Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/8763#discussion_r39554743
--- Diff: core/src/main/scala/org/apache/spark/rdd/NewHadoopRDD.scala ---
@@ -230,12 +251,16 @@ class NewHadoopRDD[K, V](
super.persist(storageLevel)
}
-
- def getConf: Configuration = confBroadcast.value.value
}
private[spark] object NewHadoopRDD {
/**
+ * Configuration's constructor is not threadsafe (see SPARK-1097 and
HADOOP-10456).
+ * Therefore, we synchronize on this lock before calling new
Configuration().
+ */
+ val CONFIGURATION_INSTANTIATION_LOCK = new Object()
--- End diff --
Ah, right. In the original version of the fix for SPARK-1097, we actually
synchronized on the Configuration being cloned. It looks like the `HadoopRDD`
object lock was added in a followup patch to avoid a deadlock caused by this
synchronization: https://github.com/apache/spark/pull/1409
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
