chamikaramj commented on code in PR #22261: URL: https://github.com/apache/beam/pull/22261#discussion_r922339733
########## sdks/python/apache_beam/io/kafka.py: ########## @@ -76,16 +76,33 @@ For more information specific to Flink runner see: - https://beam.apache.org/documentation/runners/flink/ + + Reading via Kafka SDF is currently broken, and will cause the pipeline Review Comment: I think we can remove Python updates since we made UnboundedSource wrapped SDF Kafka the default for now: https://github.com/apache/beam/pull/22286 ########## sdks/java/io/kafka/src/main/java/org/apache/beam/sdk/io/kafka/KafkaIO.java: ########## @@ -543,6 +543,11 @@ * For any significant significant updates to this I/O connector, please consider involving * corresponding code reviewers mentioned <a * href="https://github.com/apache/beam/blob/master/sdks/java/io/kafka/OWNERS";>here</a>. + * + * <h1>Reading from Kafka SDF is currently broken, as re-starting the pipeline will cause the + * consumer to start from scratch. See <a + * href="https://github.com/apache/beam/issues/21730";>this</a>. Current workaround is to use + * --experimental_option=use_unbounded_sdf_wrapper to use the Unbounded implementation</h1> Review Comment: "For runners that require SDF, current workaround is to use ..." Also, pls confirm that this works for Java pipelines that use Dataflow Runner v2. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
