joshhamann opened a new issue, #9750: URL: https://github.com/apache/hudi/issues/9750
**_Tips before filing an issue_** - Have you gone through our [FAQs](https://hudi.apache.org/learn/faq/)? Yes - Join the mailing list to engage in conversations and get faster support at [email protected]. - If you have triaged this as a bug, then file an [issue](https://issues.apache.org/jira/projects/HUDI/issues) directly. I believe this is a bug, but wanted to post here first. **Describe the problem you faced** Using Glue4.0, Hudi 0.12.1, Spark 3.3.0 When using bulk_insert in combination with the configuration "hoodie.combine.before.insert = True", we get incorrect values in _hoodie_record_key (which do not match the key column). The configuration does in fact prevent duplicates from entering the destination hudi table, but issues arise if you were to then upsert records against these garbage _hoodie_record_key values. **To Reproduce** Steps to reproduce the behavior: 1. Start with an empty table 2. Run bulk insert with the configuration hoodie.combine.before.insert = True 3. Notice where _hoodie_record_key != {particular key column} 4. Then run an upsert using the same source data 5. Notice duplicates inserted (where we would expect the upsert functionality to work) **Expected behavior** _hoodie_record_key has proper values when using configuration hoodie.combine.before.insert = True (which match the expected value from {particular key column}), this then allows hudi upsert functionality to work as expected. **Environment Description** * Hudi version : 0.12.1 * Spark version : 3.3.0 * Storage (HDFS/S3/GCS..) : S3 * Running on Docker? (yes/no) : Verified this occurs both on Docker image provided by AWS locally and on the Glue AWS service itself **Additional context** * Glue Version: 4.0 -- 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]
