Github user revans2 commented on a diff in the pull request:
https://github.com/apache/storm/pull/2433#discussion_r172934007
--- Diff: conf/defaults.yaml ---
@@ -182,7 +195,7 @@ worker.log.level.reset.poll.secs: 30
# control how many worker receiver threads we need per worker
topology.worker.receiver.thread.count: 1
-task.heartbeat.frequency.secs: 3
+task.heartbeat.frequency.secs: 60
--- End diff --
This is a bit scary. Why is the default now 20 times bigger than it was
before? If this is just for scale we should drop it back to the default and
write up documentation on how to scale a cluster much larger. Even in that
case I don't feel comfortable with waiting 10+ mins for a worker to time out in
the worst case.
---
