jiangyu created ARROW-6011: ------------------------------ Summary: Data incomplete when using pyarrow in pyspark in python 3.x Key: ARROW-6011 URL: https://issues.apache.org/jira/browse/ARROW-6011 Project: Apache Arrow Issue Type: Bug Components: Java, Python Affects Versions: 0.14.0, 0.10.0 Environment: ceonts 7.4 pyarrow 0.10.0 0.14.0 python 2.7 3.5 3.6 Reporter: jiangyu Attachments: image-2019-07-23-16-06-49-889.png
Hi, In spark 2.3, pandas udf add to pyspark and pyarrow as a default serialization and deserialization method. It is a great feature, and we use it a lot. But , when we change the default python version from 2.7 to 3.5 or 3.6 ( conda as python envs manager), we encounter a fatal problem. We use pandas udf to process batches of data, but we find the data is incompelted. At first , i think the process logical maybe wrong, so i change the code to very simple one and it has the same problem.After investigate for a week, i find it is related to pyarrow. Reproduce it: Below is how to reproduce it: 1.generate data first generate a very simple data, the data have seven column, a、b、c、d、e、f and g, every row is the same,data type is Integer a,b,c,d,e,f,g 1,2,3,4,5,6,7 1,2,3,4,5,6,7 1,2,3,4,5,6,7 1,2,3,4,5,6,7 we can produce 100,000 rows and name the file test.csv upload to hdfs, then load it , and repartition it to 1 partition. df=spark.read.format('csv').option("header","true").load('/test.csv') df=df.select(*(col(c).cast("int").alias(c) for c in df.columns)) df=df.repartition(1) spark_context = SparkContext.getOrCreate() 2.register pandas udf make a very simple pandas udf function and register it.The function is very simple , just print “iterator one time” and do nothing then return. def add_func(a,b,c,d,e,f,g): print('iterator one time') return a add = pandas_udf(add_func, returnType=IntegerType()) df_result=df.select(add(col("a"),col("b"),col("c"),col("d"),col("e"),col("f"),col("g"))) 3.trigger spark to action def trigger_func(iterator): yield iterator df_result.rdd.foreachPartition(trigger_func) 4.execute it in pyspark (local or yarn) we set spark.sql.execution.arrow.maxRecordsPerBatch=100000, and the rows is 1,000,000 , so it is should print “iterator one time” for 10 times. (1)Here is result in python 2.7 envs: PYSPARK_PYTHON=/usr/lib/conda/envs/py2.7/bin/python pyspark --conf spark.sql.execution.arrow.maxRecordsPerBatch=100000 --conf spark.executor.pyspark.memory=2g --conf spark.sql.execution.arrow.enabled=true --executor-cores 1 !image-2019-07-23-16-06-49-889.png! The result is right, 10 times of print. (2)Then change to python 3.6 envs,with the same code. PYSPARK_PYTHON=/usr/lib/conda/envs/python3.6/bin/python pyspark --conf spark.sql.execution.arrow.maxRecordsPerBatch=100000 --conf spark.executor.pyspark.memory=2g --conf spark.sql.execution.arrow.enabled=true --executor-cores 1!0pPjMgKKgEJSEACEpCABCQgAQlIQAISkMCcCaismHPrWjcJSEACEpCABCQgAQlIQAISkMAJJKCy4gQ2miJLQAISkIAEJCABCUhAAhKQgATmTEBlxZxb17pJQAISkIAEJCABCUhAAhKQgAROIAGVFSew0RRZAhKQgAQkIAEJSEACEpCABCQwZwIqK bcutZNAhKQgAQkIAEJSEACEpCABCRwAgmorDiBjabIEpCABCQgAQlIQAISkIAEJCCBORP4B5QvwTqM1wfyAAAAAElFTkSuQmCC! The data is incomplete. The exception is print by spark which have been added by us , I will explain it later. h3. Investigation So i just add some log to trace it. The “process done” is added in the worker.py. !Ae0YTBna66oMAAAAAElFTkSuQmCC! In order to get the exception, we also change the spark code, the code is under core/src/main/scala/org/apache/spark/util/Utils.scala , and add this code to print the exception. @@ -1362,6 +1362,8 @@ private[spark] object Utils extends Logging { case t: Throwable => // Purposefully not using NonFatal, because even fatal exceptions // we don't want to have our finallyBlock suppress + logInfo(t.getLocalizedMessage) + t.printStackTrace() originalThrowable = t throw originalThrowable } finally { It seems the pyspark get the data from jvm , but pyarrow get the data incomplete. Pyarrow side think the data is finished, then shutdown the socket. At the same time, the jvm side still writes to the same socket , but get socket close exception. The pyarrow part is in ipc.pxi: cdef class _RecordBatchReader: cdef: shared_ptr[CRecordBatchReader] reader shared_ptr[InputStream] in_stream cdef readonly: Schema schema def __cinit__(self): pass def _open(self, source): get_input_stream(source, &self.in_stream) with nogil: check_status(CRecordBatchStreamReader.Open( self.in_stream.get(), &self.reader)) self.schema = pyarrow_wrap_schema(self.reader.get().schema()) def __iter__(self): while True: yield self.read_next_batch() def get_next_batch(self): import warnings warnings.warn('Please use read_next_batch instead of ' 'get_next_batch', FutureWarning) return self.read_next_batch() def read_next_batch(self): """ Read next RecordBatch from the stream. Raises StopIteration at end of stream """ cdef shared_ptr[CRecordBatch] batch with nogil: check_status(self.reader.get().ReadNext(&batch)) if batch.get() == NULL: raise StopIteration return pyarrow_wrap_batch(batch)read_next_batch function get NULL, think the iterator is over. h3. RESULT Our environment is spark 2.4.3, we have tried pyarrow version 0.10.0 and 0.14.0 , python version is python 2.7, python 3.5, python 3.6. When using python 2.7, everything is fine. But when change to python 3.5,3,6, the data is wrong. The column number is critical to trigger this bug, if column number is less than 5 , this bug probably will not happen. But If the column number is big , for example 7 or above, it will happened every time. So we wonder if there is some conflict between python 3.x and pyarrow version? I have put the code and data as attachment. -- This message was sent by Atlassian JIRA (v7.6.14#76016)