Github user StefanRRichter commented on the pull request:
https://github.com/apache/flink/commit/3400a87dce9940ac8da95c89ff9791ca6a687776#commitcomment-20654494
In
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/tasks/StreamTask.java:
In
flink-streaming-java/src/main/java/org/apache/flink/streaming/runtime/tasks/StreamTask.java
on line 967:
I am not sure about the guarantees we have in Akka and the underlying
acknowledgement chain, but depending on how this works, is there a possibility
that TM acknowledges a checkpoint, the JM receives the acknowledgment, marks
the checkpoint as complete but the JM's message to the TM to acknowledge back
is lost / timeout, leading to an exception in the TM's call. Would we have a
data loss or does it work differently?
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
