viirya commented on a change in pull request #26312: [SPARK-29649][SQL] Stop
task set if FileAlreadyExistsException was thrown when writing to output file
URL: https://github.com/apache/spark/pull/26312#discussion_r340926443
##########
File path:
sql/core/src/test/scala/org/apache/spark/sql/sources/InsertSuite.scala
##########
@@ -735,4 +737,35 @@ class InsertSuite extends DataSourceTest with
SharedSparkSession {
assert(msg.contains("Cannot write nullable values to non-null column
's'"))
}
}
+
+ test("SPARK-29649: Stop task set if FileAlreadyExistsException was thrown") {
+ withSQLConf("fs.file.impl" -> classOf[FileExistingTestFileSystem].getName,
+ "fs.file.impl.disable.cache" -> "true") {
+ withTable("t") {
+ sql(
+ """
+ |create table t(i int, part1 int) using parquet
+ |partitioned by (part1)
+ """.stripMargin)
+
+ val df = Seq((1, 1)).toDF("i", "part1")
+ val err = intercept[SparkException] {
+ df.write.mode("overwrite").format("parquet").insertInto("t")
+ }
+ assert(err.getCause.getMessage.contains("can not write to output file:
" +
+ "org.apache.hadoop.fs.FileAlreadyExistsException"))
Review comment:
I tried to test with max failures config, however in the test max failure is
fixed at 1. Currently this checks the exception message is thrown by the added
code path in TaskSetManager.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
