gate tripleo-ci-centos-7-containers-multinode fails to update packages in cron container

[Gabriele Cerami]

logs for the gate at

http://logs.openstack.org/61/560261/9/gate/tripleo-ci-centos-7-containers-multinode/2884516/logs/undercloud/home/zuul/overcloud_prep_containers.log.txt.gz#_2018-04-27_09_33_58

show that the update step fails during a simple rpm query in the cron container, with this error

2018-04-27 09:33:58 | container_linux.go:247: starting container process caused "process_linux.go:258: applying cgroup configuration for process caused \"open /sys/fs/cgroup/cpuset/system.slice/cpuset.cpus: no such file or directory\""

Just before this error, it seemed like cron container had to be downloaded again as it was not downloaded properly.

[Bogdan Dobrelya (bogdando)]

This prolly related https://github.com/openshift/origin/issues/16246#issuecomment-355852817

[Bogdan Dobrelya (bogdando)]

Also seems related https://bugzilla.redhat.com/show_bug.cgi?id=1543575
We need docker-1.13.1-52.gitce62987.el7_4 prolly

[Gabriele Cerami (gcerami)]

Thanks bogdan also for this link https://bugzilla.redhat.com/show_bug.cgi?id=1523043. So it may be a race condition, that's why we're not seeing it happening in all the gates.
Suggestion: update the docker version to not make this happen again

[Matt Young (halcyondude)]

(tripleo-ci bug triage)

- This requires centos 7.5

[Alex Schultz (alex-schultz)]

We've now got 7.5 everywhere. Closing this bug out. If it's still a problem or happens again feel free to reopen.