Github user sihuazhou commented on the issue:
https://github.com/apache/flink/pull/5881
@GJL I also noticed that this PR can only solve part of the problem...it
can only make sure that the `TM` is registered with ResourceManager properly,
but it can't make sure that the `TM` could connection with JobManager
properly...
Is it possible that the problem you met is that the `TM` was killed before
connecting to `JM` successfully, that way `ResourceManager `can't be notified
to trigger a new container request and the `JM` can't be notified either...What
do you think?
---
