Improved logging for when the coordinator fails to check if the connector tasks have stopped (earlier uncaught exceptions in this path would result in the token claim thread silently dying and a false alert firing)
If there are no stopping streams, no task will be spun off to check for stop propagation. This was taking up a thread from the thread pool for no reason.
Clearly marked the start and end of waitForStopToPropagateAndMarkDatastreamsStopped. Removed redundant message
Added a confirmation message for when the leader transitions a stream to the STOPPED state.
Things that are NOT part of this PR:
Separation of the thread pools for leader's tasks to wait for stop propagation and for followers tasks to claim tokens.
Printing how much time it took the leader to mark a stream STOPPED (this is not a trivial change and will be evaluated/addressed separately)
waitForStopToPropagateAndMarkDatastreamsStopped. Removed redundant messageThings that are NOT part of this PR: