Hisoka-X opened a new pull request, #4270: URL: https://github.com/apache/flink-cdc/pull/4270
This close https://issues.apache.org/jira/browse/FLINK-38455 When the pipeline starts, Flink CDC sends CreateTableEvent (a SchemaChangeEvent) before any DataChangeEvent. The ElasticsearchEventSerializer returns null for SchemaChangeEvent since it only updates the internal schema cache and does not produce an actual ES operation. In Elasticsearch8AsyncWriter.submitRequestEntries(), operations with null BulkOperationVariant are skipped via continue. However, when all operations in a batch are SchemaChangeEvent (which is common at pipeline startup), no operation is added to the BulkRequest.Builder. Calling br.build() on an empty builder throws: ``` Missing required property 'BulkRequest.operations' ``` because operations is a required field in the ES8 Java Client's BulkRequest. The fix checks whether any valid operation was added before calling br.build(). If the batch is effectively empty, it returns success immediately without sending a request to Elasticsearch. -- 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]
