I tried to expedite a cache upgrade from a 1.6TB SAS to a 3.2TB F320 NVMe by adding the NVMe to the pool, then changing btfrs to a R1 (with the idea that I'd remove the 1.6TB once cloned and go back to single mode).
Well, that did not work as planned.

I've tried to remove the F320 NVMe and try to convert back to single mode with no luck, it errors on something?
I've tried to scrub the cache, but that quickly errors out in both cache drives populated and with the nvme removed.
I've also tried to revert back to the safe method of changing cache shares from prefer to yes and running the mover, that also errors out and leaves me with some empty and some full directories on my array but files still on the server.

 

The cache drive is reporting the wrong (smaller than 1.6TB) size and isn't allowing me to r/w to /mnt/cache just r/o.

Does anybody have any suggestions on how to recover the cache or proceed with the upgrade?

I've attached the diagnostic files

diagnostics-20230517-1720.zip

Edited May 18, 20233 yr by jonnygrube

            Data    Data    Metadata Metadata  System    System                            
Id Path      single  RAID1   single   RAID1     single    RAID1    Unallocated Total   Slack
-- --------- ------- ------- -------- --------- --------- -------- ----------- ------- -----
 1 /dev/sdc1 1.45TiB 3.00GiB  6.00GiB   1.00GiB  32.00MiB 32.00MiB     1.05MiB 1.46TiB     -
 2 missing         - 3.00GiB        -   1.00GiB         - 32.00MiB     2.91TiB 2.91TiB     -
-- --------- ------- ------- -------- --------- --------- -------- ----------- ------- -----
   Total     1.45TiB 3.00GiB  6.00GiB   1.00GiB  32.00MiB 32.00MiB     2.91TiB 4.37TiB 0.00B
   Used      1.37TiB 2.80GiB  3.01GiB 780.20MiB 144.00KiB 48.00KiB  

 

Pool is kind of a mess, there's a missing device and the one assigned is completely full, you should be able to mount it read only, copy any important data elsewhere and re-format:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=543490

 

JorgeB-
Thank you for your help, I seem to have made a series of ill advised moves here.
I thought removing the drive would allow me to remove the R1 and fix the read errors that i'm getting when trying to copy data off it.
It did not, so i readded it. See below for updated diagnostics.

For background, the 1.6TB one that's completely full had ~80TB free prior to trying to R1 on single mode.  I have no idea why it's reporting full or why the GUI is reporting 757GBused/212MBfree yet 2.4TB in total size. I keep trying to pull any files off it with midnight commander, but i keep getting readonly and io errors.

My goal is to pull the critical files and VM's off the 1.6TB and restore from an appdata backup.  How should I proceed?

diagnostics-20230518-1448.zip

Edited May 18, 20233 yr by jonnygrube

13 hours ago, jonnygrube said:

How should I proceed?

See the link above to mount the pool read only, then try and copy the data.