[
https://issues.apache.org/jira/browse/SPARK-31911?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17130164#comment-17130164
]
Brandon commented on SPARK-31911:
---------------------------------
Some more notes:
My Spark app is configured with
`spark.hadoop.mapreduce.fileoutputcommitter.algorithm.version 2`. This seems
relevant because the PendingSet files are managed with a wrapped
FileOutputCommitter. Using v2 could exacerbate the problem, because the pending
files are becoming visible to the driver immediately when a task finishes. Even
so, if my understanding is correct, v1 would still have a race condition, but
may present differently (not sure if worse or better).
An additional symptom of this bug is that data-writing tasks randomly fail with
exceptions such as
"org.apache.hadoop.ipc.RemoteException(java.io.FileNotFoundException): File
does not exist" while the task is committing its work. Of course, since spark
attempts tasks 4 times by default, it usually gets through, but slows things
down. This exception seems to be another symptom of the PendingSet files being
cleaned up by the driver while tasks are still working on it?
{noformat}
20/06/10 04:45:36 WARN TaskSetManager: Lost task 7.0 in stage 15.0 (TID 4399,
172.17.252.245, executor 1): org.apache.spark.SparkException: Task failed while
writing rows.20/06/10 04:45:36 WARN TaskSetManager: Lost task 7.0 in stage 15.0
(TID 4399, 172.17.252.245, executor 1): org.apache.spark.SparkException: Task
failed while writing rows. at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.executeTask(FileFormatWriter.scala:257)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.$anonfun$write$15(FileFormatWriter.scala:177)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90) at
org.apache.spark.scheduler.Task.run(Task.scala:123) at
org.apache.spark.executor.Executor$TaskRunner.$anonfun$run$3(Executor.scala:411)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360) at
org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:414) at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)Caused by:
org.apache.hadoop.ipc.RemoteException(java.io.FileNotFoundException): File does
not exist:
/tmp/s3a-test-staging/www-data/hooray/staging-uploads/_temporary/0/_temporary/attempt_20200610044421_0015_m_000007_4399
(inode 64406205) Holder DFSClient_NONMAPREDUCE_330704236_69 does not have any
open files. at
org.apache.hadoop.hdfs.server.namenode.FSNamesystem.checkLease(FSNamesystem.java:2800)
at
org.apache.hadoop.hdfs.server.namenode.FSDirWriteFileOp.completeFileInternal(FSDirWriteFileOp.java:691)
at
org.apache.hadoop.hdfs.server.namenode.FSDirWriteFileOp.completeFile(FSDirWriteFileOp.java:677)
at
org.apache.hadoop.hdfs.server.namenode.FSNamesystem.completeFile(FSNamesystem.java:2843)
at
org.apache.hadoop.hdfs.server.namenode.NameNodeRpcServer.complete(NameNodeRpcServer.java:928)
at
org.apache.hadoop.hdfs.protocolPB.ClientNamenodeProtocolServerSideTranslatorPB.complete(ClientNamenodeProtocolServerSideTranslatorPB.java:607)
at
org.apache.hadoop.hdfs.protocol.proto.ClientNamenodeProtocolProtos$ClientNamenodeProtocol$2.callBlockingMethod(ClientNamenodeProtocolProtos.java)
at
org.apache.hadoop.ipc.ProtobufRpcEngine$Server$ProtoBufRpcInvoker.call(ProtobufRpcEngine.java:523)
at org.apache.hadoop.ipc.RPC$Server.call(RPC.java:991) at
org.apache.hadoop.ipc.Server$RpcCall.run(Server.java:872) at
org.apache.hadoop.ipc.Server$RpcCall.run(Server.java:818) at
java.security.AccessController.doPrivileged(Native Method) at
javax.security.auth.Subject.doAs(Subject.java:422) at
org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1729)
at org.apache.hadoop.ipc.Server$Handler.run(Server.java:2678)
at org.apache.hadoop.ipc.Client.getRpcResponse(Client.java:1545) at
org.apache.hadoop.ipc.Client.call(Client.java:1491) at
org.apache.hadoop.ipc.Client.call(Client.java:1388) at
org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:233)
at
org.apache.hadoop.ipc.ProtobufRpcEngine$Invoker.invoke(ProtobufRpcEngine.java:118)
at com.sun.proxy.$Proxy28.complete(Unknown Source) at
org.apache.hadoop.hdfs.protocolPB.ClientNamenodeProtocolTranslatorPB.complete(ClientNamenodeProtocolTranslatorPB.java:557)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498) at
org.apache.hadoop.io.retry.RetryInvocationHandler.invokeMethod(RetryInvocationHandler.java:422)
at
org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invokeMethod(RetryInvocationHandler.java:165)
at
org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invoke(RetryInvocationHandler.java:157)
at
org.apache.hadoop.io.retry.RetryInvocationHandler$Call.invokeOnce(RetryInvocationHandler.java:95)
at
org.apache.hadoop.io.retry.RetryInvocationHandler.invoke(RetryInvocationHandler.java:359)
at com.sun.proxy.$Proxy29.complete(Unknown Source) at
org.apache.hadoop.hdfs.DFSOutputStream.completeFile(DFSOutputStream.java:949)
at
org.apache.hadoop.hdfs.DFSOutputStream.completeFile(DFSOutputStream.java:907)
at org.apache.hadoop.hdfs.DFSOutputStream.closeImpl(DFSOutputStream.java:890)
at org.apache.hadoop.hdfs.DFSOutputStream.close(DFSOutputStream.java:845) at
org.apache.hadoop.fs.FSDataOutputStream$PositionCache.close(FSDataOutputStream.java:72)
at org.apache.hadoop.fs.FSDataOutputStream.close(FSDataOutputStream.java:101)
at
org.apache.hadoop.util.JsonSerialization.writeJsonAsBytes(JsonSerialization.java:282)
at org.apache.hadoop.util.JsonSerialization.save(JsonSerialization.java:269)
at org.apache.hadoop.fs.s3a.commit.files.PendingSet.save(PendingSet.java:170)
at
org.apache.hadoop.fs.s3a.commit.staging.StagingCommitter.commitTaskInternal(StagingCommitter.java:724)
at
org.apache.hadoop.fs.s3a.commit.staging.StagingCommitter.commitTask(StagingCommitter.java:641)
at
org.apache.spark.mapred.SparkHadoopMapRedUtil$.performCommit$1(SparkHadoopMapRedUtil.scala:50)
at
org.apache.spark.mapred.SparkHadoopMapRedUtil$.commitTask(SparkHadoopMapRedUtil.scala:77)
at
org.apache.spark.internal.io.HadoopMapReduceCommitProtocol.commitTask(HadoopMapReduceCommitProtocol.scala:245)
at
org.apache.spark.sql.execution.datasources.FileFormatDataWriter.commit(FileFormatDataWriter.scala:78)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.$anonfun$executeTask$1(FileFormatWriter.scala:247)
at
org.apache.spark.util.Utils$.tryWithSafeFinallyAndFailureCallbacks(Utils.scala:1394)
at
org.apache.spark.sql.execution.datasources.FileFormatWriter$.executeTask(FileFormatWriter.scala:242)
... 9 more{noformat}
Finally, I've confirmed that setting `spark.sql.sources.writeJobUUID` to a
random string for each write job fixes all issues for me. For example,
{noformat}
dataframe.write.option("spark.sql.sources.writeJobUUID",
generateRandomUUID()).save("s3a://..."){noformat}
With this change, I see no issues at all, it works beautifully. Maybe it's
supposed to be set in spark too?
> Using S3A staging committer, pending uploads are committed more than once and
> listed incorrectly in _SUCCESS data
> -----------------------------------------------------------------------------------------------------------------
>
> Key: SPARK-31911
> URL: https://issues.apache.org/jira/browse/SPARK-31911
> Project: Spark
> Issue Type: Bug
> Components: Input/Output
> Affects Versions: 2.4.4
> Reporter: Brandon
> Priority: Major
>
> First of all thanks for the great work on the S3 committers. I was able set
> up the directory staging committer in my environment following docs at
> [https://github.com/apache/spark/blob/master/docs/cloud-integration.md#committing-work-into-cloud-storage-safely-and-fast]
> and tested one of my Spark applications using it. The Spark version is 2.4.4
> with Hadoop 3.2.1 and the cloud committer bindings. The application writes
> multiple DataFrames to ORC/Parquet in S3, submitting them as write jobs to
> Spark in parallel.
> I think I'm seeing a bug where the staging committer will complete pending
> uploads more than once. The main symptom how I discovered this is that the
> _SUCCESS data files under each table will contain overlapping file names that
> belong to separate tables. From my reading of the code, that's because the
> filenames in _SUCCESS reflect which multipart uploads were completed in the
> commit for that particular table.
> An example:
> Concurrently, fire off DataFrame.write.orc("s3a://bucket/a") and
> DataFrame.write.orc("s3a://bucket/b"). Suppose each table has one partition
> so writes one partition file.
> When the two writes are done,
> * /a/_SUCCESS contains two filenames: /a/part-0000 and /b/part-0000.
> * /b/_SUCCESS contains the same two filenames.
> Setting S3A logs to debug, I see the commitJob operation belonging to table a
> includes completing the uploads of /a/part-0000 and /b/part-0000. Then again,
> commitJob for table b includes the same completions. I haven't had a problem
> yet, but I wonder if having these extra requests would become an issue at
> higher scale, where dozens of commits with hundreds of files may be happening
> concurrently in the application.
> I believe this may be caused from the way the pendingSet files are stored in
> the staging directory. They are stored under one directory named by the
> jobID, in the Hadoop code. However, for all write jobs executed by the Spark
> application, the jobID passed to Hadoop is the same - the application ID.
> Maybe the staging commit algorithm was built on the assumption that each
> instance of the algorithm would use a unique random jobID.
> [~ste...@apache.org] , [~rdblue] Having seen your names on most of this work
> (thank you), I would be interested to know your thoughts on this. Also it's
> my first time opening a bug here, so let me know if there's anything else I
> can do to help report the issue.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
