fxbing opened a new pull request, #4574: URL: https://github.com/apache/flink-cdc/pull/4574
## What is the purpose of this pull request? Fixes [FLINK-40982](https://issues.apache.org/jira/browse/FLINK-40982). An ADD COLUMN applied to Fluss can be replayed after recovery from an earlier checkpoint. The existing columns then cause a duplicate-column error and prevent recovery. ## Brief change log - Attempt ADD directly; check the target schema only after `InvalidAlterTableException`. - Accept existing columns with matching Fluss types and retry missing columns only when their count decreases. - Reject conflicting types, missing tables, and unrelated failures. ## Verifying this change - Extended `FlussMetadataApplierTest` against a real Fluss test cluster to cover full and partial replay, type conflicts, unchanged schemas after failures, and error propagation. - Tests passed on Java 11 / Flink 1.20.3 (25 and 2 cases in separate runs) and Java 17 / Flink 2.2.0 (27 cases after a clean build). Spotless and Checkstyle passed. - Full job recovery E2E and deterministic concurrent DDL injection were not run. ## Documentation - Does this pull request introduce a new feature? no - If yes, how is the feature documented? not applicable --- ##### Was generative AI tooling used to co-author this PR? - [X] Yes (OpenAI Codex) -- 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]
