Always being cautious of upgrading too soon, I delayed upgrading to this subversion for some time. Did a backup of the flash through unraid connect, appdata with all vm's and dockers and then finally decided to bite the bullet and click on upgrade.

 

Boy did I regret this... the machine did not respond when booting, had to hard reset it, after which it did reboot. For some reason VM's and Dockers were not visible, could not understand why, until I noticed that the NVME SSD I run the appdata, VMs and dockers from displayed an error with regard of files not being able to read. The SATA600 SSD that ordinarily would have run as its companion failed a few weeks ago. Its replacement lies next to me on my desk, waiting to be installed (delayed due to private circumstances). The machine including this SSD is barely a year old, based on AMD 5600G / 32GB / B550 board with dedicated hardware SATA controller, 1xNVME+1xSATA600 as appdata, 2xSATA600 as cache and 5x(3x data+ 2x parity) 16TB Toshiba's as array. The NVME-drive is a Samsung 1TB drive, that had at least 93% life in it after several reinstalls and previously also having served as cache during repeated large amounts of data ingestion (now two separate SSD's).

 

Fortunately the shares still work, but I need the Dockers and VMs to also work ASAP. I am not a linux geek and would more describe myself as an enthousiast. I have some experience with this OS and have always found Unraid a breath of fresh air due to the (relatively) ease of GUI. Aside that the machine hosts literally all my private data including pictures and films of my family and important documents, I also use the machine for business purposes including running continuous offsite backups and internal backups of the workstations. I was planning to have a second machine run continuous backups of tier1 important documents just to prevent these kinds of issues.

 

Your kind assistance is greatly appreciated.

 

 

 

Edited December 8, 20232 yr by Bert Roekhoest

Post the output of:

btrfs fi show

 

In the meanwhile I had to decide to reformat the drive to make the VM's and dockers accessible again, hopefully through the backup I made minutes before the Unraid upgrade. Still working on this and no idea on how the partition apparently could just somehow disappear, which is worrying.

 

The defective SATA600 drive (second app drive) is still in the system.

 

 

Label: none  uuid: 706c869c-9c22-44b5-9cfc-80b6fcc023d8
        Total devices 2 FS bytes used 25.34GiB
        devid    1 size 931.51GiB used 27.03GiB path /dev/sdd1
        devid    2 size 931.51GiB used 27.03GiB path /dev/sdc1

Label: none  uuid: 268e6eeb-0240-4423-bf65-79c71374547e
        Total devices 1 FS bytes used 3.63GiB
        devid    1 size 30.00GiB used 4.52GiB path /dev/loop2

Label: none  uuid: ffbd2d91-6c42-41d6-9eb5-4ccfcdc55909
        Total devices 1 FS bytes used 1.08MiB
        devid    1 size 1.00GiB used 126.38MiB path /dev/loop3

Label: none  uuid: ccbd834d-2bb6-441c-86bd-b7e9e25f64a1
        Total devices 1 FS bytes used 15.10GiB
        devid    1 size 931.51GiB used 18.07GiB path /dev/nvme0n1p1

warning, device 1 is missing
Label: none  uuid: c30a7c3c-8d24-4ef4-87d4-37d8a1af58a5
        Total devices 2 FS bytes used 281.63GiB
        devid    2 size 931.51GiB used 305.03GiB path /dev/sde1
        *** Some devices missing

Edited December 7, 20232 yr by Bert Roekhoest

53 minutes ago, Bert Roekhoest said:

Label: none  uuid: ccbd834d-2bb6-441c-86bd-b7e9e25f64a1
        Total devices 1 FS bytes used 15.10GiB
        devid    1 size 931.51GiB used 18.07GiB path /dev/nvme0n1p1

warning, device 1 is missing

 

53 minutes ago, Bert Roekhoest said:

Label: none  uuid: c30a7c3c-8d24-4ef4-87d4-37d8a1af58a5
        Total devices 2 FS bytes used 281.63GiB
        devid    2 size 931.51GiB used 305.03GiB path /dev/sde1
        *** Some devices missing

 

This had to be a problem before the updated, but possibly it was no checked, so it would still mount with the device missing.

 

53 minutes ago, Bert Roekhoest said:

The defective SATA600 drive (second app drive) is still in the system.

is this sde?

 

If yes, try physically disconnecting that device from the server, then start the array and post new diags

Aside the 5xHDD array, I have a pool consisting of 2x SATA600 SSD for cache and I used to have a second pool consisting of 1xNVME SSD + SATA600 SSD for appdata dockers vm's etc.

 

The third SATA600 (indeed SDE) failed some weeks ago, so I removed it from the appdata etc. pool and continued my merry way without issues.

 

As soon as I get around to it, it will be replaced by an identical drive and then added back into the pool for appdata etc. However there was no problem, including rebooting, until I ran the update.

 

I am currently re-installing the dockers and hoping I will be able to flip the backup back onto the nvme-drive; the latter seems to work just fine, SCRUB ran for 9 seconds and tested without errors etc., although it seems not the entire drive is tested using this function. SMART-testing does not work although I understand this is particular to nvme's.

Edited December 8, 20232 yr by Bert Roekhoest