Mover not clearing cache, even when ran manually

[John9]

Greetings. Fair warning, I am very new to unRAID, so I may be congesting this forum with an ultimately dumb question. 

 

So far, all I have setup on my unRAID server is Plex, and MakeMKV (to rip dvd's into plex). I used guides on the forums to set them up, but they seemed straight forward enough. From what I have seen on google, the cache settings will cause files to remain and be skipped by mover (or even cause mover to add TO the cache rather than subtract from it). I have gone over my cache settings again and again, but I can't figure it out. I am also puzzled by the free memory column. I don't have a single VM yet, so why is it the only thing showing the right amount of free storage on my array? Everything else is showing the remainder of my NVME cache, and nothing shows the SSD cache remaining memory. What did I mess up?

Edited May 11, 2022 by John9

[trurl]

attach diagnostics to your NEXT post in this thread

[John9]

unraidserver-diagnostics-20220511-1745.zip

[trurl]

Mover ignores files that are not on the designated pool.

 

Your appdata share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover.

 

Your domains share is set to move files from cache to the array. All its files are on the array. Often people will want this share to prefer to stay on fast pool so VMs will perform better and not keep array disks spunup. Depends on how large your vdisks are. VMs can access Unraid storage so maybe not necessary to create large vdisks.

 

Your isos share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover. Maybe not even necessary to have that share on fast pool since it may not need to be accessed by VMs except for installs.

 

Your NAS share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover. Are you sure you want this share to prefer to stay on fast pool?

 

Your PLEX share is set to move files from cache to the array, but it also has files on nvme_cache, which will be ignored by mover.

 

Your system share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover.

[itimpi]

I think there must be something wrong with your cache setup as it look as if both the ssd devices are mounting into both pools which is confusing things.  Did you at any point have them both in the same pool.?

[trurl]

2 minutes ago, itimpi said:

I think there must be something wrong with your cache setup as it look as if both the ssd devices are mounting into both pools which is confusing things.  Did you at any point have them both in the same pool.?

I should have looked there first. Another user did the same thing recently.

 

[John9]

 

24 minutes ago, trurl said:

Mover ignores files that are not on the designated pool.

 

Your appdata share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover.

 

Your domains share is set to move files from cache to the array. All its files are on the array. Often people will want this share to prefer to stay on fast pool so VMs will perform better and not keep array disks spunup. Depends on how large your vdisks are. VMs can access Unraid storage so maybe not necessary to create large vdisks.

 

Your isos share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover. Maybe not even necessary to have that share on fast pool since it may not need to be accessed by VMs except for installs.

 

Your NAS share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover. Are you sure you want this share to prefer to stay on fast pool?

 

Your PLEX share is set to move files from cache to the array, but it also has files on nvme_cache, which will be ignored by mover.

 

Your system share is set to prefer nvme_cache, but it also has files on cache, which will be ignored by mover.

Why is everything going onto both pools? How do I stop this and fix it so it can be moved to the array?

12 minutes ago, itimpi said:

I think there must be something wrong with your cache setup as it look as if both the ssd devices are mounting into both pools which is confusing things.  Did you at any point have them both in the same pool.?

Yes, actually. When I first set it up, I wanted two pools, but forgot to add the second pool and both cache drives went into the same pool. I used new config to split them into separate pools. But you're right, it's acting like it's all one pool still. How do I move all the files where they belong and prevent this from occurring again? 

Edited May 11, 2022 by John9

[trurl]

See the other thread I linked above.

[John9]

15 minutes ago, itimpi said:

I think there must be something wrong with your cache setup as it look as if both the ssd devices are mounting into both pools which is confusing things.  Did you at any point have them both in the same pool.?

I just noticed another odd thing. Both my cache pools show capacities of 750gb each, but pool 1 has a 1tb drive and pool 2 has a 500 gig drive. Are these pools setup as raid1 even though they are separate pools? I am beyond confused. I messed something up for sure 

[itimpi]

The problem is that both drives are being added to both pools, so they have not been properly separated at the btrfs level from when they were in the same pool.   Hopefully the linked thread will help with a way to recover from this.

[John9]

42 minutes ago, itimpi said:

The problem is that both drives are being added to both pools, so they have not been properly separated at the btrfs level from when they were in the same pool.   Hopefully the linked thread will help with a way to recover from this.

I am still waiting for the BTRFS operation to complete. Is it also doing something to the array? I wouldn't have thought it would take so long considering the drives in question are flash storage.. Then again, like I said before, I am a real noob here. 

[John9]

13 hours ago, itimpi said:

The problem is that both drives are being added to both pools, so they have not been properly separated at the btrfs level from when they were in the same pool.   Hopefully the linked thread will help with a way to recover from this.

 

13 hours ago, trurl said:

I should have looked there first. Another user did the same thing recently.

 

Thank you both for the help! My issue is resolved, and everything is running smoothly! THANK YOU!