After weeks of being fine, the deployment stopped working and results in the message:
"Error from server (BadRequest): cannot trigger a deployment for "xxxx" because it contains unresolved images"This is an in-premises Openshift 3.5 cluster, the very same deployment works just fine from the web console. "oc get events" does not return anything, raising loglevel did not help me either. Can it be related to network setup? DNS, firewall - these are the only changes in the meantime I am aware of, but I would like to know how to investigate it from openshift perspective.
Late update - turned out that it was one of the nodes having connectivity problems with the registry. Incidentally when run from web console, the builder pod got assigned to a working node, when from cmd - to the failing one. Investigating the builds helped.