*** Bug 1283694 has been marked as a duplicate of this bug. ***

(In reply to Lars Kellogg-Stedman from comment #0)
> The diagnostic information collected by sosreport does not currently
> included information necessary for debugging rabbitmq problems.  This almost
> invariably results in multiple round-trips to the customer as we ask them to
> provide the output of various commands.
> 
> For our OpenStack deployments we should be collecting, by default, at least
> the output of "rabbitmqctl report", and possibly the output of the various
> "rabbitmqctl list_*" commands.

Thanks for the report Lars, did you see this when running sos on an OSP 5 or 6 environment? Are you able to provide the version of sos you were using etc?

I ask as `rabbitmqctl report` should be collected with the forked versions of sos we currently ship with RHEL OSP 7 and plan to provide for OSP 5 & 6 shortly.

As for the `rabbitmqctl list_*` commands my understanding is that a report should already contain all of the data provided by these commands. If that isn't the case would you mind providing some examples and I'll get these added to the rabbitmq plugin.

Upstream does this (in addition to logs & config):

    def setup(self):
        self.add_cmd_output("rabbitmqctl report")
        self.add_cmd_output("rabbitmqctl cluster_status")
        self.add_cmd_output("rabbitmqctl list_policies")

Is that sufficient?

Sure.  This request stemmed from a recent customer engagement in which we weren't getting any rabbitmq data, but the customer is running OSP 5 so that would explain it.

I guess getting these changes released in OSP 5 would make everybody happy.

Apologies for the delay here, I'm closing this out as a duplicate of BZ#1310433 where I'll chase to get an updated version of sos into OSP 5 for RHEL 7 hosts.

*** This bug has been marked as a duplicate of bug 1310433 ***