Github user manishgupta88 commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1769#discussion_r161143694
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/execution/command/management/CarbonLoadDataCommand.scala
---
@@ -251,7 +255,9 @@ case class CarbonLoadDataCommand(
case ex: Exception =>
LOGGER.error(ex)
// update the load entry in table status file for changing the
status to marked for delete
- CarbonLoaderUtil.updateTableStatusForFailure(carbonLoadModel)
+ if (isUpdateTableStatusRequired && !table.isHivePartitionTable) {
--- End diff --
!table.isHivePartitionTable...this condition is not required while calling
updateTableStatusForFailure as boolean flag isUpdateTableStatusRequired will
be set ti true only if the table is not a hive partition table
---
