AWS clean-rooms documentation change
Summary
Added troubleshooting section for AWS Clean Rooms Spark SQL regarding missing partition data in S3/Glue configurations
Security assessment
The change adds guidance about data integrity checks for partitioned tables but does not address a specific security vulnerability or document security features. It focuses on configuration validation rather than security controls.
Diff
diff --git a/clean-rooms/latest/userguide/troubleshooting.md b/clean-rooms/latest/userguide/troubleshooting.md index f135508e5..02cd2b392 100644 --- a//clean-rooms/latest/userguide/troubleshooting.md +++ b//clean-rooms/latest/userguide/troubleshooting.md @@ -5 +5 @@ -One or more tables referenced by the query is not accessible by its associated service role. The table/role owner must grant the service role access to the table.One of the underlying datasets has an unsupported file format.Query results are not as expected when using Cryptographic Computing for Clean Rooms. +One or more tables referenced by the query is not accessible by its associated service role. The table/role owner must grant the service role access to the table.One of the underlying datasets has an unsupported file format.Query results are not as expected when using Cryptographic Computing for Clean Rooms.AWS Clean Rooms Spark SQL: Missing partition data @@ -18,0 +19,2 @@ This section describes some common issues that might arise when using AWS Clean + * AWS Clean Rooms Spark SQL: Missing partition data + @@ -68,0 +71,10 @@ For more information, see [Cryptographic Computing for Clean Rooms](./crypto-com +## AWS Clean Rooms Spark SQL: Missing partition data + +If you are using the AWS Clean Rooms Spark SQL engine with partitioned tables S3 tables, all partitions in the AWS Glue Data Catalog must also have data in S3. The engine uses the Spark setting `spark.sql.files.ignoreMissingFiles=False` + +For more information, see [https://spark.apache.org/docs/latest/sql-data-sources-generic-options.html#ignore-missing-files ](https://spark.apache.org/docs/latest/sql-data-sources-generic-options.html#ignore-missing-files). + +If you encounter this error you will get the following error message: `"Missing partition data: One of the configured tables is partitioned and one or more of the partitions does not have data".` + +Compare your data present in Amazon S3 with the partitions listed in the AWS Glue Data Catalog for the table. Delete partitions that don't have corresponding data in S3. +