weixiuli commented on a change in pull request #26206: [SPARK-29551][CORE] Fix
a bug about fetch failed when an executor is lost
URL: https://github.com/apache/spark/pull/26206#discussion_r338350523
##########
File path:
core/src/test/scala/org/apache/spark/scheduler/DAGSchedulerSuite.scala
##########
@@ -537,6 +536,169 @@ class DAGSchedulerSuite extends SparkFunSuite with
LocalSparkContext with TimeLi
assert(mapStatus2(2).location.host === "hostB")
}
+ test("All shuffle files on the executor should be cleaned up when executor
lost " +
+ "and then causes 'fetch failed'") {
+ // whether to kill Executor or not before FetchFailed
+ Seq(true, false).foreach { killExecutor => {
+ afterEach()
+ val conf = new SparkConf()
+ conf.set(config.SHUFFLE_SERVICE_ENABLED.key, "true")
+ conf.set("spark.files.fetchFailure.unRegisterOutputOnHost", "false")
+ init(conf)
+ runEvent(ExecutorAdded("exec-hostA1", "hostA"))
+ runEvent(ExecutorAdded("exec-hostA2", "hostA"))
+ runEvent(ExecutorAdded("exec-hostB", "hostB"))
+ val firstRDD = new MyRDD(sc, 3, Nil)
+ val firstShuffleDep = new ShuffleDependency(firstRDD, new
HashPartitioner(3))
+ val firstShuffleId = firstShuffleDep.shuffleId
+ val shuffleMapRdd = new MyRDD(sc, 3, List(firstShuffleDep))
+ val shuffleDep = new ShuffleDependency(shuffleMapRdd, new
HashPartitioner(3))
+ val secondShuffleId = shuffleDep.shuffleId
+ val reduceRdd = new MyRDD(sc, 1, List(shuffleDep))
+
+ submit(reduceRdd, Array(0))
+ // map stage1 completes successfully, with one task on each executor
+ complete(taskSets(0), Seq(
+ (Success,
+ MapStatus(BlockManagerId("exec-hostA1", "hostA", 12345),
+ Array.fill[Long](1)(2), mapTaskId = 5)),
+ (Success,
+ MapStatus(BlockManagerId("exec-hostA2", "hostA", 12345),
+ Array.fill[Long](1)(2), mapTaskId = 6)),
+ (Success, makeMapStatus("hostB", 1, mapTaskId = 7))
+ ))
+ // map stage2 completes successfully, with one task on each executor
+ complete(taskSets(1), Seq(
+ (Success,
+ MapStatus(BlockManagerId("exec-hostA1", "hostA", 12345),
+ Array.fill[Long](1)(2), mapTaskId = 8)),
+ (Success,
+ MapStatus(BlockManagerId("exec-hostA2", "hostA", 12345),
+ Array.fill[Long](1)(2), mapTaskId = 9)),
+ (Success, makeMapStatus("hostB", 1))
+ ))
+ // make sure our test setup is correct
+ val initialMapStatus1 =
mapOutputTracker.shuffleStatuses(firstShuffleId).mapStatuses
+ // val initialMapStatus1 = mapOutputTracker.mapStatuses.get(0).get
Review comment:
done.
----------------------------------------------------------------
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]
