Github user fhueske commented on the issue:
https://github.com/apache/flink/pull/2332
Hi @delding, I'm sorry that I did not mention this earlier, but I just
noticed that the `HBaseSink` does not implement any logic for checkpointing and
fault-tolerance.
The checkpointing logic of a sink and the guarantees of the external data
store define what kind of emission guarantees a sink function can offer
(exactly once, at-least once). In order to make this PR mergable, we need to
figure out what kind of guarantees it provides under which assumptions / in
which scenarios. As an example see the [documentation of the Cassandra
sink](https://ci.apache.org/projects/flink/flink-docs-master/dev/connectors/cassandra.html).
This is very important because users will rely on the stated guarantees.
I am not very familiar with HBase and what kind of write semantics and
transnational guarantees it provides. Can you describe which kind of guarantees
an `HBaseSink` can achieve and how it would be integrated with Flink's
Checkpointing mechanism?
Thanks, Fabian
---
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.
---
