[SPARK-34349] No python3 in docker images

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Cannot Reproduce
• Affects Version/s: 3.0.1
• Component/s: Kubernetes
• Assignee: Unassigned
• Reporter: Tim Hughes
• Created: 04/Feb/21 00:16
• Updated: 12/Dec/22 18:11
• Resolved: 06/Jul/21 02:08

Description

The spark-py container image doesn't receive the instruction to use python3 and defaults to python 2.7

 

The worker container was build using the following commands

mkdir ./tmp
wget -qO- https://www.mirrorservice.org/sites/ftp.apache.org/spark/spark-3.0.1/spark-3.0.1-bin-hadoop3.2.tgz | tar -C ./tmp/ -xzf -
cd ../spark-3.0.1-bin-hadoop3.2/
./bin/docker-image-tool.sh -r docker.io/timhughes -t spark-3.0.1-bin-hadoop3.2 -p kubernetes/dockerfiles/spark/bindings/python/Dockerfile build
docker push docker.io/timhughes/spark-py:spark-3.0.1-bin-hadoop3.2

 

This is the code I am using to initialize the workers

 

import os
from pyspark import SparkContext, SparkConf
from pyspark.sql import SparkSession# Create Spark config for our Kubernetes based cluster manager
sparkConf = SparkConf()
sparkConf.setMaster("k8s://https://kubernetes.default.svc.cluster.local:443")
sparkConf.setAppName("spark")
sparkConf.set("spark.kubernetes.container.image", "docker.io/timhughes/spark-py:spark-3.0.1-bin-hadoop3.2")
sparkConf.set("spark.kubernetes.namespace", "spark")
sparkConf.set("spark.executor.instances", "2")
sparkConf.set("spark.executor.cores", "1")
sparkConf.set("spark.driver.memory", "1024m")
sparkConf.set("spark.executor.memory", "1024m")
sparkConf.set("spark.kubernetes.pyspark.pythonVersion", "3")
sparkConf.set("spark.kubernetes.authenticate.driver.serviceAccountName", "spark")
sparkConf.set("spark.kubernetes.authenticate.serviceAccountName", "spark")
sparkConf.set("spark.driver.port", "29413")
sparkConf.set("spark.driver.host", "my-notebook-deployment.spark.svc.cluster.local")
# Initialize our Spark cluster, this will actually
# generate the worker nodes.
spark = SparkSession.builder.config(conf=sparkConf).getOrCreate()
sc = spark.sparkContext