Replacing cache drive - docker scrub errors

[Nem]

I had unraid with docker running fine before my server upgrades. One of the upgrades was replacing the cache drive with a larger one. I followed the instructions here: https://lime-technology.com/wiki/index.php/Replace_A_Cache_Drive

 

After copying docker.img back to the cache drive and restarting the array everything appears to show up OK. The dockers appear to be working.

 

When I go into the docker settings and scrub, however, I get 2 uncorrectable errors:

 

scrub status for 73b6b03e-f0d6-498a-ae48-cbea42522c54
scrub started at Sat Dec  5 19:52:31 2015 and finished after 00:00:11
total bytes scrubbed: 5.66GiB with 2 errors
error details: csum=2
corrected errors: 0, uncorrectable errors: 2, unverified errors: 0

 

I check the log and I see this:

 

Dec  5 19:52:31 Server php: /usr/local/emhttp/plugins/dynamix/scripts/btrfs_scrub 'start' '/var/lib/docker'
Dec  5 19:52:36 Server kernel: BTRFS: checksum error at logical 3003174912 on dev /dev/loop0, sector 7979112, root 395, inode 1207326, offset 5152768, length 3640, links 1 (path: var/log/auth.log)
Dec  5 19:52:36 Server kernel: BTRFS: bdev /dev/loop0 errs: wr 0, rd 0, flush 0, corrupt 7, gen 0
Dec  5 19:52:36 Server kernel: BTRFS: unable to fixup (regular) error at logical 3003174912 on dev /dev/loop0
Dec  5 19:52:42 Server kernel: BTRFS: checksum error at logical 7370739712 on dev /dev/loop0, sector 16509512, root 395, inode 26832, offset 327606272, length 3936, links 1 (path: var/log/error)
Dec  5 19:52:42 Server kernel: BTRFS: bdev /dev/loop0 errs: wr 0, rd 0, flush 0, corrupt 8, gen 0
Dec  5 19:52:42 Server kernel: BTRFS: unable to fixup (regular) error at logical 7370739712 on dev /dev/loop0
Dec  5 19:52:48 Server kernel: BTRFS warning (device loop0): csum failed ino 1207326 off 5152768 csum 3845547863 expected csum 1038413045
Dec  5 19:52:48 Server kernel: BTRFS warning (device loop0): csum failed ino 1207326 off 5152768 csum 3845547863 expected csum 1038413045
Dec  5 19:52:48 Server kernel: BTRFS warning (device loop0): csum failed ino 26832 off 327606272 csum 1778903515 expected csum 790652044
Dec  5 19:52:48 Server kernel: BTRFS warning (device loop0): csum failed ino 26832 off 327606272 csum 1778903515 expected csum 790652044

 

And whenever I open up the webgui of a docker this warning just keeps repeating in the log:

 

Dec  5 19:53:48 Server kernel: BTRFS warning (device loop0): csum failed ino 1207326 off 5152768 csum 3845547863 expected csum 1038413045
Dec  5 19:53:48 Server kernel: BTRFS warning (device loop0): csum failed ino 1207326 off 5152768 csum 3845547863 expected csum 1038413045
Dec  5 19:53:48 Server kernel: BTRFS warning (device loop0): csum failed ino 26832 off 327606272 csum 1778903515 expected csum 790652044
Dec  5 19:53:48 Server kernel: BTRFS warning (device loop0): csum failed ino 26832 off 327606272 csum 1778903515 expected csum 790652044

 

Is there a way to fix this? I didn't get any errors prior to moving the image to a larger cache drive

 

I'm not sure if this is related, but as part of the upgrades I removed the parity drive and had to turn it into another data drive (temporarily until I get a larger parity drive). To do this, I went into Tools > New Config, added all my old data drives to the array without a parity

[JorgeB]

Did you ran scrub without the –r (read only) option?

 

Also, if you still have your old cache drive and that one is error free you could use it to create a mirror with the new cache drive and then remove it, as long as its data is still up-to-date.

 

 

[lionelhutz]

Scrub can't correct anything on a stand-alone filesystem.

 

This is the only way I know to fix it. Shut-down your Dockers then turn-off Docker from the settings page. Delete the Docker image on the cache drive. Then, restart Docker. On the Docker tab, add Dockers and pick the my-XXXX templates where XXXX is the names of the Dockers you were running. They will all re-install just like they were before without any data or setup loss.

 

Post-up if it corrupts again.

[Nem]

That appears to have fixed things, thanks!

 

So what did I do that caused this problem? Just trying to understand so I can avoid this type of thing in the future

[lionelhutz]

That was quick. I'm not sure what the issue is which causes corruption. Others will tell you that your hardware which works 100% fine with a different filesystem has issues.