Hi All

I am panicking now as I stopped the array, de-assigned my old small cache drive, assigned my new larger cache drive & restarted the array.

BUT... I now have no shares...

No dockers (says "Docker Service failed to start.")...

& No VMs (says "Libvirt Service failed to start.")...

Array is running and new cache disk is showing as "Unmountable: wrong file system"... but I have the format option available!

When I look directly on the disks I can see the shares;

But am at a loss what to do now?!?!

This means my production server is offline completely... I don't want to reboot and make it worse without some guidance.

Diags are attached

Can anyone help urgently?

Thanks in advance.

phoenix-diagnostics-20251009-1026.zip

Edited October 9, 2025Oct 9 by SliMat

Interestingly if I try and access the shares using \\192.168.76.25\ in Windows (the servers IP) I see no shares... but my mapped drives in Windows still work and show the data!?!?!

OK, desperate to try something as all my clients are offline... does anyone know if a host reboot will make things worse... or should I un-assign the (unformatted) large cache disk and re-assign the old smaller (empty) cache disk then restart array to see if that helps?

Any pointers appreciated.

1 hour ago, SliMat said:

de-assigned my old small cache drive, assigned my new larger cache drive & restarted the array.

Seems that pool was a single drive, how would you expect that if you remove a single drive with data on it it'd magically appear on the new one?

You need to manually copy the old drive contents to the new one.

1 minute ago, Kilrah said:

Seems that pool was a single drive, how would you expect that if you remove a single drive with data on it it'd magically appear on the new one?

You need to manually copy the old drive contents to the new one.

The old cache drive was empty... everything was moved to the array first... so I am not magically expecting any data.

My query is that after stopping the array, removing the small cache drive, assigning a larger (also empty) drive in the cache slot and restarting the array... all my shares/VMs/dockers have vanished - but shares are still accessible (at the moment) via mapped shares on a Windows machine.

Worried if I reboot it might make things worse :(

Check that nothing in Docker or VM settings directly maps to /mnt/[poolname]/[share] but map to /mnt/user/share instead.

Just now, Kilrah said:

Check that nothing in Docker or VM settings directly maps to /mnt/[poolname]/[share] but map to /mnt/user/share instead.

Pretty sure I moved everything to /mnt/user/share - but the issue is that I cant see any dockers now to check this...

Edited October 9, 2025Oct 9 by SliMat

I could stop array... disable VMs and dockers... bring array online again and then enable them one at a time to see if one works and one doesn't to try and see if its one or the other causing the issue?

OK, I have disabled dockers & disabled VMs in settings, restarted the array and still have no shares 😢!!

Edited October 9, 2025Oct 9 by SliMat

OK, I have unassigned the new large cache disk and reassigned the smaller one and still get no shares visible!!!

OK, am thinking that a complete reboot is the only thing left to try - any thoughts??

Oct 9 10:15:45 Phoenix root: mkdir: cannot create directory '/mnt/user': File exists

Reboot and the shares should come back up.

Thanks - Will try now... will let you know in 5-10 mins

Looks like everything has come back now... thanks @JorgeB

I just dont like rebooting when it goes wrong in case it makes things worse... I now also have my 2Tb SSD listed as a cache drive... I plan to leave all the dockers configs stored on /mnt/user/appdata - rather than moving back to /mnt/cache/appdata - is there a correct way to move the /mnt/user/appdata folder to use the cache disk only... or any threads you recommend to set the cache disk back up.

I plan to use the old 500Gb SSD cache disk (which is now unassigned) as a hot store for VM backups etc ;-)

Thanks