xxubai opened a new issue, #11352:
URL: https://github.com/apache/gravitino/issues/11352

   ### Version
   
   main branch
   
   ### Describe what's wrong
   
   When loading an Iceberg table from the Glue catalog, Gravitino logs a 
warning saying that partitioning and sort order information may be incomplete.
   
   However, the actual root cause is that the Iceberg metadata JSON file 
referenced by the Glue table does not exist in S3. The current warning is 
misleading because it makes the issue look like a partition/sort recovery 
problem, while the table metadata file itself is missing.
   
   ### Error message and/or stacktrace
   
   ```text
   WARN GlueCatalogOperations.loadTable - Failed to load Iceberg metadata for 
table <database>.<table>. Partitioning and sort order information may be 
incomplete.
   
   org.apache.iceberg.exceptions.NotFoundException: Location does not exist:
   s3://<bucket>/<table>/metadata/<version>.metadata.json
   
   Caused by: software.amazon.awssdk.services.s3.model.NoSuchKeyException:
   The specified key does not exist. (Service: S3, Status Code: 404)
   ```
   
   ### How to reproduce
   
   1. Configure a Gravitino Glue catalog with an Iceberg table.
   2. Make the Glue table reference an Iceberg metadata location whose S3 
object no longer exists.
   3. Load the table through Gravitino, or trigger a path that calls 
`GlueCatalogOperations.loadTable`.
   4. Check the server log.
   
   Expected behavior: the log should clearly indicate that the Iceberg metadata 
file is missing or cannot be loaded.
   
   Actual behavior: the log says partitioning and sort order information may be 
incomplete, which hides the real cause.
   
   ### Additional context
   
   The warning is emitted from `GlueCatalogOperations.loadTable` after 
`GlueIcebergTableHelper.loadTable` fails. The catch block handles all 
exceptions with the same partition/sort warning, so missing metadata files are 
reported with an inaccurate message.
   


-- 
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]

Reply via email to