In spark <http://www.orienit.com/courses/spark-and-scala-training-in-hyderabad> 2.3 with blacklist enabled this is a common problem when executor A has some problem, for instance let’s say there’s some connection issue. Tasks on executor B, executor C will fail saying cannot read from executor A. This would make the job fail due to task on executor B failed 4 times.
-- Sent from: http://apache-spark-user-list.1001560.n3.nabble.com/ --------------------------------------------------------------------- To unsubscribe e-mail: user-unsubscr...@spark.apache.org