Description of problem: Our upgrade scripts do not currently make the change in mcollective connector. This is not a problem until one of two things happens: 1. A new node is created with the new instructions which specify "activemq". This will be incompatible with the existing systems - the new nodes will not even respond to mco ping. 2. We upgrade to the next version of mcollective where "stomp" is not just deprecated removed. Steps to Reproduce: 1. Upgrade OSE 1.1 to OSE 1.2 2. Look in /etc/mcollective/server.cfg on nodes 3. Look in /etc/mcollective/client.cfg on brokers Actual results: connector = stomp plugin.stomp.pool.size = 1 plugin.stomp.pool.host1 = activemq.ose.example.com plugin.stomp.pool.port1 = 61613 [...] Expected results: connector = activemq plugin.activemq.pool.size = 1 plugin.activemq.pool.1.host = activemq.ose.example.com plugin.activemq.pool.1.port = 61613 [...] (Note the slight difference in format!) Additional info: Would be wise to add a check for the old connector in oo-diagnostics and have it refer to a helpful kbase article or at least this bug.
Fix submitted here: https://github.com/openshift/enterprise/pull/30 Awaiting review
Merged PR, tested against working single-server and server-pool mcollective configs.
Verified this bug with the following packages, and PASS. openshift-enterprise-upgrade-broker-1.2.2-1.el6op.noarch openshift-enterprise-upgrade-node-1.2.2-1.el6op.noarch Before upgrade: # cat ~/client.cfg <--snip--> connector = stomp plugin.stomp.host = activemq.rhn.com plugin.stomp.port = 61613 plugin.stomp.user = mcollective plugin.stomp.password = marionette # cat ~/server.cfg <--snip--> connector = stomp plugin.stomp.host = activemq.rhn.com plugin.stomp.port = 61613 plugin.stomp.user = mcollective plugin.stomp.password = marionette After upgrade: # cat /etc/mcollective/client.cfg <--snip--> connector = activemq plugin.activemq.pool.size = 1 plugin.activemq.pool.1.host = activemq.rhn.com plugin.activemq.pool.1.port = 61613 plugin.activemq.pool.1.user = mcollective plugin.activemq.pool.1.password = marionette # cat /etc/mcollective/server.cfg <--snip--> connector = activemq plugin.activemq.pool.size = 1 plugin.activemq.pool.1.host = activemq.rhn.com plugin.activemq.pool.1.port = 61613 plugin.activemq.pool.1.user = mcollective plugin.activemq.pool.1.password = marionette After upgrade, add a new node with the new instructions which specify "activemq" mco ping could find this new node.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-1275.html