Github user tedxia commented on the pull request:
https://github.com/apache/storm/pull/268#issuecomment-60866235
@clockfly , when worker A not get notified that worker B is not alive, the
connection from A to B should abort, but not throw RuntimeException, because
will cause the worker A exit even if the connection between worker A and other
worker are normal. There are many factors that cause nimbus not inform worker A
that worker B died, such as heavy load of zookeeper, failure of nimbus and so
on. For those reason even if network partition, I still don't think it is a
good idea for worker A exit, fault tolerance is too high a price to pay.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
