AUFS storage driver results in zombie processes #412
Comments
|
This was addressed in the MMW project with WikiWatershed/model-my-watershed@6fb93e6. |
|
Ok. Any reason why the AUFS dependencies install in the base ansible-docker role shouldn't be handling the kernel backport? |
|
I think because it was only occurring on that instance of Docker/AUFS. There are a few other setups with that combo that are working fine. My hope is that the patched kernel will be available soon. For DRIVER, adding the backported kernel install to the top level playbook may be the best bet for now. Once I see the patch kernel land, I'll open PRs against DRIVER and MMW to remove it. |
|
Ok, I'll go with a kernel backport on this project then. |
flibbertigibbet
added a commit
to flibbertigibbet/DRIVER
that referenced
this issue
Feb 22, 2016
flibbertigibbet
added a commit
to flibbertigibbet/DRIVER
that referenced
this issue
Feb 23, 2016
hectcastro
pushed a commit
that referenced
this issue
Feb 23, 2016
This reverts 17be608. The kernel version that fixes the AUFS bug referenced in that commit is in the latest Ubuntu release AMI (3.13.0-79). In addition, the Docker daemon have been upgraded to 1.10.x. See also: - https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1533043 - https://bugzilla.kernel.org/show_bug.cgi?id=109971
hectcastro
added a commit
that referenced
this issue
Feb 23, 2016
…kported-kernel Revert "Backport kernel; fixes #412"; upgrade Docker
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
See moby/moby#18180.
Java processes in Debian container hang seemingly at random, pegging CPU; container cannot be stopped. Issue only appeared when running container within vagrant, mounting volumes shared via NFS.
Switching storage driver to
devicemapperfixes the issue; however,devicemapperalso apparently causes file system corruption and nasal demons.The text was updated successfully, but these errors were encountered: