pyspark java.net.SocketException: Connection reset by peer

比眉伴天荒 2023-10-18 13:58 228阅读 0赞

在window、运行pyspark训练模型,报错
Caused by: java.net.SocketException: Connection reset by peer: socket write error

  1. py4j.protocol.Py4JJavaError: An error occurred while calling z:org.apache.spark.api.python.PythonRDD.runJob.
  2. : org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 1.0 failed 1 times, most recent failure: Lost task 0.0 in stage 1.0 (TID 1, localhost, executor driver): java.net.SocketException: Connection reset by peer: socket write error
  3. at java.net.SocketOutputStream.socketWrite0(Native Method)
  4. at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:109)
  5. at java.net.SocketOutputStream.write(SocketOutputStream.java:153)
  6. ...
  7. Caused by: java.net.SocketException: Connection reset by peer: socket write error
  8. at java.net.SocketOutputStream.socketWrite0(Native Method)
  9. at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:109)
  10. at java.net.SocketOutputStream.write(SocketOutputStream.java:153)
  11. ...
解决:修改python环境下的pyspark包中的worker. py 中的process函数

(注:python环境下的pyspark包是从spark目录拷贝过来的)
eg: E:\Anaconda3\envs\py27\Lib\site-packages\pyspark\worker. py
定位到 process 函数:

  1. def process():
  2. iterator = deserializer.load_stream(infile)
  3. serializer.dump_stream(func(split_index, iterator), outfile)
  4. # 添加以下内容
  5. for obj in interator:
  6. pass

refer:
https://issues.apache.org/jira/browse/SPARK-12261
https://stackoverflow.com/questions/26954566/apache-spark-pyspark-crash-for-large-dataset


end

发表评论

表情:
评论列表 (有 0 条评论,228人围观)

还没有评论,来说两句吧...

相关阅读