I'm making my second attempt at upgrading from 6.12.13 to 7.0. The first attempt failed when I realised that all my docker apps had vanished, and I downgraded back to 6.12.13 as I wasn't really in the right headspace to fix it. 

 

Today, after running the update assistant and uninstalling the nerdtools plugin as it recommended, and also fixing something that the fix common problems plugin had highlighted (my system and appdata shares were set to use the Music pool instead of cache, so the docker image and some of the appdata files were on the wrong pool - I didn't understand that at the time, because there's no way I would have selected the Music pool for those files) I upgraded again and have hit the same problem. This time however, I'm using my rational brain and investigating it properly, and I've seen that the 1TB btrfs SSD I use for my cache pool shows an 'Unmountable: wrong or no file system' message. That means that my shares have defaulted to the Music pool, so the docker image has moved to the wrong pool again and so has the appdata folder, although that is empty because nothing is running.

 

I'm pretty confident the SSD is fine, because I didn't have any problems running on 6.12.13, so I don't know why it's throwing this error after going to 7.0. I can't browse it to confirm that it's definitely good, but I didn't have to do anything special to fix it after I downgraded last time, so I can't see it being corrupted. I'm 99% sure that if I took that disk out and plugged it into something else, I would see all the files as normal.

 

What is the best course of action for fixing this? Do I need to copy all the data off and reformat the disk, then set up the pool again? I'd rather not do that if I don't have to, it'd just be a hassle. I'd also prefer not to downgrade again, I want to fix this properly.

 

Anonymised diagnostics are attached.

 

Thanks.

tower-diagnostics-20250216-1145.zip

Feb 16 11:18:39 Tower kernel: BTRFS error (device sdk1): parent transid verify failed on logical 317460430848 mirror 1 wanted 1745220 found 1746244
Feb 16 11:18:39 Tower kernel: BTRFS error (device sdk1): Error removing orphan entry, stopping orphan cleanup
Feb 16 11:18:39 Tower kernel: BTRFS error (device sdk1): could not do orphan cleanup -22

 

There is a filesystem issue and this error is typically fatal with btrfs, if it mounts with v6.12, the error may be undetected by the older kernel, in that case, I would recommend downgrading, backing up the pool, then upgrade re-format it and restore the data, you can use this to help with the backup/restore:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/#findComment-511923

 

Thanks JorgeB. That was a bit of a pain to go through moving all that data twice (especially when the mover just refused to bring the data back from the array to the cache pool after the upgrade), but I've re-formatted my cache SSD and got everything working again on 7.0. I did have to reinstall all my docker apps, but that's no great pain.