parisni opened a new issue #4723:
URL: https://github.com/apache/hudi/issues/4723


   hudi 0.10.1 / aws glue metastore
   
   so far, glue does not suport the syntax used by hudi to make schema 
evolution `alter table replace colums ... cascade`
   
   As a result, when a hudi table schema evolve, the glue metastore fail to be 
updated at this step.
   
   The issue has been marked as resolved in the bellow jira, which, from my 
experience is not the case.
   
   The 
[HiveSyncTool](https://github.com/apache/hudi/blob/4f6cdd73a352bee132213438e6d11a8aef56de8b/hudi-sync/hudi-hive-sync/src/main/java/org/apache/hudi/hive/HiveSyncTool.java#L184-L197)
 code start by changing the schema and then add the partition.
   
   Workaround I can think of :
   - when a schema evolution exist, drop/recreate the schema from scratch in 
glue (last resort)
   - start by adding the partitions and then change the schema so that glue 
data get added, without schema evolution (so that user can get the data with 
the old schema)
   - implement the schema DDL evolution differently (there might be a way 
without cascade ?)
   - Provide a GlueSyncTool to overwrite the default way        
"hoodie.meta.sync.classes": "org.apache.hudi.hive.HiveSyncTool"
   
   
   related:
   
   - https://issues.apache.org/jira/browse/HUDI-874
   - https://github.com/apache/hudi/issues/1856
   - https://github.com/apache/hudi/issues/1581
   


-- 
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