all again, now.

 

Thank you I really appreciate it and once it's semi working again if you have a PayPal link I will buy you several coffee's or beers or whatever you prefer! 

du -h -d 1 /var/log

 

I guess I wasn't patient enough for the "Compute All"

Delete the Mover Tuning plugin

Go to Scheduler Settings and enable Mover Logging

Run Mover, wait for it to complete or at least run for a while

Then post new diagnostics

I am working the new diagnostics it's taking forever. I don't know if this is weird or not. But I stopped my windows VM in hopes that my home assistant VM would work. It got further and look a lot more normal on the boot than before, BUT this is the weird part I told the VM not to auto start, I rebooted the system and when it came back up the docker service was started and almost all the dockers were running. I had to re-boot and the docker service is dead again. I also ran smart on the SSD in question.

 

If I were to do a re-install would the data still be on the drive? I am almost at the point of blowing everything away. Because I tried mounting a remote NAS share to get my Plex off and that doesnt even work. I'm panicking and at a total loss. I am re-thinking unRaid. I had a reservation when I installed it after I found out there is no real back up for your data other than a second server or rsync.

unraid-smart-20220316-2013.zip

I am not following on "Mover Tuning plugin". Is that "Folder Caching".. or "Dynamix SSD TRIM" ?

Figured it out, CA-mover tuner.

 

Thanks

It appears to be moving the cache, I can see the files moving when I look at the logs button.

Here's the diags after 3 minutes or so.unraid-diagnostics-20220316-2227.zip

Mar 16 21:13:01 unRAID root: Fix Common Problems: Warning: Share Backups set to not use the cache, but files / folders exist on the cache drive
Mar 16 21:13:07 unRAID root: Fix Common Problems: Warning: Deprecated plugin Unraid-Nvidia.plg

Backups is set to cache-no. Mover ignores cache-no shares. You must set it to cache-yes to get files moved from it.  And you should remove the deprecated plugin

 

The very last thing in your syslog is

Mar 16 22:21:15 unRAID root: mover: started

but there are no messages from mover logging after that even though your diagnostics have a timestamp of 22:27, several minutes later.

 

11 hours ago, trurl said:

Go to Scheduler Settings and enable Mover Logging

Did you?

 

Mar 16 21:13:07 unRAID root: Fix Common Problems: Error: Unable to write to cache

If cache really is unwriteable, mover can't move anything off cache. You will have to fix cache problem before you can make any progress. I didn't notice anything about cache corruption in these latest diagnostics, except for a corrupt docker.img which may be on cache but since your system share has files on the array also can't say for sure. system share should be cache-prefer or only instead of cache-yes as you have it. Might be best to set it to cache-only for now so mover will ignore it until we make room on cache.

 

Disable Docker in Settings and delete the corrupt docker.img. Also disable VM Manager in Settings. Leave them both disabled until all your other problems are resolved.

 

Mar 16 21:13:47 unRAID unassigned.devices: PHP Warning: syntax error, unexpected END_OF_LINE in Unknown on line 29  in /usr/local/emhttp/plugins/unassigned.devices/include/lib.php on line 2327

Not sure what the repeated error about unassigned devices is but since it is UI (PHP) related you might try clearing browser cache.

The cache did move and all my VM's now work as well as dockers. So removing CA-mover tuner seemed to be the problem.  The unassigned devices error was because I attached a 4Tb drive to try copying some files and I think it has issues. I totally forgot the drive wasn't marked. 

The latest diags is attached. Let me know what you think. Like I said I could see the being moved and some of the files were from months ago. So the cache app was bad or corrupt? 

unraid-diagnostics-20220317-2210.zip

23 hours ago, trurl said:

your system share has files on the array ... system share should be cache-prefer or only instead of cache-yes as you have it.

Dockers/VMs will perform better on fast pool and won't keep array disks spunup since these files are always open. You might consider changing that setting for your domains share also.

 

You will have to disable Docker and VM Manager since nothing can move open files, set system and domains to cache-prefer, then run mover to get those moved to cache.

 

Haven't looked much further into your diagnostics yet, will check back later when I have time.