mridulm commented on code in PR #55347:
URL: https://github.com/apache/spark/pull/55347#discussion_r3337736794
##########
core/src/main/scala/org/apache/spark/TaskEndReason.scala:
##########
@@ -107,6 +107,28 @@ case class FetchFailed(
override def countTowardsTaskFailures: Boolean = false
}
+/**
+ * :: DeveloperApi ::
+ * Task failed to fetch shuffle metadata from a remote node. This failure is
expected to recover
+ * by simply repeating the task. This does not count towards task failures, so
the stage will never
+ * fail due to this failure.
+ */
+@DeveloperApi
+case class MetadataFetchFailed(
+ shuffleId: Int,
+ reduceId: Int,
+ message: String)
+ extends TaskFailedReason {
+ override def toErrorString: String = {
+ s"MetadataFetchFailed(shuffleId=$shuffleId, reduceId=$reduceId,
message=\n$message\n)"
+ }
+
+ /**
+ * Metadata fetch failures do not count toward task failures.
+ */
+ override def countTowardsTaskFailures: Boolean = false
Review Comment:
What is preventing the application from becoming unrecoverable ? (constantly
retry on failed metadata fetches)
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
