Description of problem: I had done a deployment of RHV(self hosted) + OSE, and when it finished it displayed the installation summary. The link for the example application was not accessible from the Satellite system. All the addresses displayed in the URL's were resolvable and pingable, however when you would go to the example application URL you would get: Application is not available Here is an example of using curl to go to the link from the satellite server: [root@b ~]# curl http://hello-openshift.bubba.b.b <html> <head> <meta name="viewport" content="width=device-width, initial-scale=1"> <style type="text/css"> /*! * Bootstrap v3.3.5 (http://getbootstrap.com) * Copyright 2011-2015 Twitter, Inc. * Licensed under MIT (https://github.com/twbs/bootstrap/blob/master/LICENSE) */ /*! normalize.css v3.0.3 | MIT License | github.com/necolas/normalize.css */ html { font-family: sans-serif; -ms-text-size-adjust: 100%; -webkit-text-size-adjust: 100%; } body { margin: 0; } h1 { font-size: 1.9em; font-weight: 400; line-height: 1.3; margin: 0.68em 0; } @media (min-width: 768px) { h1 { font-size: 2.15em;} } * { -webkit-box-sizing: border-box; -moz-box-sizing: border-box; box-sizing: border-box; } *:before, *:after { -webkit-box-sizing: border-box; -moz-box-sizing: border-box; box-sizing: border-box; } html { -webkit-tap-highlight-color: rgba(0, 0, 0, 0); } body { font-family: "Helvetica Neue", Helvetica, Arial, sans-serif; line-height: 1.66666667; font-size: 13px; color: #333333; background-color: #ffffff; margin: 4em 3em; } p { margin: 0 0 10px; font-size: 13px; font-weight: 300; line-height: 1.4; } @media (min-width: 768px) { p { font-size: 14px; } } </style> </head> <body> <div> <h1>Application is not available</h1> <p>The application is currently not serving requests at this endpoint. It may not have been started or is still starting.</p> </div> </body> </html> Version-Release number of selected component (if applicable): QCI-1.0-RHEL-7-20160809.t.0 How reproducible: only tried once. Steps to Reproduce: 1. Do a RHV + OSE deployment. 2. Start a browser on the satellite host. 3. Attempt to go to the example application URL using the browser running on the satellite system. Actual results: "Application is not available" is displayed. Expected results: The output of the example application.
https://github.com/fusor/fusor/pull/1146 Problem was due to the subdomain not being replaced for the sample app.
Expected to be in QCI-1.0-RHEL-7-20160815.t.0-QCI-x86_64-dvd1.iso
Verified against QCI-1.0-RHEL-7-20160816.t.0
This is broken again. In QCI-1.0-RHEL-7-20160902.5 if you click on the link for the example application it says: 503 Service Unavailable No Server is available to handle this request. This happens running a browser from the satellite. It's been happening consistently, so there is no need for a log...just do an OSE deployment and you will get this behavior.
James, The initial issue which caused this bug was due to the subdomain changing and the route being invalid. If you get a 503 it is generally not an issue with the route but rather something in the post-installation process did not go properly. I cannot reproduce this 100 percent but it would help to SSH into the master host and run 'oc get pods' to see what is and is not running. Sometimes the usual suspect is the docker-registry pod which you can access by running 'oc login -u system:admin; oc project default; oc logs bc/docker-registry'
Had a successful deployment using the latest compose on 9/8. I spoke with James and determined its possible that he was trying to access the sample app too soon after the deployment when OSE had not finished building and deploying the sample application. After a successful deployment the sample app takes 5-10 minutes to build the image. We currently have no way to monitor that application without polling, so we plan to address this issue in the future and will keep this bug open to track it. Relevant Trello card: https://trello.com/c/z01lJplD
I have definitely had deployments that even after the 5-10 minute period the sample app is still not accessible from the summary provided link. This is not to say its never accessible, but only that it is not always so.
James, In the scenario where the sample app is not accessible it would help to get the logs from the build process of the application to determine what went wrong. If you come across another deployment where the sample application is not deployed please run 'oc logs bc/php-helloworld' and another log that is helpful would be to run 'oc login -u system:admin; oc project default; oc logs bc/docker-registry'
Marking as a duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=1373843 due to the 3.3 upgrade causing instability with the sample app. Keeping all issues with the sample app in one bug. *** This bug has been marked as a duplicate of bug 1373843 ***