I am trying to setup Openshift-origin, using the oo_deployment_guide_comprehensive.html guide, but am coming across an connection error, when my broker server tries to communicate with the application node.
This shows itself, when I am on the console via a browser, and try and click on either My Applications or Create Application (though from the logs, it is redirecting My Applications to Create Application, as I have no apps yet). The request hangs for a time, probably hitting a timeout value and returns logs like the following:
==> /var/log/openshift/console/production.log <== 2013-10-04 13:49:14.278 [ERROR] Unhandled exception reference #0a93348d85376c140d35a0922525acc4: Failed. Response code = 500. Response message = .
==> /var/log/openshift/broker/user_action.log <== RESULT=FAILURE STATUS=internal_server_error TIMESTAMP=1380890954 DATE=2013-10-04 TIME=13:49:14 ACTION="" REQ_ID=b16209fa462e1c5dac8660eebfc95442 USER_ID=524c1bc0a0908ee6d4000001 LOGIN=testuser DOMAIN= Unable to complete the requested operation due to: Node execution failure (error getting result from node)..
However when I run an mco ping, from the broker server, I get a successful response back from the application node.
I have added a test district to the cluster, and moved the app node into that district. When viewing this district in the admin_console, it shows the node is a member of the district however it says:
1 nodes failed to respond The nodes may be down, or may not have responded within the configured timeout.
I have tried reinstalling both the broker and application nodes from scratch twice, but hit the same problem, so either I am misconfiguring something the same way each time, or there is a bug I am unfortunately hitting.
Has anyone seen this behaviour? Forum posting link is as follows: