find out why rabbitmq container sometimes does not boot in time

[martenson]

we have a generous 60s startup timeout but I can see this on GCZ often, rerun usually works fine though

TASK [usegalaxy_eu.rabbitmqserver : Pause for some seconds to give time to the container to come up] ***********************************************************************************************************
Pausing for 60 seconds
(ctrl+C then 'C' = continue early, ctrl+C then 'A' = abort)
ok: [usegalaxy.cz]

...


TASK [usegalaxy_eu.rabbitmqserver : Remove guest user] *********************************************************************************************************************************************************
fatal: [usegalaxy.cz]: FAILED! => changed=true
  rc: 64
  stderr: |-
    Error: this command requires the 'rabbit' app to be running on the target node. Start it with 'rabbitmqctl start_app'.
    Arguments given:
            delete_user guest

    Usage

    rabbitmqctl [--node <node>] [--longnames] [--quiet] delete_user <username>
  stderr_lines: <omitted>
  stdout: ''
  stdout_lines: <omitted>
...ignoring

[martindemko]

Docker was moved to RBD (/rbd/docker/) on GCZ. That's probably the reason. The previous directory (/var/lib/docker) is a symlink to the new location. It was a step to free some space in the times of need. Not needed anymore.

[martenson]

we should move it back then, the space issues were resolved with shadowing hg folders

[martenson]

container rabbitmq solution replaced with native in #253