Reza Safi created SPARK-30064:
---------------------------------
Summary: Incomplete Graphical SQL Plan in Spark application UI's
SQL tab for SaveIntoDataSourceCommand
Key: SPARK-30064
URL: https://issues.apache.org/jira/browse/SPARK-30064
Project: Spark
Issue Type: Bug
Components: SQL
Affects Versions: 2.4.4, 2.3.4, 3.0.0, 3.1.0
Reporter: Reza Safi
For reproduction consider the following for jdbc datasource:
{noformat}
spark-shell --master yarn-client --jars /*/mysql-connector-java.jar
import java.util.Properties
val connectionProperties = new Properties()
connectionProperties.put("user", "***")
connectionProperties.put("driver", "com.mysql.jdbc.Driver")
val jdbc_url ="jdbc:mysql://****"
Seq(1 -> "a").toDF("i",
"j").write.mode("append").jdbc(jdbc_url,"test.test3",connectionProperties)
{noformat}
Then check the web UI's SQL tab and select the query and look at the plan. You
will see a one node tree for the plan which just says "Execute
SaveIntoDataSourceCommand".
It seems that the bug was first fixed in SPARK-20213(PR#18064), but that fix
was partially reverted in SPARK-22252(PR#19474).
It seems that the fix should be to make SaveIntoDataSourceCommand to have an
actual child for the run method, but that requires changes to other places.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
