[rabbitmq] Log collection fixes when rabbitmq runs in a container

When rabbitmq is running inside a container all the rabbitmqctl
commands will fail with some obscure (welsh?) error messages like:

$ rabbitmqctl cluster_status
erno" ienrirto rt elromgigneart ipnrge sienn td)o _booetr"r,o{r: b<a0d.ar2g.,0[>{
l_prim_loader,check_file_result,3,[]},{init,get_boot,1,[]},{init,get_boot,2,[]},{init,do_boot,3,[]}]}}
init terminating in do_boot ()

When run inside containers we need to run any rabbitmqctl command inside
the container, so that output is properly collected:
$ docker exec -t rabbitmq-bundle-docker-0 rabbitmqctl cluster_status
Cluster status of node 'rabbit@controller-0' ...
[{nodes,[{disc,['rabbit@controller-0','rabbit@controller-1',
                'rabbit@controller-2']}]},
 {running_nodes,['rabbit@controller-2','rabbit@controller-1',
                 'rabbit@controller-0']},
 {cluster_name,<<"rabbit@controller-0.localdomain">>},
 {partitions,[]},
 {alarms,[{'rabbit@controller-2',[]},
          {'rabbit@controller-1',[]},
          {'rabbit@controller-0',[]}]}]

While we're at it we also collect the logs of any rabbitmq container.
This is particularly useful in containerized openstack deployments
where short-lived containers are used to set up the cloud.

Since the info output by 'rabbitmqctl cluster_status' and 'rabbitmqctl
list_policies' is already contained in 'rabbitmqctl report', we just
remove the former two commands.

First reported via RHBZ#1525620

Signed-off-by: Michele Baldessari <michele@acksyn.org>
Signed-off-by: John Eckersberg <jeckersb@redhat.com>

0 files changed, 0 insertions, 0 deletions