cloud-fan commented on a change in pull request #24699: [SPARK-27666][CORE] Do
not release lock while TaskContext already completed
URL: https://github.com/apache/spark/pull/24699#discussion_r293217786
##########
File path: core/src/test/scala/org/apache/spark/rdd/RDDSuite.scala
##########
@@ -1176,6 +1176,24 @@ class RDDSuite extends SparkFunSuite with
SharedSparkContext {
}.collect()
}
+ test("SPARK-27666: Do not release lock while TaskContext already completed")
{
+ val rdd = sc.parallelize(Range(0, 10), 1).cache()
+ // validate cache
+ rdd.collect()
+ rdd.mapPartitions { iter =>
+ val t = new Thread(() => {
+ while (iter.hasNext) {
+ iter.next()
+ Thread.sleep(100)
+ }
+ })
+ t.setDaemon(false)
+ t.start()
+ Iterator(0)
+ }.collect()
+ Thread.sleep(10 * 150)
Review comment:
we shouldn't use sleep in tests, as the test will become flaky sooner or
later. If `CountDownLatch` doesn't work, can we use Spark `Accumulator` as
signals?
----------------------------------------------------------------
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]
