Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/22001#discussion_r209460397
--- Diff: core/src/main/scala/org/apache/spark/scheduler/DAGScheduler.scala
---
@@ -929,11 +963,38 @@ class DAGScheduler(
// HadoopRDD whose underlying HDFS files have been deleted.
finalStage = createResultStage(finalRDD, func, partitions, jobId,
callSite)
} catch {
+ case e: Exception if e.getMessage.contains(
+
DAGScheduler.ERROR_MESSAGE_BARRIER_REQUIRE_MORE_SLOTS_THAN_CURRENT_TOTAL_NUMBER)
=>
+ logWarning(s"The job $jobId requires to run a barrier stage that
requires more slots " +
+ "than the total number of slots in the cluster currently.")
+ jobIdToNumTasksCheckFailures.compute(jobId, new BiFunction[Int,
Int, Int] {
+ override def apply(key: Int, value: Int): Int = value + 1
+ })
+ val numCheckFailures = jobIdToNumTasksCheckFailures.get(jobId)
+ if (numCheckFailures <= maxFailureNumTasksCheck) {
+ messageScheduler.schedule(
+ new Runnable {
+ override def run(): Unit =
eventProcessLoop.post(JobSubmitted(jobId, finalRDD, func,
+ partitions, callSite, listener, properties))
+ },
+ timeIntervalNumTasksCheck * 1000,
--- End diff --
minor: how about removing `1000` and changing the time unit to `SECONDS`?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
