Michael Allman created SPARK-18202:
--------------------------------------
Summary: Spark throws a mysterious system error when a Hive
command has at least 100,000 results
Key: SPARK-18202
URL: https://issues.apache.org/jira/browse/SPARK-18202
Project: Spark
Issue Type: Bug
Components: SQL
Affects Versions: 2.0.1
Reporter: Michael Allman
There's a hard coded limit (100,000) to the number of results a hive command
can return. If the number of results is greater than or equal this limit, Spark
throws a system exception with the somewhat mysterious message "RESULTS
POSSIBLY TRUNCATED".
This problem manifests itself when calling `SHOW PARTITIONS <table>`, where
<table> is a Hive table with not less than 100,000 partitions.
The code in question can be found here:
https://github.com/apache/spark/blob/ccb11543048dccd4cc590a8db1df1d9d5847d112/sql/hive/src/main/scala/org/apache/spark/sql/hive/client/HiveClientImpl.scala#L560-L566
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
