Bug 1577959

Description of problem:

when installation task "[openshift_hosted : Poll for OpenShift pod deployment success]" is executed it tries to deploy router pods in all infra region nodes and during this time it also tries to deploy router pods in master nodes, in which by default "infra" region is selected, but as master servers are set to unschedulable these task fails.

Error message in verbose -
~~~
"_raw_params": "oc get replicationcontroller router-20 --namespace default --config /etc/origin/master/admin.kubeconfig -o jsonpath='{ .metadata.annotations.openshift\\.io/deployment\\.phase }'

                "images": "openshift3/ose-${component}:${version}", 
                "name": "router", 
                "namespace": "default", 
                "ports": [
                    "80:80", 
                    "443:443"
                ], 
                "replicas": "5", 
                "selector": "region=infra", 
                "serviceaccount": "router", 
                "stats_port": 1936
~~~

How reproducible: 
Steps to Reproduce:

1. Setup multi master installation with three master and three nodes, assign "infra" region to two nodes in inventory file.
2. Installer will try to deploy router pods on 5 nodes, and will finish with below message.
~~~
Failure summary:


  1. Hosts:    redhat.example.com
     Play:     Create Hosted Resources - router
     Task:     Poll for OpenShift pod deployment success
     Message:  All items completed
~~~

Actual results:
Please include the entire output from the last TASK line through the end of output if an error is generated

Get below error message for task - "[openshift_hosted : Poll for OpenShift pod deployment success]".
~~~ 
"_raw_params": "oc get replicationcontroller router-20 --namespace default --config /etc/origin/master/admin.kubeconfig -o jsonpath='{ .metadata.annotations.openshift\\.io/deployment\\.phase }'

                "images": "openshift3/ose-${component}:${version}", 
                "name": "router", 
                "namespace": "default", 
                "ports": [
                    "80:80", 
                    "443:443"
                ], 
                "replicas": "5", 
                "selector": "region=infra", 
                "serviceaccount": "router", 
                "stats_port": 1936
~~~

Expected results: Installation should not try to deploy router pods on master servers, as they are by default set to "unschedulable" due to which router pod deployment fails.

Additional info: We can use variable "openshift_router_selector='region=primary'" as workaround in this case, router pod will only get deployed on node with primary region and installer will not try to deploy them on master nodes.


Please attach logs from ansible-playbook with the -vvv flag


Description of problem:

Version-Release number of the following components:
rpm -q openshift-ansible
rpm -q ansible
ansible --version

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:
Please include the entire output from the last TASK line through the end of output if an error is generated

Expected results:

Additional info:
Please attach logs from ansible-playbook with the -vvv flag


Description of problem:

Version-Release number of the following components:
rpm -q openshift-ansible
rpm -q ansible
ansible --version

How reproducible:

Steps to Reproduce:
1.
2.
3.

Actual results:
Please include the entire output from the last TASK line through the end of output if an error is generated

Expected results:

Additional info:
Please attach logs from ansible-playbook with the -vvv flag