Could you please post the node log file with more context? I'd like to see the following, if possible: - Node service startup before it was restarted. This will be sometime before Feb 06 23:56:10 (beginning of attached log file) - Node service startup after it was restarted - Log lines around the time the pod actually started (post node service restart)
Description of problem: Without any reason we are getting multiple pods in pending state and the state never changes. The only solution to solve the problem was to restart the service atomic-openshift-node on the node. It was like something on the node was blocked but we don't know what and we consider that if it was really something blocked a timeout should be implemented to avoid unlimited "pending" state on a node (other nodes were available). The solution was to restart the atomic-openshift-node service. Version-Release number of selected component (if applicable): 3.1.0.4 How reproducible: At certain times, we don't know how to reproduce this problem Steps to Reproduce: 1. At certain times, we don't know how to reproduce this problem 2. 3. Actual results: The pods shall run fine Expected results: The pods remain in pending state