Description of problem: rabbitmq server has a possibility to enable REST API of the service by allowing rabbitmq management plugin. The only way right now how to query and set rabbitmq cluster is through rabbitmqctl command line tool which output is erlang formatted. By enabling REST API, json formatted output could be got and easily parsable for automation purposes. How to enable the management pluging: 1. Create a file in /etc/rabbitmq/enabled_plugins with content of: [rabbitmq_management]. 2. Make it rw for rabbitmq user. 3. Edit rabbitmq_management section in /etc/rabbitmq/rabbitmq.config to look like: {rabbitmq_management, [ {listener, [{port, 15672}, {ip, "127.0.0.1"}]} ]} - So for now It's allowed only on localhost, possibly It could be allowed on internal interface used for rabbitmq communication.
So the puppet-rabbitmq does support the enabling of the management plugin via the $admin_enable class parameter. The issue is that it only works when: $admin_enable and $service_manage are true. Since we manage rabbitmq with pacemaker we do set $service_manage to false, so this approach won't work. The simplest thing will be to wait for the puppet-pacemaker changes to manage all pacemaker services via puppet. I need to think about this one a bit more.
This bug did not make the OSP 8.0 release. It is being deferred to OSP 10.
So if I tweak the puppet-rabbitmq module to not require service_manage to be true (we are using pacemaker so it is always false), but during the deployment I get an error because the module insists on verifying it is up: Error: curl -k --noproxy 172.16.2.6 --retry 30 --retry-delay 6 -f -L -o /var/lib/rabbitmq/rabbitmqadmin http://guest:DRzBkenNTDk9ybGRZUktQn9y4@172.16.2.6:15672/cli/rabbitmqadmin returned 7 instead of one of [0]\u001b[0m\n\u001b[1;31mError: /Stage[main]/Rabbitmq::Install::Rabbitmqadmin/Staging::File[rabbitmqadmin]/Exec[/var/lib/rabbitmq/rabbitmqadmin]/returns: change from notrun to 0 failed: curl -k --noproxy 172.16.2.6 --retry 30 --retry-delay 6 -f -L -o /var/lib/rabbitmq/rabbitmqadmin http://guest:DRzBkenNTDk9ybGRZUktQn9y4@172.16.2.6:15672/cli/rabbitmqadmin returned 7 instead of one of [0]\u001b[0m\n\u001b[1;31mWarning: /Stage[main]/Rabbitmq::Install::Rabbitmqadmin/File[/usr/local/bin/rabbitmqadmin]: Skipping because of failed dependencies\u001b[0m\n" So the changes to the puppet module might need to be a bit more involved
@michele should we move this out to OSP 11?
This bugzilla has been removed from the release and needs to be reviewed and Triaged for another Target Release.
Michele: any plans for this or is it no longer needed?
verified with RHOS_TRUNK-15.0-RHEL-8-20190913.n.3.
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. https://access.redhat.com/errata/RHEA-2019:2811