Github user aarondav commented on a diff in the pull request:
https://github.com/apache/spark/pull/101#discussion_r10546476
--- Diff: core/src/main/scala/org/apache/spark/rdd/HadoopRDD.scala ---
@@ -165,12 +174,29 @@ class HadoopRDD[K, V](
override def compute(theSplit: Partition, context: TaskContext) = {
val iter = new NextIterator[(K, V)] {
+
+ private def localizeConfiguration(conf: JobConf) {
+ //generate job id
+ val stageId = context.stageId
+ val dummyJobTrackerID = new
SimpleDateFormat("yyyyMMddHHmm").format(createTime)
+ val jobId = new JobID(dummyJobTrackerID, stageId)
+ val splitID = theSplit.index
+ val attemptId = (context.attemptId % Int.MaxValue).toInt
--- End diff --
Why do you mod by MaxValue? Any casting to integer will have similar
behavior. This will not make attemptId positive, for instance, but nor is it
likely for a stage to be reattempted more than 2 billion times.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
