rangareddy commented on issue #16554:
URL: https://github.com/apache/hudi/issues/16554#issuecomment-4900534144
This issue was reviewed as part of the JIRA-migrated backlog triage.
Findings: row writer for clustering is already enabled by default on the
master / 1.x line. `MultipleSparkJobExecutionStrategy.performClustering` gates
on `hoodie.datasource.write.row.writer.enable`, which defaults to `true` on
master (MultipleSparkJobExecutionStrategy.java:109), and the row-writer
clustering read path was reworked to read explicit file slices through
`HoodieFileGroupReader` (`readRecordsForGroupAsRow`) instead of issuing a
`TIMESTAMP_AS_OF` query. That means the time-travel-query problem that
originally motivated disabling row writer (HUDI-7812) no longer applies on 1.x.
For context, the disable was applied only on `branch-0.x` (#11360 /
HUDI-7812, which sets the default to `false`), and the related 0.x
timestamp-based-read bug was separately fixed by #18475 / #18476 ("Fix
clustering row writer to avoid using timestamp based reads").
Since the 1.x line already ships row-writer clustering enabled with the root
cause removed, closing as fixed. If you still need it enabled/validated on the
0.x line, please reopen with details.
[MultipleSparkJobExecutionStrategy.java](https://github.com/apache/hudi/blob/master/hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/MultipleSparkJobExecutionStrategy.java#L109)
--
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]
