火花在本地运行，但在纱线运行时找不到文件

Question

我一直试图提交一个简单的python脚本，以便在一个带有Yarn的集群中运行它。当我在本地执行作业时，没有问题，一切都很好，但当我在集群中运行它时，它就失败了。

17/11/07 13:02:48 INFO yarn.Client: Application report for application_1510046813642_0010 (state: ACCEPTED)
17/11/07 13:02:49 INFO yarn.Client: Application report for application_1510046813642_0010 (state: ACCEPTED)
17/11/07 13:02:50 INFO yarn.Client: Application report for application_1510046813642_0010 (state: FAILED)
17/11/07 13:02:50 INFO yarn.Client: 
     client token: N/A
     diagnostics: Application application_1510046813642_0010 failed 2 times due to AM Container for appattempt_1510046813642_0010_000002 exited with  exitCode: -1000
For more detailed output, check application tracking page:http://myserver:8088/proxy/application_1510046813642_0010/Then, click on links to logs of each attempt.
**Diagnostics: File does not exist: hdfs://myserver:8020/user/josholsan/.sparkStaging/application_1510046813642_0010/test.py**
java.io.FileNotFoundException: File does not exist: hdfs://myserver:8020/user/josholsan/.sparkStaging/application_1510046813642_0010/test.py
    at org.apache.hadoop.hdfs.DistributedFileSystem$20.doCall(DistributedFileSystem.java:1266)
    at org.apache.hadoop.hdfs.DistributedFileSystem$20.doCall(DistributedFileSystem.java:1258)
    at org.apache.hadoop.fs.FileSystemLinkResolver.resolve(FileSystemLinkResolver.java:81)
    at org.apache.hadoop.hdfs.DistributedFileSystem.getFileStatus(DistributedFileSystem.java:1258)
    at org.apache.hadoop.yarn.util.FSDownload.copy(FSDownload.java:251)
    at org.apache.hadoop.yarn.util.FSDownload.access$000(FSDownload.java:61)
    at org.apache.hadoop.yarn.util.FSDownload$2.run(FSDownload.java:359)
    at org.apache.hadoop.yarn.util.FSDownload$2.run(FSDownload.java:357)
    at java.security.AccessController.doPrivileged(Native Method)
    at javax.security.auth.Subject.doAs(Subject.java:422)
    at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1917)
    at org.apache.hadoop.yarn.util.FSDownload.call(FSDownload.java:356)
    at org.apache.hadoop.yarn.util.FSDownload.call(FSDownload.java:60)
    at java.util.concurrent.FutureTask.run(FutureTask.java:266)
    at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
    at java.util.concurrent.FutureTask.run(FutureTask.java:266)
    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)

Failing this attempt. Failing the application.
     ApplicationMaster host: N/A
     ApplicationMaster RPC port: -1
     queue: root.users.josholsan
     start time: 1510056155796
     final status: FAILED
     tracking URL: http://myserver:8088/cluster/app/application_1510046813642_0010
     user: josholsan
Exception in thread "main" org.apache.spark.SparkException: Application application_1510046813642_0010 finished with failed status
    at org.apache.spark.deploy.yarn.Client.run(Client.scala:1025)
    at org.apache.spark.deploy.yarn.Client$.main(Client.scala:1072)
    at org.apache.spark.deploy.yarn.Client.main(Client.scala)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:730)
    at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:181)
    at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:206)
    at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:121)
    at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)
17/11/07 13:02:50 INFO util.ShutdownHookManager: Shutdown hook called
17/11/07 13:02:50 INFO util.ShutdownHookManager: Deleting directory /tmp/spark-5cc8bf5e-216b-4d9e-b66d-9dc01a94e851

诊断：文件不存在:hdfs://myserver:8020/user/josholsan/.sparkstaging/application_1510046813642_0010/test.py

我不知道为什么它找不到test.py，我也试着把它放在HDFS中执行作业的用户目录下：/user/josholsan/

为了完成我的文章，我还想分享我的test.py脚本：

from pyspark import SparkContext

file="/user/josholsan/concepts_copy.csv"
sc = SparkContext("local","Test app")
textFile = sc.textFile(file).cache()

linesWithOMOP=textFile.filter(lambda line: "OMOP" in line).count()
linesWithICD=textFile.filter(lambda line: "ICD" in line).count()

print("Lines with OMOP: %i, lines with ICD9: %i" % (linesWithOMOP,linesWithICD))

Answer 1

转自评论科：

>

sc=SparkContext（“local”,“test app”）:在这里使用“local”将覆盖任何命令行设置；从文档中：

指定为标志或在属性文件中的任何值都将传递到应用程序，并与通过Sparkconf指定的值合并。直接在SparkConf上设置的属性优先，然后是传递给spark-submit或spark-shell的标志，然后是spark-defaults.conf文件中的选项。

可以使用--py-files参数指定任何其他文件和资源（在mesos中测试，不幸的是，不在yarn中测试），例如--py-files http://somewhere/accessible/to/all/extra_python_code_my_code_uses.zip

编辑：正如@DesertNaut评论的那样，这个参数应该在脚本执行之前使用。

希望这对你有帮助，祝你好运！