Hello Flink community, I am encountering an issue when trying to run a SQL job on *Flink 2.0.0 (standalone, macOS)* with a *Kafka sink (Docker).*
*Setup:* - Flink 2.0.0 installed locally (~/apps/flink-2.0.0) - Kafka/Zookeeper running via Docker Compose - Job defined in job.sql (5s tumbling window over event time, Kafka source and sink) *Submission command:* ./sql-client.sh -f ~/flink-lab/sql/job.sql * Error: * ClassNotFoundException: org.apache.flink.api.connector.sink2.StatefulSink >From what I can tell, this seems like a *compatibility issue between Flink 2.0.0 and the Kafka connector JAR*. *What I have tried so far:* - Verified job SQL syntax and configs - Confirmed Flink installation and classpath - Checked the Kafka connector JARs bundled with Flink - Searched documentation and mailing list archives *Next step I am considering:* downgrading to *Flink 1.19.1*, since it appears to be the stable recommended version, while 2.0.0 is still very new. But before I take that route, I wanted to ask: - Is this a known issue with Flink 2.0.0 and the Kafka connector? - Is there a specific Kafka JAR/connector version that should be used with Flink 2.0.0? - Would downgrading to Flink 1.19.1 be the best workaround at this point? Any guidance or confirmation would be greatly appreciated. Best regards, Tomas Nader
