codope commented on issue #7191: URL: https://github.com/apache/hudi/issues/7191#issuecomment-1524969555
> a. Few records from a table are unpartitioned and are stored in S3 with partition name "default". i.e. year/month/day --> default/default/default b. These records are not reflected in Athena but can be queried from Spark SQL Hudi used to have `default` partition value in case the partitionpath field was null for a record in partitioned table. This value is not compatible with the HIve-based engines. So, we switched to the Hive sentinel value `__HIVE_DEFAULT_PARTITION__` in https://github.com/apache/hudi/pull/5954 There is a [check](https://github.com/apache/hudi/pull/6248) during upgrade of Hudi which will fail if a partition with old default value is present in the table. When the upgrade fails, [hudi-cli](https://github.com/apache/hudi/pull/6438) can be used to repair the table. -- 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]
