New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Random "Cannot start container" Errors on 1.9.0-rc5 CentOS7 #17653
Comments
I do see the same here with 1.9.0 on CentOS 7:
Unfortunately random, not really reproducible.
|
Added this to the 1.9.1 milestone so we don't loose sight. ping @LK4D4 care to have a look? |
Quick update: I've seen that now also during "docker build":
Some config / setup as written in my previous comment. |
this maybe something related to centos because i cannot repo on my systemd based ubuntu system. |
@crosbymichael pretty sure, it seems to be a timing issue, as repeating the same command (docker build / run) after such a failure mostly succeeds then... |
ya, i'll have to setup a centos box and try to reproduce this |
Let me know if I can support in any way |
@crosbymichael Assigning this to you, please let us know how the attempts to reproduce go. |
Same here:
On a Centos machine. Happens occasionally.
|
Was happening randomly on Docker version 1.9.0, build 76d6bc9 running Centos 7.1, 3.10.0-229.20.1.el7.x86_64. Error response from daemon: Cannot start container 54fafc9cdea4167c6be9fd5162e5db7ec76a58be86993b1b5ce5086be4f89a40: [8] System error: write /sys/fs/cgroup/devices/system.slice/docker-54fafc9cdea4167c6be9fd5162e5db7ec76a58be86993b1b5ce5086be4f89a40.scope/cgroup.procs: no such device Rolled back to 1.8.3 and can't repro. |
same here:
docker:
running on CentOS 7.1:
|
ok, i can reproduce. looking for the cause |
@LK4D4 can you help me with this. cannot find anything yet that would cause this on systemd only |
@crosbymichael how you reproduced it? |
So far I can reproduce but cannot find the root cause. A workaround for everyone is to launch the daemon with
Sorry about not finding a better fix, i'll keep looking for clues. However, if you make this change, i have been running tests for a day now and there are zero errors. |
I'm running into the same issue with
@crosbymichael 's workaround "resolves" the issue |
I'll echo having the same problem since yesterday, when I upgraded to Docker 1.9. Containers randomly fail to start, but usually succeed on the first or second retry. The workaround suggested by @crosbymichael seems to "resolve" the issue for me as well.
|
Log for ref:
|
@crosbymichael Thanks, the |
Wondering if this is fixed for all following this by changing to |
@thaJeztah changing to |
Adding "--exec-opt native.cgroupdriver=cgroupfs" to my Docker unit file worked for me, too. Am running CentOS 7. |
I was hitting this with docker 1.9.1 on RHEL 7.2 but the "--exec-opt native.cgroupdriver=cgroupfs" option has fixed it for me |
@woshihaoren does using |
@thaJeztah No,just docker rm it, and create again |
@thaJeztah I use cgroupfs , create 10 times, not see this problem. Looks good, Thank you. |
Ran into this issue with centos. I edited the unit file and problem was fixed. This was with version |
Encountered: System error: open /sys/fs/cgroup/devices/system.slice/docker [...].scope/cgroup.pcros: no such file or directory with docker 1.9.1 on 3.10.0-327.4.5.el7.x86_64 I will try the workaround. |
Just a note for others: Trying to edit the unit file for this workaround I got the error: docker: "daemon" requires 0 arguments when trying to start the service, but that is due to options now requiring an "=" sign. Seems to work so far. |
ping @LK4D4 @crosbymichael is this resolved with the switching to cgroupfs as default in 1.10? #17704 |
We received these messages occasionally as well:
After upgrading to the latest available (docker-engine-1.10.1) our issues seems to have dissapeared. From the release notes:
So this seems resolved at least on rhel7 using docker-engine-1.10.1-1.el7 |
I'm seeing this error message on CoreOS 983.0.0 with Docker 1.10.2, where
Setting
edit This was an issue with a bad |
Also see this on docker 1.8.3 using Kubernetes.
Error message:
|
can you try using |
Hey y'all. ● docker.service - Docker Application Container Engine Sep 16 13:51:06 COMPNAME systemd[1]: Starting Docker Application Container Engine... Any help would be appreciated. |
@WigiPedia |
Issue still happens on CentOS based image (AMI) even with
In next attempt docker can start without any issues. Issue can be observed in ~1% of all executions.
@crosbymichael, I think to create PR with retry-mechanism for docker run. Does such PR have any chance to be accepted? |
Let me close this ticket for now, as it looks like it went stale. |
Since upgrading to 1.9.0 RCs on a CentOS7 host (3.10.0-229.14.1.el7.x86_64), I've been seeing random Cannot start container errors, e.g.:
Relevant log entries are below for reference:
I'm able to reproduce this fairly reliably on CentOS7 using the technique mentioned in #17387:
This issue does not occur with Docker 1.8.3 installed, nor can I reproduce it with 1.9.0-rc5 on a system without systemd using an AUFS storage backend (Ubuntu 14.04).
Bug Report Information
The text was updated successfully, but these errors were encountered: