Messed up cache pools - getting errors on system shares

[Fotis Yanchev]

Hi Newbie here, 

have my first Unraid server for several weeks, all ok until today. The Fix Common errors plugin found errors on appdata, system and domains.

 

 

Attaching also the share settings for your ref. 

I have 2 cache pools, one is cache_nvme and one is cache_sata. The cache_sata is used with qbittorent to only download files and when ready are moved to the arrays for storage. Its an old sata I had laying around. The cache_nvme is a new nvme disk used for cache for the rest and also for system chares. Checked Scrutiny and it shows the disk powered on 1 day?!? Although the system is running for 4 weeks. The other new disks in the system show the correct powered on date of 32 days.  

Any suggestions or ideas are welcomed and highly appreciated. As far as I could check all installed docker apps are OK and running. I do not have any VMs yet. 

[itimpi]

Unraid will not automatically move existing files when you change the pool associated with a share.

 

The easiest (and most efficient) thing to do is to move them manually to the correct location.  Alternatively you can exploit mover as described here in the online documentation that can be accessed via the Manual link at the bottom of the GUI

[Fotis Yanchev]

I did not change the pool, 30 days ago it was set as described, cache_sata only for Downloads and the anything else uses cache_nvme. Above errors are from today without any changes made to shares. Will look up in above links. 

[Squid]

14 minutes ago, Fotis Yanchev said:

Checked Scrutiny and it shows the disk powered on 1 day

I've got a nvme that does the exact same thing.  Had it for around a month or 2 and is currently showing 9 days.  Too lazy to investigate why (and it doesn't particularly bother me)

 

To use the mover method as linked above, you stop the docker and VM services in settings, then set those 3 shares to use the cache_sata pool with a use cache setting of yes.

(if you're running 6.10, stop and start the array after doing this)

Run mover and the contents will be moved from the sata pool to the array.  

 

Next, set those 3 shares to use the cache_nvme pool with a use cache setting of prefer. 

(if you're running 6.10, stop and start the array after doing this)

Run mover and the contents will get moved from the array to the proper pool

 

You're now good to go.

 

This happened because you had previously had the settings reflecting the cache_sata pool (or renamed a pool etc) but didn't move all of the files off of it

 

 

[Fotis Yanchev]

Thank you both.

 

The 1 days of nvme does not bother me at all, mentioned it in case it causes the problem.

 

Running 6.9.2, will try the solution and if it works I will mark it as solved.

Regards