February 7, 20242 yr I noticed on start up today that there was a bdev/dev/loop2 error and that the docker.img does not exist? All my dockers seem to be running just fine. Every few days I was having lockups, but I did a Memtest and found bad ram, so I swapped that out. Is this something that I should be worried about? Docker is on cache drive NVME Raid 1. I included diag and syslog. Any in-site would be appreciated. Feb 6 19:23:45 Goliath emhttpd: shcmd (144): /usr/local/sbin/mount_image '/mnt/user/system/docker/docker.img' /var/lib/docker 20 Feb 6 19:23:45 Goliath root: Specified filename /mnt//system/docker/docker.img does not exist. Feb 6 19:23:45 Goliath kernel: loop2: detected capacity change from 0 to 62914560 Feb 6 19:23:45 Goliath kernel: BTRFS: device fsid 8872dcb8-526f-41b4-9cc5-796095101428 devid 1 transid 1087604 /dev/loop2 scanned by mount (19888) Feb 6 19:23:45 Goliath kernel: BTRFS info (device loop2): using crc32c (crc32c-intel) checksum algorithm Feb 6 19:23:45 Goliath kernel: BTRFS info (device loop2): using free space tree Feb 6 19:23:45 Goliath kernel: BTRFS info (device loop2): bdev /dev/loop2 errs: wr 0, rd 0, flush 0, corrupt 1, gen 0 Feb 6 19:23:45 Goliath kernel: BTRFS info (device loop2): enabling ssd optimizations Feb 6 19:23:45 Goliath root: Resize device id 1 (/dev/loop2) from 30.00GiB to max Feb 6 19:23:45 Goliath emhttpd: shcmd (146): /etc/rc.d/rc.docker start Feb 6 19:23:45 Goliath root: starting dockerd ... goliath-diagnostics-20240206-1958.zip goliath-syslog-20240207-0056.zip
February 7, 20242 yr maybe off topic, but my docker image file corrupted 4 times at least in 2 years and i have given up on BTFRS suggest to use XFS from the dropdown and recreate all the stuff, never had any issue again with corruption seems to come cause of disk filled up with logs and shit.
February 7, 20242 yr Author 5 minutes ago, astronax said: maybe off topic, but my docker image file corrupted 4 times at least in 2 years and i have given up on BTFRS suggest to use XFS from the dropdown and recreate all the stuff, never had any issue again with corruption seems to come cause of disk filled up with logs and shit. If I switch to XFS from the drop down, will I have to re-configure everything? It's not a big deal, just some extra work. lol.
February 7, 20242 yr normally you should only have to go to the docker tab, click add below and select from the dropdown settings should be save, so just hit apply and see download running
February 7, 20242 yr Community Expert Memory can corrupt everything. You shouldn't even attempt to run any computer unless memory is working perfectly. The syslog snippet you posted seems to indicate it recreated docker.img, maybe you changed its size? I didn't notice anything wrong. 56 minutes ago, soulstrife said: All my dockers seem to be running just fine.
February 7, 20242 yr Author 2 minutes ago, trurl said: Memory can corrupt everything. You shouldn't even attempt to run any computer unless memory is working perfectly. The syslog snippet you posted seems to indicate it recreated docker.img, maybe you changed its size? I didn't notice anything wrong. Yeah, I kept getting corrupt / memory errors, so I ran Memtest and it almost instantly failed. I have new memory that has passed 2x. I have not changed the size of docker.img, I believe I changed it to 30GB way way way back when I first started the server and have not touched it since.
February 7, 20242 yr Community Expert config/docker.cfg has DOCKER_IMAGE_SIZE="20" That is what it would use if you recreated it. But since it is already 30, it uses that, and that is what the syslog entries are indicating.
February 7, 20242 yr Author 1 minute ago, trurl said: config/docker.cfg has DOCKER_IMAGE_SIZE="20" That is what it would use if you recreated it. But since it is already 30, it uses that, and that is what the syslog entries are indicating. Interesting. I have never changed that back to 20. Is there any way that it could have defaulted back to that? I did have to change the macvlan back to ipvlan a few days ago which I thought was odd. If I change that 20 back to 30, do you think it may fix the issue?
February 7, 20242 yr Author 6 minutes ago, trurl said: config/docker.cfg has DOCKER_IMAGE_SIZE="20" That is what it would use if you recreated it. But since it is already 30, it uses that, and that is what the syslog entries are indicating. I changed it to back to 30. I know that wouldn't take care of the corruption, but switching to XFS instead of BTRFS for docker will fix the issue? Or is there a way I can scan the docker.img and purge / fix the corruption? Also, why does it say "root: Specified filename /mnt//system/docker/docker.img does not exist." ?
February 11, 20242 yr Author Solution I moved everything to Docker Folders and got rid of BTRFS. All good now.
February 11, 20242 yr Community Expert On 2/6/2024 at 9:05 PM, soulstrife said: ran Memtest and it almost instantly failed Probably bad memory was the only problem, and the changes you made didn't really do anything except recreate docker image, which got rid of the corruption caused by bad memory. Probably just recreate https://docs.unraid.net/unraid-os/manual/docker-management/#re-create-the-docker-image-file and reinstall https://docs.unraid.net/unraid-os/manual/docker-management/#re-installing-docker-applications would have fixed it also.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.