- Minor
Need a second set of eyes.
I upgraded from Unraid 6.x to 7.2.3 and dockers did not start (surprise!!!), searching I found the supposed cause was corrupt files/errors in the btrfs docker.img file.
I deleted the old one, create a new docker.img, and one by one added the dockers back, and I arrived at the same problem.
Log shows:
# cat /var/log/docker.log
failed to start containerd: timeout waiting for containerd to start
I do have a fair number of images in the docker file and suspect it might be a timeout issue, as I can start up docker fine manually or by turning it on/off in the GUI.
root@cerebellum:~# losetup /dev/loop3 /mnt/disks/HITACHI_SSD/docker/docker.img
root@cerebellum:~# mount /dev/loop3 /var/lib/docker
root@cerebellum:~# /etc/rc.d/rc.docker start
rc.docker: Starting Docker daemon...
rc.docker: Docker daemon... Started.
rc.docker: Starting network...
rc.docker: Processing... br0
rc.docker: connecting mysql to network br0
rc.docker: connecting librenms to network br0
rc.docker: connecting zabbix-agent to network br0
rc.docker: connecting Zabbix-Server to network br0
rc.docker: connecting Zabbix-Webinterface to network br0
rc.docker: Processing... eth2
rc.docker: Network started.
root@cerebellum:~# dmesg | grep -i btrfs
[ 44.611532] Btrfs loaded, zoned=no, fsverity=no
[ 710.105795] BTRFS: device fsid 7425f554-59f8-49c9-9635-a33a5ebebc83 devid 1 transid 75417380 /dev/sdc1 (8:33) scanned by mount (28559)
[ 710.106374] BTRFS info (device sdc1): first mount of filesystem 7425f554-59f8-49c9-9635-a33a5ebebc83
[ 710.106406] BTRFS info (device sdc1): using crc32c (crc32c-intel) checksum algorithm
[ 711.153698] BTRFS info (device sdc1): enabling free space tree
[ 711.159638] BTRFS info (device sdc1 state M): turning on async discard
[ 738.470578] BTRFS: device fsid 572852c1-eb8b-4771-87e3-4da2a4b35e36 devid 1 transid 20345 /dev/loop2 (7:2) scanned by mount (30434)
[ 738.473595] BTRFS info (device loop2): first mount of filesystem 572852c1-eb8b-4771-87e3-4da2a4b35e36
[ 738.473618] BTRFS info (device loop2): using crc32c (crc32c-intel) checksum algorithm
[ 738.496042] BTRFS info (device loop2): enabling ssd optimizations
[ 738.496049] BTRFS info (device loop2): turning on async discard
[ 738.496052] BTRFS info (device loop2): enabling free space tree
[ 803.176936] BTRFS info (device loop2): last unmount of filesystem 572852c1-eb8b-4771-87e3-4da2a4b35e36
[ 804.761909] BTRFS: device fsid e36c48e9-7d9d-4242-9ffd-80105f180a30 devid 1 transid 3103 /dev/loop2 (7:2) scanned by mount (31778)
[ 804.766091] BTRFS info (device loop2): first mount of filesystem e36c48e9-7d9d-4242-9ffd-80105f180a30
[ 804.766181] BTRFS info (device loop2): using crc32c (crc32c-intel) checksum algorithm
[ 804.874207] BTRFS info (device loop2): turning on async discard
[ 804.874222] BTRFS info (device loop2): enabling free space tree
[ 1207.603329] BTRFS: device fsid 572852c1-eb8b-4771-87e3-4da2a4b35e36 devid 1 transid 20346 /dev/loop3 (7:3) scanned by mount (35543)
[ 1207.603751] BTRFS info (device loop3): first mount of filesystem 572852c1-eb8b-4771-87e3-4da2a4b35e36
[ 1207.603772] BTRFS info (device loop3): using crc32c (crc32c-intel) checksum algorithm
[ 1207.612578] BTRFS info (device loop3): enabling ssd optimizations
[ 1207.612585] BTRFS info (device loop3): turning on async discard
[ 1207.612587] BTRFS info (device loop3): enabling free space tree
root@cerebellum:~# dmesg | grep -i loop3
[ 1186.960501] loop3: detected capacity change from 0 to 167772160
[ 1207.603329] BTRFS: device fsid 572852c1-eb8b-4771-87e3-4da2a4b35e36 devid 1 transid 20346 /dev/loop3 (7:3) scanned by mount (35543)
[ 1207.603751] BTRFS info (device loop3): first mount of filesystem 572852c1-eb8b-4771-87e3-4da2a4b35e36
[ 1207.603772] BTRFS info (device loop3): using crc32c (crc32c-intel) checksum algorithm
[ 1207.612578] BTRFS info (device loop3): enabling ssd optimizations
[ 1207.612585] BTRFS info (device loop3): turning on async discard
[ 1207.612587] BTRFS info (device loop3): enabling free space tree
If I revert back to the previous version of unraid 6.8.14, or 6.8.15, my dockers start up just fine.
I'd like to upgrade instead of reverting, but seeing so many posts on unraid 7 having this problem is somewhat concerning.
I'd also like to avoid having to tack those 3 lines into my startup script as a work around and fix whatever underlying problem is causing the issue.
Diagnostics attached.
Thanks
cerebellum-diagnostics-20260202-0101.zip
Edit: March 18 2026 - Unable to post further follow ups, but this seems to be the problem. Is there anyway to have the docker image work on unassigned devices and not required to be part of the cache pool? Seems like a bug...
thanks,
Jonathan