- Minor
I recently exchanged one of my cache disks from a Samsung EVO 850 500GB to a nvme 1TB Samsung EVO 860.
The first issue I noticed was that none of the docker containers worked as it seems that the docker.img file became read only while doing the initial replace (You probably understand better what happened from the log). Should the mounting of docker.img maybe wait until device replacement is finished? Anyway, that is not the main issue I wanted to report, and it was fixed by stopping and starting the docker service.
I followed the FAQ entry here:
After I replaced the drive I still got drive full messages when copying to a share with cache enabled. When I checked the unraid gui, the size of the cache pool was still saying 750 GB even though I now had two 1TB disks. After a quick check, I saw that the filesystem on the new 1TB drive was the same size as the old 500GB drive. After running btrfs filesystem resize 1:max /mnt/cache the issue was resolved.
I can see from the syslog that it does the resize command, but it looks like it's using the old device /dev/sdc1 and after the resize the device is changed to /dev/nvme0n1p1.
I know this is 6.7.2 and this might be changed in 6.8, but I'll post it in case it's not.
Below is the output of some btrfs commands before and after the resize command.
root@Server1:~# btrfs fi usage -T /mnt/cache
Overall:
Device size: 1.41TiB
Device allocated: 931.46GiB
Device unallocated: 512.39GiB
Device missing: 0.00B
Used: 624.15GiB
Free (estimated): 408.50GiB (min: 408.50GiB)
Data ratio: 2.00
Metadata ratio: 2.00
Global reserve: 130.98MiB (used: 0.00B)
Data Metadata System
Id Path RAID1 RAID1 RAID1 Unallocated
-- -------------- --------- --------- -------- -----------
1 /dev/nvme0n1p1 463.70GiB 2.00GiB 32.00MiB 465.78GiB
2 /dev/sdb1 463.70GiB 2.00GiB 32.00MiB 512.36GiB
-- -------------- --------- --------- -------- -----------
Total 463.70GiB 2.00GiB 32.00MiB 978.14GiB
Used 311.39GiB 699.41MiB 96.00KiB
root@Server1:~# btrfs filesystem show /mnt/cache/
Label: none uuid: ca04e038-59fa-4d03-a900-74b50150a144
Total devices 2 FS bytes used 312.07GiB
devid 1 size 465.76GiB used 465.73GiB path /dev/nvme0n1p1
devid 2 size 978.09GiB used 465.73GiB path /dev/sdb1
root@Server1:~# btrfs filesystem resize 1:max /mnt/cache
Resize '/mnt/cache' of '1:max'
root@Server1:~# btrfs filesystem show /mnt/cache/
Label: none uuid: ca04e038-59fa-4d03-a900-74b50150a144
Total devices 2 FS bytes used 312.28GiB
devid 1 size 931.51GiB used 465.73GiB path /dev/nvme0n1p1
devid 2 size 978.09GiB used 465.73GiB path /dev/sdb1
The command below was ran today as I forgot to run it after the resize.
root@Server1:~# btrfs fi usage -T /mnt/cache
Overall:
Device size: 1.86TiB
Device allocated: 959.46GiB
Device unallocated: 950.14GiB
Device missing: 0.00B
Used: 641.83GiB
Free (estimated): 632.56GiB (min: 632.56GiB)
Data ratio: 2.00
Metadata ratio: 2.00
Global reserve: 140.84MiB (used: 0.00B)
Data Metadata System
Id Path RAID1 RAID1 RAID1 Unallocated
-- -------------- --------- --------- --------- -----------
1 /dev/nvme0n1p1 477.70GiB 2.00GiB 32.00MiB 451.78GiB
2 /dev/sdb1 477.70GiB 2.00GiB 32.00MiB 498.36GiB
-- -------------- --------- --------- --------- -----------
Total 477.70GiB 2.00GiB 32.00MiB 950.14GiB
Used 320.21GiB 725.50MiB 112.00KiB