Glue ETL Not Update Data Catalog

0

Hi, I am creating a Glue Job that is transforming CSV files to partitioned Parquet files and I want to update the Data Catalog from the ETL. I am using this snipe of code to do that:

    dynamic_frame: DynamicFrame = DynamicFrame.fromDF(final_data, glue_context, f"{file_type}_dataset")
    sink = glue_context.getSink(
        connection_type="s3",
        enableUpdateCatalog=True,
        updateBehavior="UPDATE_IN_DATABASE",
        path=f"{target}",
        partitionKeys=partition_cols)

    sink.setFormat("glueparquet")
    sink.setCatalogInfo(catalogDatabase=conf.get_db_name(),
                        catalogTableName=conf.get_table_name_by_source(file_type))
    sink.writeFrame(dynamic_frame)

As you can see I am transforming a Spark DF to a glue DynamicFrame and later writing it to parquet.

The output parquet files are written but I am getting this error and the tables are not created in the Data Catalog:

Exception: Problem processing file type cdr_cs because An error occurred while calling o477.pyWriteDynamicFrame.
: scala.MatchError: (null,false) (of class scala.Tuple2)
	at com.amazonaws.services.glue.DataSink.forwardPotentialDynamicFrameToCatalog(DataSink.scala:177)
	at com.amazonaws.services.glue.DataSink.forwardPotentialDynamicFrameToCatalog(DataSink.scala:135)
	at com.amazonaws.services.glue.sinks.HadoopDataSink.$anonfun$writeDynamicFrame$2(HadoopDataSink.scala:302)
	at com.amazonaws.services.glue.util.FileSchemeWrapper.$anonfun$executeWithQualifiedScheme$1(FileSchemeWrapper.scala:77)
	at com.amazonaws.services.glue.util.FileSchemeWrapper.executeWith(FileSchemeWrapper.scala:70)
	at com.amazonaws.services.glue.util.FileSchemeWrapper.executeWithQualifiedScheme(FileSchemeWrapper.scala:77)
	at com.amazonaws.services.glue.sinks.HadoopDataSink.$anonfun$writeDynamicFrame$1(HadoopDataSink.scala:157)
	at org.apache.spark.sql.execution.SQLExecution$.withSQLConfPropagated(SQLExecution.scala:253)
	at com.amazonaws.services.glue.sinks.HadoopDataSink.writeDynamicFrame(HadoopDataSink.scala:151)
	at com.amazonaws.services.glue.DataSink.pyWriteDynamicFrame(DataSink.scala:64)
	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
	at java.lang.reflect.Method.invoke(Method.java:498)
	at py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:244)
	at py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:357)
	at py4j.Gateway.invoke(Gateway.java:282)
	at py4j.commands.AbstractCommand.invokeMethod(AbstractCommand.java:132)
	at py4j.commands.CallCommand.execute(CallCommand.java:79)
	at py4j.GatewayConnection.run(GatewayConnection.java:238)
	at java.lang.Thread.run(Thread.java:750)
Rafael
已提问 2 年前1679 查看次数
1 回答
2
已接受的回答

This does not give all the information. Given the error happens while writing to the DynamicFrame, there might be a difference in the file layout and the table definition in the Glue Data Catalog.

In some situations this could be access issues as well. Check the IAM access and if Lake formation if turned on, check on Lake formation as well.

profile pictureAWS
已回答 2 年前
  • Thanks for your answer. There are no tables in the database in the Data Catalog so maybe the problem has to do with a connectivity or conf problem. The thing is that the error above is quite difficult to troubleshoot because has no meaning at all.

  • Finally the problem was that Lake formation was not activated. Thanks @ananthtm for your help.

您未登录。 登录 发布回答。

一个好的回答可以清楚地解答问题和提供建设性反馈,并能促进提问者的职业发展。

回答问题的准则