Description of problem: Pull image failed when building Version-Release number of selected component (if applicable): OpenShift Master: v3.10.0-0.38.0 (online version 3.6.0.83) Kubernetes Master: v1.10.0+b81c8f8 OpenShift Web Console: v3.10.0-0.47.0 How reproducible: always Steps to Reproduce: 1.oc new project cindytest 2.oc new-app centos/ruby-22-centos7~https://github.com/openshift/ruby-ex.git 3.oc logs -f bc/ruby-ex Actual results: Cloning "https://github.com/openshift/ruby-ex.git" ... Commit: bbb670185b6ce67294cc461ae9c18710e6f26089 (Merge pull request #18 from durandom/master) Author: Ben Parees <bparees.github.com> Date: Thu Dec 7 14:53:36 2017 -0500 Pulling image "centos/ruby-22-centos7@sha256:58e87b94eefb5eb68846b2d66f2152dc7e2ea2ce247f8a6c2cc7649073a34537" ... /bin/bash: error while loading shared libraries: libtinfo.so.5: cannot open shared object file: Permission denied error: build error: non-zero (13) exit code from centos/ruby-22-centos7@sha256:58e87b94eefb5eb68846b2d66f2152dc7e2ea2ce247f8a6c2cc7649073a34537 Expected results: Build succeed
80% DevExp subteam cases are blocked by it.
I was able to reproduce this on cri-o scale group nodes. This might be a cri-o only problem, so I've made those nodes unschedulable.
This is the selinux issue which should have had already been fixed ~2 months ago. There's should be a chcon command run with ansible during install. I'll check why this hasn't run during the deploy I guess.
This looks like the chcon on the docker storage directory did not get invoked. Do we the logs from the ansible playbook run?
mwoodson would be best source for that information.
I have opened https://github.com/openshift/openshift-ansible/pull/8417 as a fix.
When tested in v3.10.0-0.47.0 (online version 3.6.0.83) , still met error as follow: Pulling image "registry.access.redhat.com/rhscl/php-71-rhel7@sha256:ac07d765077ec1cf670c36677e4308adcd1c3a527000bd316bf6731cc5477ee0" ... /bin/bash: error while loading shared libraries: libtinfo.so.5: cannot open shared object file: Permission denied ERROR: Error writing header for ".env.default": io: read/write on closed pipe ERROR: Error writing tar: io: read/write on closed pipe error: build error: write unix @->/var/run/docker.sock: write: broken pipe
Pods can run on cri-o nodes on 0.50.0, but QE is still seeing a problem using them for builds.
Opened https://github.com/openshift/openshift-ansible/pull/8519
Verified on free-int OpenShift Master: v3.10.0-0.66.0 Kubernetes Master: v1.10.0+b81c8f8 OpenShift Web Console: v3.10.0-0.66.0