Github user ravipesala commented on a diff in the pull request:
https://github.com/apache/carbondata/pull/1476#discussion_r149909385
--- Diff:
integration/spark2/src/main/scala/org/apache/spark/sql/execution/command/schema/AlterTableDropColumnCommand.scala
---
@@ -55,6 +55,14 @@ private[sql] case class AlterTableDropColumnCommand(
carbonTable = metastore
.lookupRelation(Some(dbName),
tableName)(sparkSession).asInstanceOf[CarbonRelation]
.tableMeta.carbonTable
+ if (carbonTable.isPreAggregateTable) {
--- End diff --
we should check the children tables columns and throw the exception if drop
column matches any child table column along with information of tables
---