Github user rmetzger commented on the pull request:
https://github.com/apache/flink/pull/474#issuecomment-82621443
First of all: Great work @gaborhermann! The whole Kafka connector code is
well written and contains logic for a lot of cases. I think it was already
quite some work to get it to the state it currently is.
I took this pull request as a base to add an integration test which is
starting a local Zookeeper and Kafka instance.
I also fixed a critical issue in the streaming code, regarding exception
forwarding.
My current WIP is here: https://github.com/rmetzger/flink/tree/kafka-gabor
The code needs some more care until its in a good state. The following is
missing:
- More integration tests: Testing of the `PersistentKafkaSource` with
taskmanager failure and job restart, also tests for other stuff (partitioner as
class & as instance)
- Add means to `KafkaSource` and `KafkaSink` for users to pass their own
configuration properties to Kafka.
- There are some TODOs in the code
- Testing of the code in a distributed setup.
I'll work on the cluster testing tomorrow. In the course of that, I'm
probably going to add more code to my "wip" branch.
---
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.
---
