You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Vitaliy Migov (JIRA)" <ji...@apache.org> on 2014/11/04 17:10:38 UTC
[jira] [Commented] (SPARK-3958) Possible stream-corruption issues
in TorrentBroadcast
[ https://issues.apache.org/jira/browse/SPARK-3958?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14196288#comment-14196288 ]
Vitaliy Migov commented on SPARK-3958:
--------------------------------------
Observed the same exception ( Spark 1.1.0 ). After changing broadcast factory to the HttpBroadcastFactory ( as suggested in the SPARK-4133 ), exception looks like:
java.io.FileNotFoundException: http://10.8.0.22:44907/broadcast_0
Full logs attached to issue: spark_ex.logs
Seems that something wrong with the handling of the "broadcast_0" in the BlockManager:
I think that something wrong with the handing of the "broadcast_0" in the BlockManager:
14/11/04 17:20:39 INFO MemoryStore: Block broadcast_0 stored as values in memory (estimated size 1216.0 B, free 983.1 MB)
14/11/04 17:20:39 DEBUG BlockManager: Put block broadcast_0 locally took 84 ms
14/11/04 17:20:39 DEBUG BlockManager: Putting block broadcast_0 without replication took 86 ms
14/11/04 17:20:39 DEBUG BlockManager: Getting local block broadcast_0
14/11/04 17:20:39 DEBUG BlockManager: Level for block broadcast_0 is StorageLevel(true, true, false, true, 1)
14/11/04 17:20:39 DEBUG BlockManager: Getting block broadcast_0 from memory
14/11/04 17:20:39 INFO BlockManager: Found block broadcast_0 locally
14/11/04 17:20:57 WARN BlockManager: Block broadcast_0 already exists on this machine; not re-adding it
14/11/04 17:20:57 DEBUG BlockManager: Getting local block broadcast_0
14/11/04 17:20:57 DEBUG BlockManager: Block broadcast_0 not registered locally
14/11/04 17:20:57 DEBUG BlockManager: Getting remote block broadcast_0
14/11/04 17:20:57 DEBUG BlockManagerMasterActor: [actor] received message GetLocations(broadcast_0) from Actor[akka://sparkDriver/temp/$l]
14/11/04 17:20:57 DEBUG BlockManager: Block broadcast_0 not found
14/11/04 17:20:57 DEBUG BlockManagerMasterActor: [actor] handled message (0.117676 ms) GetLocations(broadcast_0) from Actor[akka://sparkDriver/temp/$l]
14/11/04 17:20:57 INFO HttpBroadcast: Started reading broadcast variable 0
14/11/04 17:20:57 DEBUG HttpBroadcast: broadcast read server: http://10.8.0.22:44907 id: broadcast-0
14/11/04 17:20:57 DEBUG HttpBroadcast: broadcast not using security
14/11/04 17:20:57 DEBUG RecurringTimer: Callback for BlockGenerator called at time 1415114457200
14/11/04 17:20:57 ERROR Executor: Exception in task 0.0 in stage 0.0 (TID 0)
java.io.FileNotFoundException: http://10.8.0.22:44907/broadcast_0
> Possible stream-corruption issues in TorrentBroadcast
> -----------------------------------------------------
>
> Key: SPARK-3958
> URL: https://issues.apache.org/jira/browse/SPARK-3958
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 1.1.0, 1.2.0
> Reporter: Josh Rosen
> Assignee: Josh Rosen
> Priority: Blocker
> Attachments: spark_ex.logs
>
>
> TorrentBroadcast deserialization sometimes fails with decompression errors, which are most likely caused by stream-corruption exceptions. For example, this can manifest itself as a Snappy PARSING_ERROR when deserializing a broadcasted task:
> {code}
> 14/10/14 17:20:55.016 DEBUG BlockManager: Getting local block broadcast_8
> 14/10/14 17:20:55.016 DEBUG BlockManager: Block broadcast_8 not registered locally
> 14/10/14 17:20:55.016 INFO TorrentBroadcast: Started reading broadcast variable 8
> 14/10/14 17:20:55.017 INFO TorrentBroadcast: Reading broadcast variable 8 took 5.3433E-5 s
> 14/10/14 17:20:55.017 ERROR Executor: Exception in task 2.0 in stage 8.0 (TID 18)
> java.io.IOException: PARSING_ERROR(2)
> at org.xerial.snappy.SnappyNative.throw_error(SnappyNative.java:84)
> at org.xerial.snappy.SnappyNative.uncompressedLength(Native Method)
> at org.xerial.snappy.Snappy.uncompressedLength(Snappy.java:594)
> at org.xerial.snappy.SnappyInputStream.readFully(SnappyInputStream.java:125)
> at org.xerial.snappy.SnappyInputStream.readHeader(SnappyInputStream.java:88)
> at org.xerial.snappy.SnappyInputStream.<init>(SnappyInputStream.java:58)
> at org.apache.spark.io.SnappyCompressionCodec.compressedInputStream(CompressionCodec.scala:128)
> at org.apache.spark.broadcast.TorrentBroadcast$.unBlockifyObject(TorrentBroadcast.scala:216)
> at org.apache.spark.broadcast.TorrentBroadcast.readObject(TorrentBroadcast.scala:170)
> at sun.reflect.GeneratedMethodAccessor92.invoke(Unknown Source)
> at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> at java.lang.reflect.Method.invoke(Method.java:606)
> at java.io.ObjectStreamClass.invokeReadObject(ObjectStreamClass.java:1017)
> at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1893)
> at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1798)
> at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1350)
> at java.io.ObjectInputStream.defaultReadFields(ObjectInputStream.java:1990)
> at java.io.ObjectInputStream.readSerialData(ObjectInputStream.java:1915)
> at java.io.ObjectInputStream.readOrdinaryObject(ObjectInputStream.java:1798)
> at java.io.ObjectInputStream.readObject0(ObjectInputStream.java:1350)
> at java.io.ObjectInputStream.readObject(ObjectInputStream.java:370)
> at org.apache.spark.serializer.JavaDeserializationStream.readObject(JavaSerializer.scala:62)
> at org.apache.spark.serializer.JavaSerializerInstance.deserialize(JavaSerializer.scala:87)
> at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:164)
> 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:745)
> {code}
> SPARK-3630 is an umbrella ticket for investigating all causes of these Kryo and Snappy deserialization errors. This ticket is for a more narrowly-focused exploration of the TorrentBroadcast version of these errors, since the similar errors that we've seen in sort-based shuffle seem to be explained by a different cause (see SPARK-3948).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org