Note that frameworks can set their own failover timeout. If a framework is disconnected for more than the failover timeout duration, all its executors/tasks are killed.
If a framework tries to re-register with a "framework id", *currently* the master allows it. In the future (when master has persistent state about the frameworks) this will be disallowed. On Wed, Apr 16, 2014 at 12:42 PM, David Greenberg <[email protected]>wrote: > I don't recall the exact timeout of framework IDs, but what I'm wondering > is what happens if a scheduler tries to "failover", but the failover grace > period has elapsed? Does it fail to register, or does it successfully > register and all the old executors are just gone? >
