Python Hadoop Streaming Error“ERROR streaming.StreamJob：Job not successful！”和堆栈跟踪：ExitCodeException exitCode = 134

Question

我正在尝试使用Hadoop Streaming在Hadoop集群上运行python脚本以进行情绪分析。 我正在运行正常运行并提供输出的本地计算机上运行的相同脚本。
要在本地计算机上运行，​​我使用此命令。

$ cat /home/MB/analytics/Data/input/* | ./new_mapper.py

并且在hadoop集群上运行我使用下面的命令

$ hadoop jar /usr/lib/hadoop-0.20-mapreduce/contrib/streaming/hadoop-streaming-2.5.0-mr1-cdh5.2.0.jar -mapper "python $PWD/new_mapper.py" -reducer "$PWD/new_reducer.py" -input /user/hduser/Test_04012015_Data/input/* -output /user/hduser/python-mr/out-mr-out

我的脚本的示例代码是

#!/usr/bin/env python
import sys


def main(argv):
##    for line in sys.stdin:
##        print line
    for line in sys.stdin:
        line = line.split(',')
        t_text      = re.sub(r'[?|$|.|!|,|!|?|;]',r'',line[7])
        words    = re.findall(r"[\w']+", t_text.rstrip())
        predicted = classifier.classify(feature_select(words))
        i=i+1
        referenceSets[predicted].add(i)
        testSets[predicted].add(i)
        print line[7] +'\t'+predicted

if __name__ == "__main__":
    main(sys.argv)

Exception的堆栈跟踪是：

    15/04/22 12:55:14 INFO mapreduce.Job: Task Id : attempt_1429611942931_0010_m_000001_0, Status : FAILED
    Error: java.io.IOException: Stream closed at java.lang.ProcessBuilder$NullOutputStream.write(ProcessBuilder.java:434)
    ...

    Exit code: 134
    Exception message: /bin/bash: line 1:  1691 Aborted 
(core dumped) /usr/lib/jvm/java-7-oracle-cloudera/bin/java
-Djava.net.preferIPv4Stack=true -Dhadoop.metrics.log.level=WARN -Djava.net.preferIPv4Stack=true -Xmx525955249
-Djava.io.tmpdir=/yarn/nm/usercache/hduser/appcache/application_1429611942931_0010/container_1429611942931_0010_01_000016/tmp
-Dlog4j.configuration=container-log4j.properties
-Dyarn.app.container.log.dir=/var/log/hadoop-yarn/container/application_1429611942931_0010/container_1429611942931_0010_01_000016 -Dyarn.app.container.log.filesize=0 
-Dhadoop.root.logger=INFO,CLA org.apache.hadoop.mapred.YarnChild 192.168.0.122 48725 attempt_1429611942931_0010_m_000006_1 16 > /var/log/hadoop-yarn/container/application_1429611942931_0010/container_1429611942931_0010_01_000016/stdout 2> /var/log/hadoop-yarn/container/application_1429611942931_0010/container_1429611942931_0010_01_000016/stderr
    ....

    15/04/22 12:55:47 ERROR streaming.StreamJob: Job not Successful!
    Streaming Command Failed!

我试图查看日志但是在色调中它显示了我的错误。 请建议我，出了什么问题。

Answer 1

您似乎忘了将文件new_mapper.py添加到您的工作中。

基本上，您的作业尝试运行python脚本new_mapper.py，但运行映射器的服务器上缺少此脚本。

您必须使用选项-file <local_path_to_your_file>将此文件添加到作业中。

请参阅此处的文档和示例：https://hadoop.apache.org/docs/stable/hadoop-mapreduce-client/hadoop-mapreduce-client-core/HadoopStreaming.html#Streaming_Command_Options