You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Bryan Cutler (JIRA)" <ji...@apache.org> on 2019/04/23 18:24:00 UTC

[jira] [Created] (SPARK-27548) PySpark toLocalIterator does not raise errors from worker

Bryan Cutler created SPARK-27548:
------------------------------------

             Summary: PySpark toLocalIterator does not raise errors from worker
                 Key: SPARK-27548
                 URL: https://issues.apache.org/jira/browse/SPARK-27548
             Project: Spark
          Issue Type: Bug
          Components: PySpark
    Affects Versions: 2.4.1
            Reporter: Bryan Cutler


When using a PySpark RDD local iterator and an error occurs on the worker, it is not picked up by Py4J and raised in the Python driver process. So unless looking at logs, there is no way for the application to know the worker had an error. This is a test that should pass if the error is raised in the driver:
{code}
def test_to_local_iterator_error(self):

    def fail(_):
        raise RuntimeError("local iterator error")

    rdd = self.sc.parallelize(range(10)).map(fail)

    with self.assertRaisesRegexp(Exception, "local iterator error"):
        for _ in rdd.toLocalIterator():
            pass{code}
but it does not raise an exception:
{noformat}
Caused by: org.apache.spark.api.python.PythonException: Traceback (most recent call last):
  File "/home/bryan/git/spark/python/lib/pyspark.zip/pyspark/worker.py", line 428, in main
    process()
  File "/home/bryan/git/spark/python/lib/pyspark.zip/pyspark/worker.py", line 423, in process
    serializer.dump_stream(func(split_index, iterator), outfile)
  File "/home/bryan/git/spark/python/lib/pyspark.zip/pyspark/serializers.py", line 505, in dump_stream
    vs = list(itertools.islice(iterator, batch))
  File "/home/bryan/git/spark/python/lib/pyspark.zip/pyspark/util.py", line 99, in wrapper
    return f(*args, **kwargs)
  File "/home/bryan/git/spark/python/pyspark/tests/test_rdd.py", line 742, in fail
    raise RuntimeError("local iterator error")
RuntimeError: local iterator error

    at org.apache.spark.api.python.BasePythonRunner$ReaderIterator.handlePythonException(PythonRunner.scala:453)
...
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)
FAIL

======================================================================
FAIL: test_to_local_iterator_error (pyspark.tests.test_rdd.RDDTests)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/bryan/git/spark/python/pyspark/tests/test_rdd.py", line 748, in test_to_local_iterator_error
    pass
AssertionError: Exception not raised{noformat}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org