1362227089 commented on issue #12267: URL: https://github.com/apache/seatunnel/issues/12267#issuecomment-5695294355
We understand the current situation: 1. The previous missing‑class error has been resolved, and standalone JDBC connections with standard postgresql driver work fine. 2. Originally, the SeaTunnel Postgres‑CDC connector aborted at authentication type 13 before snapshot, slot creation, publication and WAL decoding. This auth‑type‑13 is the SM3 national cryptographic authentication extended by HighGo (fork of PostgreSQL), which is not recognized by the official `org.postgresql.Driver`. 3. We have completed verification in our test environment: after disabling SM3 and switching HighGo to MD5 authentication mode, the SM3 auth‑type‑13 authentication error disappeared and JDBC handshake succeeded. However, the job failed at the source‑initialization phase with a replica‑identity related check error. Our test table already has a valid primary key. Therefore, we suspect this error is caused by confusion between database‑name and schema‑name in connector configuration rather than a database‑protocol defect. We are currently checking table metadata and connector configuration. 4. Simply putting both `com.highgo.jdbc.Driver` and `org.postgresql.Driver` on the classpath is not a supported upstream solution, and we will not submit connector patches or modify replication ports for upstream merging at this moment. We recognize that HighGo cannot be treated as a drop‑in replacement for the existing Postgres‑CDC connector in the upstream SeaTunnel codebase. Even if the authentication issue is resolved, other potential kernel‑level logical‑replication differences still need full validation. -- 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]
