Hello,

 

I think there is something generally wrong with my cache pool.  For a general timeline:

• I started my build with a few drives in the array and a 1 TB and 500 GB SSD in a pool.  I set all my shares to Primary cache and Cache->Array move action (though i have hence switched to Array as primary for the share i'm moving to bypass the cache due to the issues below)
• I began (and still am) in the process of migrating files from my old server to the unraid machine.
• While that was ongoing I had a few hardware issues with PCIe bus spamming my log that caused the unraid machine to halt with 100% log usage.  I found the issue was my SATA expansion card and the PCI slot it was in; moved to a different slot and now works fine.  However, I did have to do a few hard reboots in the interim which I suspect stranded some of my files on the cache disk (and i can see them there)
• All my data seem to be there, but the cache has files on it and when I run mover, nothing happens.
• MOreover, when I go to VM/Docker config, since as early as I can remember it always says "share not present" (for /appdata etc), at first I thought I needed to manually create these so I put it off, but now as I search through docs and videos etc. they all seem to assume that those shares are auto-created when unraid is installed (and I was getting frustrated!) but i gather those shares should be there when a pool is made
• I've also added a new 128 GB SSD to my cache pool hoping that might "fix" it but that didn't do anything other than add it to the pool (which is fine)
• Finally, I never understood what unraid is reporting as far as my cache pool size is, from what i understand, a 1TB + 500 GB + 128 GB pool should have a capacity of 500+128=628GB but it doesn't seem to report that.

 

I am running unraid trial 7.0.0.  Below is a screenshot of Main.  When I run mover (or schedule runs) it won't move those 357 GB off the cache.

Ive also attached diagnostics.

 

My question:  How do I get those files off the cache (to the array) and then rebuild/fix the cache to get all shares needed for Docker/VMs to work?

 

THANKS!

 

 

ds9-diagnostics-20250117-2007.zip

Which share name(s) do you expect to be moved?

The only share that has files on the cache is the one that in the diagnostics shows as:

v----s                            shareUseCache="no"      # Share exists on disk1, disk2, disk3, cache

With these settings mover will ignore the share so any files on the cache will stay there.

 

Move will only take action on shares which have both primary and secondary storage set and the desired mover direction specified.

 

Perhaps a more pertinent issue is how did those files end up on the cache in the first place?   Something must have happened to stop Unraid writing them directly to the array in the first place.

 

 

The share is "videos", see screenshot below for it's current setting (primary = array), however to clarify my OP, this share was originally set as cache->array, and I had moved some files over, but I suspect something happened when an unclean shutdown occurred.  I have hence changed the share to primary=array while I continue the migration.

 

Should I set the share back to primary=cache and then run mover again? 

 

The larger issue though is that none of the Docker/VM shares seem to be present (see second sreenshot) and I'm not sure why or how to fix (as I gather these should've been created at pool creation but doesn't seem like it)

 

Thanks!

 

 

 

With only primary storage set mover will take no action on the share (note that setting for primary mentions it only applies to NEW files/folders).   To get files/folders already on cache moved you need cache set as primary, array as secondary, and mover direction as cache->array.    When everything is moved to the array you can then (optionally) having just the array as primary storage.

 

15 minutes ago, darwinsbeard said:

The larger issue though is that none of the Docker/VM shares seem to be present (see second sreenshot) and I'm not sure why or how to fix (as I gather these should've been created at pool creation but doesn't seem like it)

They only get created when you enable the service (docker or VMs) that need them.   You currently have these disabled.

Ha...well that was easy (and stupid of me).  I saw those warnings and thought that they would need to be resolved *first* so didn't even bother trying to start it.  Seems to work now