Skip to main content

Jupyter Magics for EMR Notebooks.

Project description

EMR Notebooks iPython Magics

This repository contains iPython magics that can be used in Amazon EMR Notebooks.

Table of Contents

  1. Installation
  2. Usage
  3. Security
  4. License


Installing Dependencies

%mount_workspace_dir magic mounts the Workspace using S3-FUSE or Goofys.

  • Installing S3-FUSE

    Add the following lines to your cluster bootstrap action script.

    sudo amazon-linux-extras install epel -y
    sudo yum install s3fs-fuse -y
  • Installing Goofys

    Add the following lines to your cluster bootstrap action script.

    sudo wget -P /usr/bin/
    sudo chmod ugo+x /usr/bin/goofys

Installing iPython magics

  • Using EMR Step.

    EMR step script

    sudo -u emr-notebook /mnt/notebook-env/bin/pip install emr-notebooks-magics
  • From Jupyter Notebook

    %pip install emr-notebooks-magics

The magics are loaded using kernel startup script. If you install magics from Jupyter Notebook, you will need to restart the kernel before using the magic.

Note: EMR-notebook-magics cannot be installed through bootstrap actions as JEG and Notebook environments are installed after the bootstrap.


  • %generate_s3_download_url magic generates presigned url for S3 objects so that it can be downloaded from the Jupyter Notebook. Refer %generate_s3_download_url? for help.

    • Generate download url for a S3 object specifying full S3 path.

      %generate_s3_download_url s3://my_bucket/path/to/s3/object
    • Generate download url for a file in the Workspace specifying path relative to Workspace root.

      %generate_s3_download_url relative/path/to/workspace/file
  • %mount_workspace_dir magic mounts Workspace files on the EMR cluster instance using FUSE based filesystem. Refer %mount_workspace_dir? for help.

    • Mount the entire Workspace onto EMR cluster instance.

      %mount_workspace_dir .
    • Mount a sub-directory mydirectory and add use_cache mount option of S3-FUSE

      %mount_workspace_dir mydirectory --params use_cache=/tmp/
    • Mount a sub-directory mydirectory and add cheap, region mount option for Goofys.

      %mount_workspace_dir mydirectory --use goofys --params cheap,region=us-east-1
  • %execute_notebook magic executes another notebook in the background. Consider executing long-running notebooks in the background to ensure that the output is continuously captured even in case of a local network disruption. The output of the executed cells are incrementally captured in a new notebook with the same name as the executed notebook. The output notebook is placed inside a separate folder within the Workspace. Additional permissions are required for EMR-EC2 instance role to execute this magic. Refer %execute_notebook? for help.

    • Execute a notebook in the Workspace
      %execute_notebook <relative-file-path>
    • Execute a notebook specific cluster id and notebook service role
      %execute_notebook <notebook_name>.ipynb --cluster-id <emr-cluster-id> --service-role <emr-notebook-service-role>
:exclamation: Warnings
When the write access is enabled, any changes made to the mount directory are applied to the S3 Workspace. These changes are irreversible, please enable S3 versioning to your S3 Workspace as a pre-caution.
Once the Workspace is mounted on the EMR cluster, it can be accessed from all EMR Notebooks in your account that can attach to that cluster.
When you install S3-FUSE or Goofys, its your responsibility to keep those package up to date for new patches. Since Goofys is not managed by any package managers, take necessary steps to upgrade Goofys binaries.


See CONTRIBUTING for more information.


This project is licensed under the Apache-2.0 License.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

emr-notebooks-magics-0.2.4.tar.gz (14.0 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page