There should be more segregated networks within that cluster.
This could be achieved by creating additional subnets for each layer e.g. 1 for masters, 1 for nodes etc.

Need to update the ansible hosts template to take into account if deploy_prometheus is set to false.

The grafana svc pod does get deployed successfully, but the following error is thrown:

TASK [openshift_grafana : Wait for grafana service] *****************************************************
Monday 01 October 2018 17:00:27 -0400 (0:00:00.181) 2:06:55.901 ********
fatal: [ocp-master-1]: FAILED! => {"msg": "The conditional check 'result.status == 200' failed. The error was: error while evaluating conditional (result.status == 200): 'dict object' has no attribute 'status'"}

vars.yml:

http://pastebin.test.redhat.com/653780

===============
Setup Info

I am using the v3.10 branch without CNS i.e all the variables in vars realted to cns are commented out except 'deploy_cns', 'deploy_cns_on_infra' which are set to false.

deploy_cns: false
deploy_cns_on_infra: false

========================
Observation related to failure

 What i see is this failed on ocp-app-1, ocp-app-2, ocp-app3 basically the app nodes. 
 It's trying to find hosts under : 
      (( inventory_hostname in groups.masters ) or 
       ( inventory_hostname in groups.infra ) or 
       ( inventory_hostname in groups.glusterfs )):

Since we don't have cns, we don't have glusterfs section in the hosts file. There is no check to find in the groups.nodes.

The app nodes doesn't exists under any of these groups in the hosts file. (Partial output from hosts file on Bastion node)

[masters]
ocp-master-1
ocp-master-2
ocp-master-3

[infra]
ocp-infra-1
ocp-infra-2
ocp-infra-3

[etcd]
ocp-master-1
ocp-master-2
ocp-master-3

[nodes]
ocp-master-1 openshift_node_group_name='node-config-master' openshift_schedulable=true
ocp-master-2 openshift_node_group_name='node-config-master' openshift_schedulable=true
ocp-master-3 openshift_node_group_name='node-config-master' openshift_schedulable=true
ocp-infra-1 openshift_node_group_name='node-config-infra'
ocp-infra-2 openshift_node_group_name='node-config-infra'
ocp-infra-3 openshift_node_group_name='node-config-infra'
ocp-app-1 openshift_node_group_name='node-config-compute'
ocp-app-2 openshift_node_group_name='node-config-compute'
ocp-app-3 openshift_node_group_name='node-config-compute'

====================
Output from ansible.log:

2018-10-04 14:53:13,591 p=11352 u=root | TASK [Find all hostnames for bootstrapping] *********************************************************************
2018-10-04 14:53:13,591 p=11352 u=root | Thursday 04 October 2018 14:53:13 -0400 (0:00:00.376) 1:34:47.856 ******
2018-10-04 14:53:14,468 p=11352 u=root | fatal: [ocp-master-1]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'openshift'\n\nThe error appears to have been in '/usr/share/ansible/openshift-ansible/playbooks/openshift-node/private/join.yml': line 32, column 5, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n - name: Find all hostnames for bootstrapping\n ^ here\n"}
2018-10-04 14:53:14,470 p=11352 u=root | to retry, use: --limit @/root/ansible-ocp-azure-3_10/ansible-ocp-azure/deploy.retry

2018-10-04 14:53:14,470 p=11352 u=root | PLAY RECAP ******************************************************************************************************
2018-10-04 14:53:14,470 p=11352 u=root | localhost : ok=76 changed=35 unreachable=0 failed=0
2018-10-04 14:53:14,470 p=11352 u=root | ocp-app-1 : ok=8 changed=2 unreachable=0 failed=1
2018-10-04 14:53:14,471 p=11352 u=root | ocp-app-2 : ok=8 changed=2 unreachable=0 failed=1
2018-10-04 14:53:14,471 p=11352 u=root | ocp-app-3 : ok=8 changed=2 unreachable=0 failed=1
2018-10-04 14:53:14,471 p=11352 u=root | ocp-infra-1 : ok=201 changed=92 unreachable=0 failed=0
2018-10-04 14:53:14,471 p=11352 u=root | ocp-infra-2 : ok=201 changed=92 unreachable=0 failed=0
2018-10-04 14:53:14,471 p=11352 u=root | ocp-infra-3 : ok=201 changed=92 unreachable=0 failed=0
2018-10-04 14:53:14,471 p=11352 u=root | ocp-master-1 : ok=580 changed=249 unreachable=0 failed=1
2018-10-04 14:53:14,471 p=11352 u=root | ocp-master-2 : ok=393 changed=177 unreachable=0 failed=0
2018-10-04 14:53:14,471 p=11352 u=root | ocp-master-3 : ok=393 changed=177 unreachable=0 failed=0
2018-10-04 14:53:14,472 p=11352 u=root | INSTALLER STATUS ************************************************************************************************
2018-10-04 14:53:14,475 p=11352 u=root | Initialization : Complete (0:01:12)
2018-10-04 14:53:14,475 p=11352 u=root | Health Check : Complete (0:05:09)
2018-10-04 14:53:14,476 p=11352 u=root | Node Bootstrap Preparation : Complete (0:32:19)
2018-10-04 14:53:14,476 p=11352 u=root | etcd Install : Complete (0:01:46)
2018-10-04 14:53:14,476 p=11352 u=root | Master Install : Complete (0:07:18)
2018-10-04 14:53:14,476 p=11352 u=root | Master Additional Install : Complete (0:00:56)
2018-10-04 14:53:14,476 p=11352 u=root | Node Join : In Progress (0:00:11)
2018-10-04 14:53:14,477 p=11352 u=root | This phase can be restarted by running: playbooks/openshift-node/join.yml
2018-10-04 14:53:14,477 p=11352 u=root | Thursday 04 October 2018 14:53:14 -0400 (0:00:00.885) 1:34:48.742 ******
2018-10-04 14:53:14,477 p=11352 u=root | ===============================================================================
2018-10-04 14:53:14,481 p=11352 u=root | openshift_node : install needed rpm(s) ---------------------------------------------------------------- 1009.21s
2018-10-04 14:53:14,481 p=11352 u=root | azure_infra : Azure | Bastion install required packages ------------------------------------------------ 515.06s
2018-10-04 14:53:14,481 p=11352 u=root | Ensure openshift-ansible installer package deps are installed ------------------------------------------ 307.61s
2018-10-04 14:53:14,481 p=11352 u=root | Run health checks (install) - EL ----------------------------------------------------------------------- 305.37s
2018-10-04 14:53:14,481 p=11352 u=root | container_runtime : Install Docker --------------------------------------------------------------------- 283.81s
2018-10-04 14:53:14,481 p=11352 u=root | openshift_node : Install iSCSI storage plugin dependencies --------------------------------------------- 247.06s
2018-10-04 14:53:14,481 p=11352 u=root | azure_infra : Azure | Wait for master VM creation ------------------------------------------------------ 189.86s
2018-10-04 14:53:14,481 p=11352 u=root | openshift_node : Install node, clients, and conntrack packages ----------------------------------------- 141.25s
2018-10-04 14:53:14,481 p=11352 u=root | openshift_node : Check status of node image pre-pull --------------------------------------------------- 122.40s
2018-10-04 14:53:14,482 p=11352 u=root | azure_infra : Azure | Manage Node Nics ---------------------------------------------------------------- 110.43s
2018-10-04 14:53:14,482 p=11352 u=root | azure_infra : Azure | Manage Infra Nics ---------------------------------------------------------------- 106.71s
2018-10-04 14:53:14,482 p=11352 u=root | azure_infra : Azure | Manage Master Nics --------------------------------------------------------------- 106.36s
2018-10-04 14:53:14,482 p=11352 u=root | os_firewall : Install iptables packages ----------------------------------------------------------------- 95.19s
2018-10-04 14:53:14,482 p=11352 u=root | openshift_node : Install dnsmasq ------------------------------------------------------------------------ 83.10s
2018-10-04 14:53:14,482 p=11352 u=root | ocp_pre : Azure | Install OCP client -------------------------------------------------------------------- 78.89s
2018-10-04 14:53:14,482 p=11352 u=root | openshift_node : Install GlusterFS storage plugin dependencies ------------------------------------------ 77.69s
2018-10-04 14:53:14,482 p=11352 u=root | openshift_node : Install Ceph storage plugin dependencies ----------------------------------------------- 70.74s
2018-10-04 14:53:14,482 p=11352 u=root | azure_infra : Azure | Bastion disable all repos --------------------------------------------------------- 52.86s
2018-10-04 14:53:14,482 p=11352 u=root | openshift_control_plane : Wait for all control plane pods to become ready ------------------------------- 52.69s
2018-10-04 14:53:14,483 p=11352 u=root | openshift_excluder : Install docker excluder - yum ------------------------------------------------------ 51.51s
2018-10-04 14:53:14,483 p=11352 u=root | Playbook run took 0 days, 1 hours, 34 minutes, 48 seconds
2018-10-04 14:53:14,483 p=11352 u=root | Failure summary:

1. Hosts: ocp-app-1, ocp-app-2, ocp-app-3
   Play: nodes
   Task: Azure | RHSM register (username-passwd) Master-Infra-CNS
   Message: The conditional check '( inventory_hostname in groups.masters ) or ( inventory_hostname in groups.infra ) or ( inventory_hostname in groups.glusterfs )' failed. The error was: error while evaluating conditional (( inventory_hostname in groups.masters ) or ( inventory_hostname in groups.infra ) or ( inventory_hostname in groups.glusterfs )): Unable to look up a name or access an attribute in template string ({% if ( inventory_hostname in groups.masters ) or ( inventory_hostname in groups.infra ) or ( inventory_hostname in groups.glusterfs ) %} True {% else %} False {% endif %}).
   Make sure your variable name does not contain invalid characters like '-': argument of type 'StrictUndefined' is not iterable

         The error appears to have been in '/root/ansible-ocp-azure-3_10/ansible-ocp-azure/roles/ocp_pre/tasks/subscribe.yml': line 36, column 3, but may
         be elsewhere in the file depending on the exact syntax problem.
         
         The offending line appears to be:
         
         
         - name: Azure | RHSM register (username-passwd) Master-Infra-CNS
           ^ here
   

2. Hosts: ocp-master-1
   Play: Approve any pending CSR requests from inventory nodes
   Task: Find all hostnames for bootstrapping
   Message: The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'openshift'

         The error appears to have been in '/usr/share/ansible/openshift-ansible/playbooks/openshift-node/private/join.yml': line 32, column 5, but may
         be elsewhere in the file depending on the exact syntax problem.
         
         The offending line appears to be:
         
         
           - name: Find all hostnames for bootstrapping
             ^ here
   

CNS should be running on Infra nodes to cut down resource costs in Azure

Components should be split to their own resource groups based on their lifecycle.

I have fixed the indentation on your recent commit, moving the lines to the left

Need the ability to create custom RHEL images and upload them to Azure so user isn't double billed for RH Subscriptions.