Docker Service failed to start

[razorhazor]

I've come back to my server after a while and the Docker service appears to be stopped.

 

I've tried deleting the docker.img, resizing it to 30GB and moving it off my cache drive to no avail.

 

I've also remove my parity drive from my array as its starting to kick out SMART errors.

 

Can someone see any clues in my diagnostic log?

 

Thanks!

 

tower-diagnostics-20180627-2023.zip

[razorhazor]

Bump! 

[Squid]

The image is messed up.  This is what you're going to have to do

 

1. On the flash drive, edit the file called config/docker.cfg   Change DOCKER_ENABLED from "yes" to be "no"
2. Restart the system.
3. Once its back up, go to Settings, Docker, advanced view and then delete the image
4. Re-enable Docker
5. Go to Apps, Previous Apps, check off what you want, and you'll be back in business in a couple of minutes.

 

If that doesn't get you anywhere, then I'd say it's some sort of corruption on the cache drive preventing the image from forming correctly (I'm not a big fan of BTRFS for cache drives unless you are planning on running a cache pool - XFS anecdotally causes people far less problems

Edited June 29, 2018 by Squid

[razorhazor]

On 6/29/2018 at 10:51 PM, Squid said:

The image is messed up.  This is what you're going to have to do

 

1. On the flash drive, edit the file called config/docker.cfg   Change DOCKER_ENABLED from "yes" to be "no"
2. Restart the system.
3. Once its back up, go to Settings, Docker, advanced view and then delete the image
4. Re-enable Docker
5. Go to Apps, Previous Apps, check off what you want, and you'll be back in business in a couple of minutes.

 

If that doesn't get you anywhere, then I'd say it's some sort of corruption on the cache drive preventing the image from forming correctly (I'm not a big fan of BTRFS for cache drives unless you are planning on running a cache pool - XFS anecdotally causes people far less problems

 

That sorted it thanks.

[docbrown]

Same thing has happened to me, I'm in the process of taking the above steps but I thought I'd submit my diagnostics as well in case it becomes a bigger issue for everyone.

 

EDIT: followed the steps, but no luck for me unfourtunatley, I'm still getting the same BTRFS error

 

Quote

Jul 4 21:18:39 trustno1 kernel: BTRFS error (device loop3): bad tree block start 0 50819923968
Jul 4 21:18:39 trustno1 kernel: BTRFS error (device loop3): bad tree block start 0 50819923968
Jul 4 21:18:39 trustno1 root: mount: /var/lib/docker: wrong fs type, bad option, bad superblock on /dev/loop3, missing codepage or helper program, or other error.

 

Any ideas? Any help is much appreciated.

 

trustno1-diagnostics-20180704-2110.zip

Edited July 4, 2018 by docbrown

[lionelhutz]

If you have a single cache drive then you have to re-format it and start over with it. If you have a cache pool then try the scrub on it.

[JorgeB]

It's enough to delete and recreate the docker image.

[docbrown]

I appreciated everyones help. So XFS is preferred when I reformat? Does it matter if all the other drives are BTRFS?

[JorgeB]

You don't need to reformat, just delete and recreate the docker image.

[docbrown]

8 minutes ago, johnnie.black said:

You don't need to reformat, just delete and recreate the docker image.

 

Sorry as I mentioned above in an edit I followed all the steps, one of which was deleting the image and then I still got the quoted error in red. 

[JorgeB]

1 minute ago, docbrown said:

 

Sorry as I mentioned above in an edit I followed all the steps, one of which was deleting the image and then I still got the quoted error in red. 

Thta's weird, don't know if reformatting will help with that, but you can try, as for btrfs or xfs if it's single device and you don't care for checksums and snapshots you can use xfs which is usually more resilient than btrfs and more easily fixable when there is corruption.

[docbrown]

2 minutes ago, johnnie.black said:

Thta's weird, don't know if reformatting will help with that, but you can try, as for btrfs or xfs if it's single device and you don't care for checksums and snapshots you can use xfs which is usually more resilient than btrfs and more easily fixable when there is corruption.

 

Thanks, I’ll give it another shot deleting it tonight if that doesn’t work I’ll then reformat. 

[JorgeB]

53 minutes ago, docbrown said:

 

Thanks, I’ll give it another shot deleting it tonight if that doesn’t work I’ll then reformat. 

Yes, make sure it really was deleted since it wasn't on the diags you posted, unless you rebooted after doing it, and there are no apparent issues with the cache filesystem.

[docbrown]

6 hours ago, johnnie.black said:

Yes, make sure it really was deleted since it wasn't on the diags you posted, unless you rebooted after doing it, and there are no apparent issues with the cache filesystem.

 

Thanks for your help, I went through the steps again and this time it worked. Not sure what I did wrong the first time. Thanks again.