Github user frreiss commented on the issue:
https://github.com/apache/bahir/pull/13
I don't think a warning is going to be enough. As far as I can see, this
code will exhaust the Java heap and crash the Spark executor processes every
time it runs. And when those processes restart, they will not be able to replay
any lost data, because there is no code to repopulate the messages buffer. Even
if there was code to repopulate the buffer, filling the buffer up again would
only result in exhausting the heap again. Can MQTT's client-side persistence be
used to provide the ability to replay arbitrarily back in time?
---
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.
---
