Cache Drive Full - need help with some commands

[ledfortr]

Hi - I'm having some challenges with my cache drive. I noticed the other day that it's full (250gb).  I ran the mover and nothing came off the drive.

I set all my shares that were cache only to prefer cache to allow for the main array to be used until I figured out what was going on.  That didn't make a difference.

I had an extra drive laying around, so I added it to the cache pool and now its 750gb.  I ran a parity check and the mover again and nothing happened.  I checked again this morning and the entire cache drive is full again.  

Is there a command that I can run to see what is taking up the space? I'm not great on linux, so I'm not sure how to do this.

See attached diag.

[JorgeB]

15 minutes ago, ledfortr said:

I set all my shares that were cache only to prefer cache to allow for the main array to be used

You must set them to cache=yes and run the mover, cache=prefer moves from array to pool.

[itimpi]

14 minutes ago, ledfortr said:

I set all my shares that were cache only to prefer cache to allow for the main array to be used until I figured out what was going on.

That will not cause anything to be moved off the cache (the Prefer setting means you want files moved TO the cache).   If you want them moved off the cache then use the Yes setting.  The help built into the GUI explains what the various settings do for new files and how they interact with the mover application.
 

You can use the Compute All button on the Shares tab to find out how much of each share is on the cache.

[ledfortr]

16 minutes ago, JorgeB said:

You must set them to cache=yes and run the mover, cache=prefer moves from array to pool.

Shares were set to cache=yes when I attempted the mover earlier. I can swap them back and try again. I would expect the same result.

 

[trurl]

attach diagnostics to your NEXT post in this thread

[trurl]

48 minutes ago, ledfortr said:

See attached diag.

Looks like you forgot to attach.

[ledfortr]

11 minutes ago, trurl said:

Looks like you forgot to attach.

I see why it didn't attach, I ran the diag and copied it into the thread, however, it didn't upload because the diag is apparently empty.  0 kb  Probably because of the disk full problem. 

Is there a way to run the diag command from the CLI? Then I click the download, I actually see the script running and data being looked at.

[trurl]

The word Diagnostics in every post is a link explaining how to get them 

 

What do you get from command line with this? 

df -h

 

[trurl]

1 hour ago, ledfortr said:

diag is apparently empty.  0 kb  Probably because of the disk full problem. 

That shouldn't cause diagnostics to fail, possibly you have filled rootfs somehow, which is why I asked

 

49 minutes ago, trurl said:

What do you get from command line with this? 

df -h

 

[trurl]

If necessary, reboot and get diagnostics. In any case, stop anything writing to your server until you get this resolved or you may get corruption

[ledfortr]

35 minutes ago, trurl said:

That shouldn't cause diagnostics to fail, possibly you have filled rootfs somehow, which is why I asked

Looks like its not full.

The mover is still running.  I'll reboot after its finished and report back.

 

[ledfortr]

The mover finally finished last night and it appears to be successful.

It looks like some of my dockers aren't functioning (unifi), so I have some troubleshooting to do. 

 

I'll keep updating this thread.  I think I may have to restore from backup to fix all my dockers.

[trurl]

Diagnostics would help us give advice