Panoply Slugrunner

The Slugrunner is a Docker container that will run Heroku-like slugs. A slug is essentially a gzipped tarball of an application that is self-contained and includes everything it needs to run. These slugs can be built from panoplymedia/slugbuilder, lxfontes/slugbuilder, deis/slugbuilder and the like.

Requirements

You will need to have Docker installed and set up on your system.

Installation

docker pull panoplymedia/slugrunner

Usage

You will use docker run to run your slug in this container, passing a string argument at the end, which will be run.

# ellipses notate additional args
docker run ... panoplymedia/slugrunner npm start

docker run ... panoplymedia/slugrunner bundle exec rails s

Alternatively, if your slug contains a Procfile you can use the start command followed by the name of the process.

# Procfile
web: npm start
worker: node worker.js

# ellipses notate additional args
docker run ... panoplymedia/slugrunner start web

There are various ways you can pass the slug in to be run:

STDIN

cat myslug.tgz | docker run -i -a stdin -a stdout panoplymedia/slugrunner start worker

SLUG_URL Environment Variable

By specifying the SLUG_URL as an environment variable, it will use the slug located at that URL.

docker run -e SLUG_URL=http://example.org/myslug.tgz panoplymedia/slugrunner bundle exec rails s

Amazon S3

You can also use a slug from Amazon S3 with the following environment variables:

• SLUG_KEY: s3-bucket-name/path/to/slug.tgz
• SLUG_AWS_KEY: your AWS access key
• SLUG_AWS_SECRET: your AWS secret access key

docker run -e SLUG_KEY=my-bucket/items/stuff/someslug.tgz -e SLUG_AWS_KEY=XXX -e SLUG_AWS_SECRET=XXX panoplymedia/slugrunner start web

Google Cloud Storage

You may use Google Cloud with a developer access key and secret key. Alternatively you may use service account credentials.

To use a slug from Google Cloud Storage with a developer access key and secret key, follow Google's documentation and then provide the following environment variables:

• SLUG_KEY: cloud-storage-bucket-name/path/to/slug.tgz
• SLUG_GOOGLE_KEY: your Cloud Storage access key
• SLUG_GOOGLE_SECRET: your Cloud Storage secret access key

docker run -e SLUG_KEY=my-bucket/items/stuff/someslug.tgz -e SLUG_GOOGLE_KEY=XXX -e SLUG_GOOGLE_SECRET=XXX panoplymedia/slugrunner start web

To use a slug from Google Cloud Storage with service account credentials, follow Google's documentation to generate a JSON service account credential and convert the credential into a Base64 string. Then provide the following environment variables:

• SLUG_KEY: cloud-storage-bucket-name/path/to/slug.tgz
• SLUG_GOOGLE_ACCOUNT_KEY_B64: your Base64 encoded JSON-formatted Google service account credentials

Debugging

Set environment variable DEBUG=true to get debug output from the script.

Motivation and Thanks

This library is heavily based on deis/slugrunner, and most notably adds the option to run a slug off of Amazon S3 and Google Cloud Storage.