Hi Mohit,
Try increasing the *executor* memory instead of the worker memory - the
most appropriate place to do this is actually when you're creating your
SparkContext, something like:
conf = pyspark.SparkConf()
.setMaster("spark://master:7077")
.setAppName("Example")
.setSparkHome("/your/path/to/spark")
.set("spark.executor.memory", "20G")
.set("spark.logConf", "true")
sc = pyspark.SparkConf(conf = conf)
Hope that helps,
Bryn
On Wed, Feb 26, 2014 at 2:39 PM, Mohit Singh <[email protected]> wrote:
> Hi,
> I am experimenting with pyspark lately...
> Every now and then, I see this error bieng streamed to pyspark shell ..
> and most of the times.. the computation/operation completes.. and
> sometimes, it just gets stuck...
> My setup is 8 node cluster.. with loads of ram(256GB's) and space( TB's)
> per node.
> This enviornment is shared by general hadoop and hadoopy stuff..with
> recent spark addition...
>
> java.lang.OutOfMemoryError: Java heap space
> at
> com.ning.compress.BufferRecycler.allocEncodingBuffer(BufferRecycler.java:59)
> at com.ning.compress.lzf.ChunkEncoder.<init>(ChunkEncoder.java:93)
> at
> com.ning.compress.lzf.impl.UnsafeChunkEncoder.<init>(UnsafeChunkEncoder.java:40)
> at
> com.ning.compress.lzf.impl.UnsafeChunkEncoderLE.<init>(UnsafeChunkEncoderLE.java:13)
> at
> com.ning.compress.lzf.impl.UnsafeChunkEncoders.createEncoder(UnsafeChunkEncoders.java:31)
> at
> com.ning.compress.lzf.util.ChunkEncoderFactory.optimalInstance(ChunkEncoderFactory.java:44)
> at
> com.ning.compress.lzf.LZFOutputStream.<init>(LZFOutputStream.java:61)
> at
> org.apache.spark.io.LZFCompressionCodec.compressedOutputStream(CompressionCodec.scala:60)
> at
> org.apache.spark.storage.BlockManager.wrapForCompression(BlockManager.scala:803)
> at
> org.apache.spark.storage.BlockManager$$anonfun$5.apply(BlockManager.scala:471)
> at
> org.apache.spark.storage.BlockManager$$anonfun$5.apply(BlockManager.scala:471)
> at
> org.apache.spark.storage.DiskBlockObjectWriter.open(BlockObjectWriter.scala:117)
> at
> org.apache.spark.storage.DiskBlockObjectWriter.write(BlockObjectWriter.scala:174)
> at
> org.apache.spark.scheduler.ShuffleMapTask$$anonfun$runTask$1.apply(ShuffleMapTask.scala:164)
> at
> org.apache.spark.scheduler.ShuffleMapTask$$anonfun$runTask$1.apply(ShuffleMapTask.scala:161)
> at scala.collection.Iterator$class.foreach(Iterator.scala:727)
> at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
> at
> org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:161)
> at
> org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:102)
> at org.apache.spark.scheduler.Task.run(Task.scala:53)
> at
> org.apache.spark.executor.Executor$TaskRunner$$anonfun$run$1.apply$mcV$sp(Executor.scala:213)
> at
> org.apache.spark.deploy.SparkHadoopUtil.runAsUser(SparkHadoopUtil.scala:49)
> at
> org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:178)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:744)
>
>
>
> Most of the settings in spark are default.. So i was wondering if maybe,
> there is some configuration that needs to happen?
> There is this one config I have addded to spark_env file
> SPARK_WORKER_MEMORY=20g
>
> Also, I see tons of these errors as well..
> 14/02/26 14:33:17 INFO TaskSetManager: Loss was due to
> java.lang.OutOfMemoryError: Java heap space [duplicate 1]
> 14/02/26 14:33:17 INFO TaskSetManager: Starting task 996.0:278 as TID 1792
> on executor 9: node02 (PROCESS_LOCAL)
> 14/02/26 14:33:17 INFO TaskSetManager: Serialized task 996.0:278 as 4070
> bytes in 0 ms
> 14/02/26 14:33:17 WARN TaskSetManager: Lost TID 1488 (task 996.0:184)
> 14/02/26 14:33:17 INFO TaskSetManager: Loss was due to
> java.lang.OutOfMemoryError: Java heap space [duplicate 2]
> 14/02/26 14:33:17 INFO TaskSetManager: Starting task 996.0:247 as TID 1793
> on executor 9: node02 (PROCESS_LOCAL)
> 14/02/26 14:33:17 INFO TaskSetManager: Serialized task 996.0:247 as 4070
> bytes in 0 ms
> 14/02/26 14:33:17 WARN TaskSetManager: Lost TID 1484 (task 996.0:82)
> 14/02/26 14:33:17 INFO TaskSetManager: Loss was due to
> java.lang.OutOfMemoryError: Java heap space [duplicate 3]
> 14/02/26 14:33:17 INFO TaskSetManager: Starting task 996.0:116 as TID 1794
> on executor 9: node02 (PROCESS_LOCAL)
> 14/02/26 14:33:17 INFO TaskSetManager: Serialized task 996.0:116 as 4070
> bytes in 1 ms
> 14/02/26 14:33:17 WARN TaskSetManager: Lost TID 1475 (task 996.0:157)
> 14/02/26 14:33:17 INFO TaskSetManager: Loss was due to
> java.lang.OutOfMemoryError: Java heap space [duplicate 4]
> 14/02/26 14:33:17 INFO TaskSetManager: Starting task 996.0:98 as TID 1795
> on executor 9: node02 (PROCESS_LOCAL)
> 14/02/26 14:33:17 INFO TaskSetManager: Serialized task 996.0:98 as 4070
> bytes in 1 ms
> 14/02/26 14:33:17 WARN TaskSetManager: Lost TID 1492 (task 996.0:17)
>
>
> and then...
>
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1649 (task 996.0:115)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1666 (task 996.0:32)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1675 (task 996.0:160)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1657 (task 996.0:349)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1660 (task 996.0:141)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1651 (task 996.0:55)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1669 (task 996.0:126)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1678 (task 996.0:173)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1663 (task 996.0:128)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1672 (task 996.0:28)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1654 (task 996.0:96)
> 14/02/26 14:33:20 WARN TaskSetManager: Lost TID 1699 (task 996.0:294)
> 14/02/26 14:33:20 INFO DAGScheduler: Executor lost: 12 (epoch 16)
> 14/02/26 14:33:20 INFO BlockManagerMasterActor: Trying to remove executor
> 12 from BlockManagerMaster.
> 14/02/26 14:33:20 INFO BlockManagerMaster: Removed 12 successfully in
> removeExecutor
> 14/02/26 14:33:20 INFO Stage: Stage 996 is now unavailable on executor 12
> (0/379, false)
>
>
> which looks like warnings..
>
>
> The code I tried to run was:
> subs_count = complex_key.map( lambda x:
> (x[0],int(x[1])).reduceByKey(lambda a,b:a+b))
> subs_count.take(20)
>
> Thanks
>
> --
> Mohit
>
> "When you want success as badly as you want the air, then you will get it.
> There is no other secret of success."
> -Socrates
>
