Hi, In our Titan Framework, when a scheduler instance looses leadership, because of zookeeper connectivity etc, we stop the Mesos driver by calling driver.stop(true)
And if the same instance gets leadership back again, we call driver.start(). Note that the driver instance remains the same, as in it's the same object. However when I call driver.start() after the driver is stopped the first time, I see the status of the driver remains to be DRIVER_STOPPED I am wondering if I should re-create the driver object after I stop it. -- Thanks, Diptanu Choudhury Web - www.linkedin.com/in/diptanu Twitter - @diptanu <http://twitter.com/diptanu>
