Description of problem: When use `oc cluter up` command failed with 400 error. Version-Release number of selected component (if applicable): oc v3.11.0-0.32.0 kubernetes v1.11.0+d4cacc0 How reproducible: always Steps to Reproduce: 1. Use command `oc cluster up --image='brew-pulp-docker01.web.prod.ext.phx2.redhat.com:8888/openshift3/ose-${component}:${version}' --loglevel=6` Actual results: 1. Failed with 400 error. Expected results: 1. Succeed without any error. Additional info: I0910 04:58:30.852011 24841 web_console_operator.go:65] polling for web-console availability ... I0910 04:58:30.885121 24841 round_trippers.go:405] GET https://127.0.0.1:8443/apis/apps/v1/namespaces/openshift-web-console/deployments/webconsole 200 OK in 32 milliseconds I0910 04:58:31.851986 24841 web_console_operator.go:65] polling for web-console availability ... I0910 04:58:31.883109 24841 round_trippers.go:405] GET https://127.0.0.1:8443/apis/apps/v1/namespaces/openshift-web-console/deployments/webconsole 200 OK in 30 milliseconds I0910 04:58:31.933095 24841 round_trippers.go:405] GET https://127.0.0.1:8443/apis/webconsole.operator.openshift.io/v1alpha1/openshiftwebconsoleconfigs/instance 200 OK in 47 milliseconds I0910 04:58:32.025133 24841 round_trippers.go:405] PUT https://127.0.0.1:8443/apis/webconsole.operator.openshift.io/v1alpha1/openshiftwebconsoleconfigs/instance 200 OK in 89 milliseconds I0910 04:58:32.025368 24841 interface.go:41] Finished installing "openshift-image-registry" "rhel-imagestreams" "sample-templates" "persistent-volumes" "openshift-router" "openshift-web-console-operator" I0910 04:58:33.047074 24841 round_trippers.go:405] GET https://127.0.0.1:8443/apis/user.openshift.io/v1/users/developer 503 Service Unavailable in 1020 milliseconds W0910 04:58:33.073131 24841 up.go:1064] error checking user: the server is currently unable to handle the request (get users.user.openshift.io developer) Login to server ... I0910 04:58:33.073863 24841 loader.go:359] Config loaded from file /root/openshift.local.clusterup/kube-apiserver/admin.kubeconfig I0910 04:58:33.123130 24841 round_trippers.go:405] HEAD https://127.0.0.1:8443/ 403 Forbidden in 48 milliseconds I0910 04:58:33.133072 24841 round_trippers.go:405] GET https://127.0.0.1:8443/.well-known/oauth-authorization-server 200 OK in 9 milliseconds I0910 04:58:33.152107 24841 round_trippers.go:405] GET https://127.0.0.1:8443/oauth/authorize?client_id=openshift-challenging-client&code_challenge=TMlgnBuN3wYrrHYPTyil32QQ5icTeYoyf1Vy4mAnabw&code_challenge_method=S256&redirect_uri=https%3A%2F%2F127.0.0.1%3A8443%2Foauth%2Ftoken%2Fimplicit&response_type=code 400 Bad Request in 18 milliseconds
After delete the exist images, then `oc cluster up` succeed.
@mfojtik as I recall this behavior is intentional. If you specify exactly which images to use, we assume that you want something in particular and we do *not* repull them. We do the same thing when --tag is specified. Choosing exactly which images to use is a very advanced usage.
Yes, if the tag is not latest, we do IfNotExists pull policy, I would close this as not a bug.