Hello Cory- What does `journalctl -u docker` say?
My hunch is that `/var/lib/docker` got corrupted in some way on this host. There are several issues over several releases of Docker that can cause this to happen and the issues are itermittent or racy. If you don't care about any of the volumes or data in your containers I would suggest stopping docker (`systemctl stop docker.service docker.socket`), remove all of the files (`rm -Rf /var/lib/docker/*`), and restart the machine. Hope that helps. Thank You, Brandon On Sun, Aug 7, 2016 at 1:23 AM Cory Keane <[email protected]> wrote: > Hello. > > I am experiencing an issue whenever trying to use the "docker run" command > on one of my Fleet CoreOS machines. It is only happening to this specific > CoreOS machine in the cluster, no other ones. The only difference between > this CoreOS machine and the others is that it seems to have automatically > updated its CoreOS version. I am thinking this might be related. > > The issue is: Whenever I try and use the "docker run" command, no matter > how I format it (I've tried everything, and am positive that I am not using > the command incorrectly), the command fails and I get the following message > "docker: Error response from daemon: Error relabeling upper directory: > operation not supported.". > > I cannot find anything on the Internet with this exact error message. I > was thinking I could try and revert the CoreOS version for this machine, > but no matter which release channel I try (stable, beta, or alpha), I still > get the same issue. Is there a way to revert to a specific version of > CoreOS? All my other CoreOS machines are on a version below the current > stable release. Ideally, I'd like to figure out the issue and be able to > use this current stable release for all our CoreOS machines. > > Thanks! >
