Bug 1367173

Summary: RHEV+CFME+OSE deployment reported success but OpenShift install is broken
Product: Red Hat Quickstart Cloud Installer Reporter: Landon LaSmith <llasmith>
Component: Installation - OpenShiftAssignee: John Matthews <jmatthew>
Status: CLOSED DUPLICATE QA Contact: Sudhir Mallamprabhakara <smallamp>
Severity: high Docs Contact: Derek <dcadzow>
Priority: unspecified    
Version: 1.0CC: arubin, bthurber, dymurray, jmatthew, tsanders
Target Milestone: ---Keywords: Triaged
Target Release: 1.1   
Hardware: All   
OS: All   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-12-06 16:05:25 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Landon LaSmith 2016-08-15 19:04:05 UTC 
Description of problem: A deployment of RHEV + CFME + OSE reported success but the OSE hello world sample application is inaccessible and 'oc' commands fail to run with a message about missing a config until a successful 'oc login -u admin' command.  After 'oc login' running 'oc get pods' on the master node as the OSE user(admin) reports nothing unless the command is run with sudo which then reports failures for docker-registry-*-deploy

$ sudo oc get pods                                                                                                                                                                                                                            
NAME                         READY     STATUS             RESTARTS   AGE                                                                                                                                                                      
docker-registry-1-deploy     0/1       DeadlineExceeded   0          2d                                                                                                                                                                       
docker-registry-2-deploy     0/1       Error              0          2d                                                                                                                                                                       
management-metrics-1-9vlfx   1/1       Running            0          2d                                                                                                                                                                       
router-1-oj0bm               1/1       Running            0          2d


ISO Version: QCI-1.0-RHEL-7-20160812.t.0

How reproducible: First encounter

Steps to Reproduce:
1. Deploy RHEV + CFME + OSE (with hello world)
2. After deployment success try to access the OSE sample application which fails
3. ssh into the OSE master node and run 'oc get pods' which shows failures status for docker-register-1-deploy

Actual results: OSE sample application is inaccessible and 'oc' commands fail until 'oc login' is successful which will allow the user to query OSE

Expected results: Deployment will report failure if the OSE sample app is inaccessible or the OSE install fails on the master/worker nodes
Comment 4 Landon LaSmith 2016-08-15 19:35:24 UTC 
This was a fresh deployment on LibVirt.
Specs for VMs:
Satellite: 4 CPU, 16GB memory and 260GB disk (self hosted NFS)
RHV Engine: 2 CPU, 4GB memory and 30GB disk
RHV Hypervisor: 8 CPU, 32GB memory and 70GB disk
OSE: 1 Master, 1 Worker
 - Docker Storage per Worker: 30GB
 - CPU: 3
 - RAM: 16 GB
 - Disk: 75 GB

CFME was installed successfully and RHEV was added as provider
Comment 5 Landon LaSmith 2016-08-15 19:44:31 UTC 
# Summarized snapshot of the initial attempt to run 'oc whoami' when logged in as admin


$ oc whoami
error: No configuration file found, please login or point to an existing file:

  1. Via the command-line flag --config
  2. Via the KUBECONFIG environment variable
  3. In your home directory as ~/.kube/config

To view or setup config directly use the 'config' command.

$ oc login -u admin

Server [https://localhost:8443]: https://192.168.0.100:8443
The server uses a certificate signed by an unknown authority.
You can bypass the certificate check, but any data you send to the server could be intercepted by others.
Use insecure connections? (y/n): 
...
Login successful
Using project "helloworld"

$ oc whoami
admin

$ sudo oc get pods
NAME                         READY     STATUS             RESTARTS   AGE
docker-registry-1-deploy     0/1       DeadlineExceeded   0          2d
docker-registry-2-deploy     0/1       Error              0          2d
management-metrics-1-9vlfx   1/1       Running            0          2d
router-1-oj0bm               1/1       Running            0          2d
Comment 6 John Matthews 2016-08-17 18:18:22 UTC 
Moving this to post-ga because we have been unable to reproduce.

If we learn of details to reproduce we can revisit for GA.
Comment 8 Dylan Murray 2016-12-06 16:05:25 UTC 
Closing because I have not seen this reproduced and sample application errors are being kept in https://bugzilla.redhat.com/show_bug.cgi?id=1373843.

*** This bug has been marked as a duplicate of bug 1373843 ***