Periodic failed to get CPU consumption and request: metrics obtained for 0/X of pods logs in OpenShift

Issue

• Periodic failed to get CPU consumption and request: metrics obtained for 0/X of pods logs in master
• The heapster returns empty metrics like "metrics": []
• This log appears for 2 min when new pod is deployed
• This log appears in master log every 2.5 min * n intervals (5.0 min, 7.5 min etc.), the heapster returns false empry metrics like "metrics": [] for long-running pods