CODEWITHSUNDEEP
pythonapache-sparkversioncluster-computing
user1780424
user1780424

Reputation: 413

Spark Exception: Python in worker has different version 3.4 than that in driver 3.5

I am using Amazon EC2, and I have my master and development servers as one. And I have another instance for a single worker.

I am new to this, but I have managed to make spark work in a standalone mode. Now I am trying cluster. the master and worker are active (I can see the webUI for them and they are functioning).

I have Spark 2.0, and I have installed the latest Anaconda 4.1.1 which comes with Python 3.5.2. In both worker and master, if I go to pyspark and do os.version_info, I will get the 3.5.2, I also have set all the environment variables correctly (as seen in other posts on stackoverflow and google) (e.g., PYSPARK_PYTHON).

There is no 3.4 version of python anywhere anyways. So I am wondering how I can fix this.

I get the error by running this command:

rdd = sc.parallelize([1,2,3])
rdd.count()

error happens for the count() method:

16/08/13 18:44:31 ERROR Executor: Exception in task 1.0 in stage 2.0 (TID 17)
org.apache.spark.api.python.PythonException: Traceback (most recent call last):
  File "/opt/spark/python/lib/pyspark.zip/pyspark/worker.py", line 123, in main
    ("%d.%d" % sys.version_info[:2], version))
Exception: Python in worker has different version 3.4 than that in driver 3.5, PySpark cannot run with different minor versions

at org.apache.spark.api.python.PythonRunner$$anon$1.read(PythonRDD.scala:193)
at org.apache.spark.api.python.PythonRunner$$anon$1.<init>(PythonRDD.scala:234)
at org.apache.spark.api.python.PythonRunner.compute(PythonRDD.scala:152)
at org.apache.spark.api.python.PythonRDD.compute(PythonRDD.scala:63)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:319)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:283)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:70)
at org.apache.spark.scheduler.Task.run(Task.scala:85)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:274)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
16/08/13 18:44:31 ERROR Executor: Exception in task 1.1 in stage 2.0 (TID 18)
org.apache.spark.api.python.PythonException: Traceback (most recent call last):
  File "/opt/spark/python/lib/pyspark.zip/pyspark/worker.py", line 123, in main
    ("%d.%d" % sys.version_info[:2], version))
Exception: Python in worker has different version 3.4 than that in driver 3.5, PySpark cannot run with different minor versions
at org.apache.spark.api.python.PythonRunner$$anon$1.read(PythonRDD.scala:193)
at org.apache.spark.api.python.PythonRunner$$anon$1.<init>(PythonRDD.scala:234)
at org.apache.spark.api.python.PythonRunner.compute(PythonRDD.scala:152)
at org.apache.spark.api.python.PythonRDD.compute(PythonRDD.scala:63)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:319)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:283)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:70)
at org.apache.spark.scheduler.Task.run(Task.scala:85)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:274)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)

Upvotes: 2

Views: 30371

Answers (3)

Kolaaa
Kolaaa

Reputation: 256

Just install the required python version and update it in your environment

Upvotes: -3

Shyu Kevin
Shyu Kevin

Reputation: 53

I've had the same issue. There are several possible reasons:

  • One of your workers had python3.4 but you didn't notice it. Please go to each worker to check PYSPARK_PYTHON. That is, if you set PYSPARK_PYTHON=python3, go to each worker to type python3 and check the version.

  • You connect to the wrong workers. Check your SparkContext config and make sure what your workers are.

I have spent more than ten hours to find the same issue with you. But the root cause for me is the wrong connection.... TAT

Upvotes: 1

zero323
zero323

Reputation: 330113

Since you already use Anaconda you can simply create an environment with the desired Python version:

conda create --name foo python=3.4
source activate foo

python --version
## Python 3.4.5 :: Continuum Analytics, Inc

and use it as PYSPARK_DRIVER_PYTHON:

export PYSPARK_DRIVER_PYTHON=/path/to/anaconda/envs/foo/bin/python

Upvotes: 10

Related Questions