exception - 应用程序运行一段时间后Pyspark套接字超时异常

Question

我正在使用 pyspark 来估计逻辑回归模型的参数。我使用spark来计算可能性和梯度，然后使用scipy的最小化函数进行优化(L-BFGS-B)。

我使用 yarn 客户端模式来运行我的应用程序。我的应用程序可以毫无问题地开始运行。然而，一段时间后，它报告以下错误:

Traceback (most recent call last):
  File "/home/panc/research/MixedLogistic/software/mixedlogistic/mixedlogistic_spark/simulation/20160716-1626/spark_1m_data.py", line 115, in <module>
    res = trainEM2(distData, params0, verbose=True, em_tol=1e-5, opt_method='L-BFGS-B')
  File "/home/panc/research/MixedLogistic/software/mixedlogistic/mixedlogistic_spark/Train2.py", line 166, in trainEM
    options={'disp': False})
  File "/apps/rhel6/Anaconda-2.0.1/lib/python2.7/site-packages/scipy/optimize/_minimize.py", line 447, in minimize
    callback=callback, **options)
  File "/apps/rhel6/Anaconda-2.0.1/lib/python2.7/site-packages/scipy/optimize/lbfgsb.py", line 330, in _minimize_lbfgsb
    f, g = func_and_grad(x)
  File "/apps/rhel6/Anaconda-2.0.1/lib/python2.7/site-packages/scipy/optimize/lbfgsb.py", line 278, in func_and_grad
    f = fun(x, *args)
  File "/apps/rhel6/Anaconda-2.0.1/lib/python2.7/site-packages/scipy/optimize/optimize.py", line 289, in function_wrapper
    return function(*(wrapper_args + args))
  File "/home/panc/research/MixedLogistic/software/mixedlogistic/mixedlogistic_spark/Train2.py", line 146, in fun_observedQj
    return dataAndWeightsj_old.map(lambda _: calObservedQj(_[0], _[1], vparamsj, params0)).sum()
  File "/apps/hathi/spark-1.6.2/python/pyspark/rdd.py", line 995, in sum
    return self.mapPartitions(lambda x: [sum(x)]).fold(0, operator.add)
  File "/apps/hathi/spark-1.6.2/python/pyspark/rdd.py", line 869, in fold
    vals = self.mapPartitions(func).collect()
  File "/apps/hathi/spark-1.6.2/python/pyspark/rdd.py", line 772, in collect
    return list(_load_from_socket(port, self._jrdd_deserializer))
  File "/apps/hathi/spark-1.6.2/python/pyspark/rdd.py", line 142, in _load_from_socket
    for item in serializer.load_stream(rf):
  File "/apps/hathi/spark-1.6.2/python/pyspark/serializers.py", line 139, in load_stream
16/07/16 20:59:10 ERROR python.PythonRDD: Error while sending iterator
java.net.SocketTimeoutException: Accept timed out
    at java.net.PlainSocketImpl.socketAccept(Native Method)
    at java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:409)
    at java.net.ServerSocket.implAccept(ServerSocket.java:545)
    at java.net.ServerSocket.accept(ServerSocket.java:513)
    at org.apache.spark.api.python.PythonRDD$$anon$2.run(PythonRDD.scala:645)
    yield self._read_with_length(stream)
  File "/apps/hathi/spark-1.6.2/python/pyspark/serializers.py", line 156, in _read_with_length
    length = read_int(stream)
  File "/apps/hathi/spark-1.6.2/python/pyspark/serializers.py", line 543, in read_int
    length = stream.read(4)
  File "/apps/rhel6/Anaconda-2.0.1/lib/python2.7/socket.py", line 384, in read
    data = self._sock.recv(left)
socket.timeout: timed out

当我将 Spark 日志级别设置为“ALL”时，我还发现了 python Broken Pipe 错误。

我使用的是 Spark 1.6.2 和 Java 1.8.0_91。知道发生了什么吗？

--更新--

我发现这与我在程序中使用的优化例程有关。

我所做的是使用 EM 算法(作为迭代算法)通过最大似然法估计统计模型。在每次迭代期间，我需要通过解决最小化问题来更新参数。 Spark 负责计算我的可能性和梯度，然后将其传递给 Scipy 的最小化例程，我在其中使用 L-BFGS-B 方法。看来这个例程中的某些东西使我的 Spark 工作崩溃了。但我不知道例程的哪一部分导致了这个问题。

另一个观察结果是，在使用相同的示例和相同的程序时，我更改了分区的数量。当分区数量较小时，我的程序可以毫无问题地完成。然而，当分区数量变大时，程序开始崩溃。

Answer 1

我也遇到过类似的问题。我进行了一次迭代，有时执行时间太长以至于超时。增加 Spark.executor.heartbeatInterval 似乎可以解决问题。我将其增加到 3600 秒，以确保我不会再次遇到超时，并且从那时起一切都工作正常。

来自:http://spark.apache.org/docs/latest/configuration.html :

spark.executor.heartbeatInterval 10s Interval between each executor's heartbeats to the driver. Heartbeats let the driver know that the executor is still alive and update it with metrics for in-progress tasks.