1

I'm trying to run some jobs on aws cli using a virtual environment where I installed some libraries. I followed this guide; the same is here. But when I run the job I have this error:

Job execution failed, please check complete logs in configured logging destination. ExitCode: 1. Last few exceptions: Caused by: java.io.IOException: error=2, No such file or directory Exception in thread "main" java.io.IOException: Cannot run program "./environment/bin/python"

I also tried /home/hadoop/environment/bin/python as path but I obtain the same result. My job conf are:

--conf spark.archives=s3://mybucket/dependencies/myenv.tar.gz#environment --conf spark.emr-serverless.driverEnv.PYSPARK_DRIVER_PYTHON=./environment/bin/python --conf spark.emr-serverless.driverEnv.PYSPARK_PYTHON=./environment/bin/python --conf spark.emr-serverless.executorEnv.PYSPARK_PYTHON=./environment/bin/python

If I run in the job

os.listdir("./environment/bin/)

The result is

['python3.9', 'pip', 'pip3.9', 'rst2xetex.py', 'rstpep2html.py', 'f2py3', 'rst2latex.py', 'f2py', 'rst2odt.py', 'rst2html4.py', 'pip3', 'aws', 'python3', 'jp.py', 'rst2odt_prepstyles.py', 'pyrsa-encrypt', 'activate', 'rst2man.py', 'pyrsa-priv2pub', 'python', 'pyrsa-keygen', 'pyrsa-verify', 'rst2html.py', 'aws_completer', 'f2py3.9', 'venv-pack', 'rst2pseudoxml.py', 'aws_bash_completer', 'aws_zsh_completer.sh', 'aws.cmd', 'rst2s5.py', 'rst2xml.py', 'pyrsa-decrypt', 'rst2html5.py', 'Activate.ps1', '__pycache__', 'pyrsa-sign']

So the path should be correct. I also tried to set the PYSPARK_DRIVER_PYTHON inside the script as

os.environ['PYSPARK_PYTHON'] = "./environment/bin/python"
os.environ['PYSPARK_DRIVER_PYTHON'] = "./environment/bin/python"

But in this case the error is when I import libraries that I installed in the virtualenv, so it runs the script with the standard python.

Can you help me?

solopiu
  • 718
  • 1
  • 9
  • 28
  • I have the same problem. even in the logs it says unpacking but still fails: `Unpacking an archive s3://space/spark-env/pyspark_venv.tar.gz#environment from /tmp/spark-02908b0e-9b64-469d-b094-edee291a2426/pyspark_venv.tar.gz to /home/hadoop/./environment` – Amin Jul 20 '22 at 13:29

1 Answers1

5

The problem is that you probably haven't used Amazon Linux 2 to create the venv. Using Amazon Linux and Python 3.7.10 did it for me.

As detailed here you can use similar to this docker file to generate such a venv. you better use a requirements.txt to make it more reusable but it gives you the idea.

FROM --platform=linux/amd64 amazonlinux:2 AS base

RUN yum install -y python3

ENV VIRTUAL_ENV=/opt/venv
RUN python3 -m venv $VIRTUAL_ENV
ENV PATH="$VIRTUAL_ENV/bin:$PATH"

RUN python3 -m pip install --upgrade pip && \
    python3 -m pip install \
    great_expectations==0.15.6 \
    venv-pack==0.2.0

RUN mkdir /output && venv-pack -o /output/pyspark_ge.tar.gz

FROM scratch AS export
COPY --from=base /output/pyspark_ge.tar.gz /
Amin
  • 763
  • 7
  • 22