Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Failing tests

See original GitHub issue

What happened?

TASK [ansible-node-exporter : Create node_exporter config directory] ***********
    fatal: [centos8]: UNREACHABLE! => {"changed": false, "msg": "Failed to create temporary directory.In some cases, you may have been able to authenticate and did not have permissions on the target directory. Consider changing the remote tmp path in ansible.cfg to a path rooted in \"/tmp\", for more error information use -vvv. Failed command was: ( umask 77 && mkdir -p \"` echo OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused \"process_linux.go:90: adding pid 28027 to cgroups caused \\\"failed to write 28027 to cgroup.procs: write /sys/fs/cgroup/cpu,cpuacct/docker/6bffe942ab289f6018c910fcb7b67957333e050d435ffbe3bbfedb72effead8d/cgroup.procs: invalid argument\\\"\": unknown/.ansible/tmp `\"&& mkdir OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused \"process_linux.go:90: adding pid 28027 to cgroups caused \\\"failed to write 28027 to cgroup.procs: write /sys/fs/cgroup/cpu,cpuacct/docker/6bffe942ab289f6018c910fcb7b67957333e050d435ffbe3bbfedb72effead8d/cgroup.procs: invalid argument\\\"\": unknown/.ansible/tmp/ansible-tmp-1595086095.7307746-27669-8276321652003 && echo ansible-tmp-1595086095.7307746-27669-8276321652003=\"` echo OCI runtime exec failed: exec failed: container_linux.go:348: starting container process caused \"process_linux.go:90: adding pid 28027 to cgroups caused \\\"failed to write 28027 to cgroup.procs: write /sys/fs/cgroup/cpu,cpuacct/docker/6bffe942ab289f6018c910fcb7b67957333e050d435ffbe3bbfedb72effead8d/cgroup.procs: invalid argument\\\"\": unknown/.ansible/tmp/ansible-tmp-1595086095.7307746-27669-8276321652003 `\" ), exited with result 1", "unreachable": true}

Did you expect to see some different?

A working test.

Maybe fix?

I think I had similar issues with Github Actions. Could be multiple things:

I remember that essentially the Docker in Docker failed because of the filesystem. You can’t have overlay on overlay. One of them has to be vfs for example.
Or it’s buggy Docker?
Or it’s Travis-CI.

Should we try to fix this?