rabbitmq hartbeat implementation is incomplete

[Saurabh Surana]

We have added support for rabbit_heartbeat configuration in trove, but there is nothing that sends the heartbeats to rabbit resulting in the connection being dropped after the timeout.

[Michael Basnight (hubcap)]

Do you mean the conductor work? Is this bug still relevant?

[Launchpad Janitor (janitor)]

[Expired for Trove because there has been no activity for 60 days.]

[Nikhil Manchanda (slicknik)]

I'm seeing rabbitMQ disconnects from time to time in all of the Trove Services (including the guest agent) because of this. Oslo.messaging now natively supports rabbitmq heartbeats, so we should just make use of that.

[Nikhil Manchanda (slicknik)]

I've confirmed that oslo.messaging now sets the rabbit heartbeat values by default so this is not an issue any longer. The two config values relating to this that oslo.messaging recently introduced are:
    cfg.IntOpt('heartbeat_timeout_threshold',
               default=60,
               help="Number of seconds after which the Rabbit broker is "
               "considered down if heartbeat's keep-alive fails "
               "(0 disable the heartbeat). EXPERIMENTAL"),
    cfg.IntOpt('heartbeat_rate',
               default=2,
               help='How often times during the heartbeat_timeout_threshold '
               'we check the heartbeat.'),