Created attachment 1498583 [details]
Cluster console Crashlooping Pod counter

Description of problem: 
Crashlooping Pod number in Cluster console doesn't display the real value. It displays "0" instead of the real pod number value in "CrashLoopBackOff" state.


Version-Release number of selected component (if applicable):
Openshift 3.11


How reproducible:
Create a new app, restart a pod may times so it becomes in Crahsloop state, and check Openshift Cluster Console. Instead of showing a Crashlooping value of "1", it always displays "0".


Steps to Reproduce:
1. Create a new project test:
oc new-project test
2. Create a new test app:
oc new-project testoc new-app https://github.com/openshift/sti-ruby.git --context-dir=2.0/test/puma-test-app
3. Get POD Container ID:
docker ps -a | grep ruby | grep ose-pod | grep Up
4. Kill POD Container ID with SIGTERM (I used SIGHUP):
docker kill --signal=SIGHUP CONTAINER-ID
5. Repeat 3 and 4 until POD status changes to "CrashLoopBackOff"
watch -n 5 "docker kill --signal=SIGHUP $(docker ps -a | grep ruby | grep ose-pod | grep Up | awk '{print $1}')"
oc get pods | grep Crash
6. Check Cluster Console (make sure Project is the new one, "test")


Actual results:
Crashlooping Pods number in Cluster Console remains as "0" instead of "1"


Expected results:
Crashlooping Pods number in Cluster Console should be "1"