Export - CSV (All fields)
Export - CSV (Current fields)
SPARK-554
Add aggregateByKeySPARK-571
Forbid return statements when cleaning closuresSPARK-2849
bin/spark-submit should respect spark.driver.* for client mode
getAllowedLocalityLevel() throws ArrayIndexOutOfBoundsException
TorrentBroadcast should use the user specified serializer
Utils.getLocalDir() may return non-existent spark.local.dir directory
Snappy is now default codec - could lead to conflicts since uses /tmp
Removing broadcast in quick successions causes Akka timeout
sparkEventToJson should support SparkListenerExecutorMetricsUpdate
Deal with `--conf` options in spark-submit that relate to flags
ArrayIndexOutOfBoundsException in scheduler
LZ4 compression cause the the ExternalSort exception
Port recent spark-submit changes to windows
Akka timeouts from ContextCleaner when cleaning shuffles
org.apache.spark.broadcast.TorrentBroadcast does use the serializer class specified in the spark option "spark.serializer"
Should not swallow exception when cannot find custom Kryo registrator
RDD.countApproxDistinct() is wrong when RDD has more one partitionSPARK-1680
Clean up use of setExecutorEnvs in SparkConf
Multi-statement input to spark repl does not work
SizeBasedRollingPolicy throw an java.lang.IllegalArgumentException
API Stability in Spark 1.X (Umbrella)
BasicBlockFetchIterator#next can wait forever
Spark submit standalone-cluster mode is broken
Type mismatch in Spark shell when using case class defined in shell
Disable local execution of Spark jobs by default
Pass "cached" blocks directly to disk if memory is not large enough
Support external sorting for RDD#sortByKey()
ConnectionManager cannot distinguish whether error occurred or not
SPARK_LOCAL_DIRS may cause problems when running in local mode
Support transferring large blocks in Netty network module
Create a ShuffleMemoryManager that allocates across spilling collections in the same task
Support displaying accumulator contents in the web UI
Report TaskMetrics for running tasks
StorageStatusListener should avoid O(blocks) operations
TorrentBroadcast does not support broadcast compression
Inconsistent Kryo serialisation with custom Kryo Registrator
Can't write tight firewall rules for Spark
Add support for SequenceFiles and binary Hadoop InputFormats in PySpark
Preemptively serialize closures to help users identify non-serializable errors early on
Exponential recursion in getPreferredLocations
Spark shell prints error when :4040 port already in use
History Server enhancements
Warning when spark.storage.memoryFraction is not between 0 and 1
Missing Spark-Shell Configure Options
Users should be allowed to cogroup at least 4 RDDs
Fix reading too much or too little data from each stream in ExternalMap / Sorter
Fix committing, reverting and state tracking in shuffle file consolidation
Make sort-based shuffle write files directly when there is no sorting / aggregation and # of partitions is small
Spearman correlation computes wrong ranks when numPartitions > RDD size
ScalaReflectionSuite throw an llegalArgumentException in JDK 6