The issue exist on ded-int-aws (OpenShift Master: v3.7.23, Kubernetes Master: v1.7.6+a08f5eeb62) too, thus cloning it: $ oc new-app centos/ruby-22-centos7~https://github.com/openshift/ruby-ex.git $ oc get pod NAME READY STATUS RESTARTS AGE ruby-ex-1-build 0/1 Completed 0 3m ruby-ex-2-8r6qb 1/1 Running 0 2m The deployment is triggered twice +++ This bug was initially created as a clone of Bug #1543317 +++ Description of problem: ... snipped... --- Additional comment from ge liu on 2018-03-07 10:31:51 CST --- Verified that this issue have been fixed in ded-stage-aws, thx
Stephanie, I presume this env needs the same fix as https://bugzilla.redhat.com/show_bug.cgi?id=1543317 1) the master must have the OPENSHIFT_DEFAULT_REGISTRY env var set or it must be set in the master-config via: imagePolicyConfig: internalRegistryHostname: docker-registry.default.svc.cluster.local:5000 2) the registry DC must have OPENSHIFT_DEFAULT_REGISTRY or REGISTRY_OPENSHIFT_SERVER_ADDR set (both to the same value) I'd suggest you go check what was done in ded-stage-aws and make ded-int-aws match it.
Reproduce this bug on free-int
Fixed in ded-int-aws and free-int.
Verified in ded-int-aws (server v3.7.23). Now deployment is only triggered once. In free-int, bug 1579107 still exists. Will verify this bug when that unblocked
Noticed reporter of bug 1579107 communicated with Dev in online and made it addressed. Now verified this bug in free-int; deployment is only triggered once. OpenShift Master: v3.10.0-0.50.0 (online version 3.6.0.83) Kubernetes Master: v1.10.0+b81c8f8 OpenShift Web Console: v3.10.0-0.50.0 Thank you