Help! Cache drive lost, then recovered, but now Unraid seems to want to reformat it?

[Skeptic]

I searched the forums for an answer, but this appears to be a pretty unique situation, so here goes. 

There were other problems, all now apparently ok, but this is what remains. 

I've got a cache pool of 2 drives, a 120gb SSD and a 240gb SSD. The 120 was the original, and the 240 was added later. I only mention this before the bulk of the data appears to have been the 120gb drive, which failed. 

I had a power outage (I've since added a UPS so this hopefully won't happen again), and the 120gb cache drive did not show up in the pool. It was just missing, along with all my dockers and VMs. Apparently, this is a known issue with SSDs, and with the "power cycle method", I was able to get the cache drive back. Unraid listed it as an unassigned drive. When I attempt to add it back to its slot in the cache pool, Unraid tells me it will be formatted when I start the array. 

I was able to mount the drive, add it as a share, and copy all the data off, and it appears to be intact. I've got the appdata, domains, and system folders, and I was able to copy the data off to another drive. 

So, is it possible to re-add the drive to the cache pool, with file structure intact? Or, should I just let Unraid format it, and then somehow restore the data to it, hopefully restoring my dockers and VMs? (and if so, how?)

I also now have another 1tb SSD, which I could use rather than a pool of the 2 drives, if that ends up being easier.

 

I've been running Unraid for just over a year, and this is my first problem (Other than the fact that now Unraid always tells me the UPS cable is unplugged, but one thing at a time). Any assistance you can offer to a relative newb would be greatly appreciated.

 

 

[trurl]

What was the raid configuration of the pool? I ask because there are several possibilities, with the default being raid1. But that default would just give you a mirror and so the total capacity of the pool for those different sized disks would just be the size of the smaller, 120GB.

 

 

 

[Skeptic]

The cache pool size was the total of 360gb. Thanks for the response!

 

[JorgeB]

6 hours ago, Skeptic said:

So, is it possible to re-add the drive to the cache pool, with file structure intact?

If the pool was single profile like you suggest it won't be possible to join the 2 devices leaving the data intact, but it also doesn't make much sense it mounted by itself, it's possible UD mounted the pool, post diagnostics with the device mounted.

[Skeptic]

 

This is what it looked like originally.

 

This is what it looks like now. The 120gb OCZ shows as unassigned, but if I assigned it back to the first slot in the cache pool, it looks like this:

 

 

Does this help? I'd really like to get my dockers and VMs back. Should I just let Unraid reformat it, and can I then just copy the data back? 

 

Thanks for your assistance!

 

[JorgeB]

Likely the pool is still working, but if you assign the missing device as is it will format it and make the pool unmountable, to confirm mount the device again with the UD and post the output of:

 

btrfs fi usage -T /mnt/disks/UD_path

 

[Skeptic]

root@UnRaid:~# btrfs fi usage -T /mnt/disks/OCZ-VERTEX3_OCZ-RTM5B812R019QDQ2
Overall:
    Device size:                 335.37GiB
    Device allocated:            174.06GiB
    Device unallocated:          161.31GiB
    Device missing:                  0.00B
    Used:                        155.75GiB
    Free (estimated):            176.64GiB      (min: 95.99GiB)
    Data ratio:                       1.00
    Metadata ratio:                   2.00
    Global reserve:               73.38MiB      (used: 0.00B)

             Data      Metadata  System
Id Path      single    RAID1     RAID1    Unallocated
-- --------- --------- --------- -------- -----------
 2 /dev/sdn1 141.00GiB   2.00GiB 32.00MiB    80.55GiB
 1 /dev/sdo1  29.00GiB   2.00GiB 32.00MiB    80.76GiB
-- --------- --------- --------- -------- -----------
   Total     170.00GiB   2.00GiB 32.00MiB   161.31GiB
   Used      154.66GiB 554.38MiB 48.00KiB

 

Again, thanks!

 

[JorgeB]

So as expected the pool is still functioning, just the Unraid cache config was lot, easiest way to fix it to unassign all cache devices, start the array so that Unraid "forgets" current cache config, then stop the array, assign both devices and start again.

[Skeptic]

That worked, THANK YOU! 

Everything is back. I wish I could shake your hand, and buy you a beer or coffee!

 

Thanks again!!

[JorgeB]

👍