orlade/cloudchart repository overview

⁠Cortex

Cortex is an application for monitoring and managing AWS infrastructure.

Cortex accompanies the WIP book on AWS configuration and deployment, Modern Deployment⁠.

⁠Disclaimer

Cortex is under development as a side project. It should not be considered safe for production use. Providing full access your AWS account to any person or application, even if read-only, is a risky proposition.

As per the MIT license, this software comes with absolutely no warranty, and the author will not be held liable if you manage to stuff up any aspect of your AWS environment.

If you choose to run Cortex at all, I would recommend a private instance with read-only access and a very strong admin password.

⁠Configuration

The following environment variables must be set on the Cortex server:

• AWS IAM credentials for a user with at least read-only access to EC2, ECS and S3:
  • AWS_ACCESS_KEY_ID
  • AWS_SECREY_ACCESS_KEY
• Cortex admin user account details (for login):
  • METEOR_ADMIN_USERNAME
  • METEOR_ADMIN_PASSWORD

The upcoming multi-tenancy feature will allow Cortex users to register their AWS access credentials within the app, rather than having to set them on the server.

Some of the /common/configs specifications are deliberately incomplete. The missing values will be populated from the process.env of the host machine. Refer to the /common/configs to determine which environment variables must be set to create new services.

⁠Build and Run

Cortex is designed to be built with Meteor and run as a Node.js app. To simplify this process, we use Docker for both building and deploying the app using the Dockerfile.

Refer to docker-meteor⁠ for more detailed instructions, but in summary:

1. Install Docker, then cd to the Cortex repo root directory
2. Build the orlade/cortex Docker image with # docker build -t orlade/cortex .
3. Run a Cortex container with # docker run --rm -p 3000:3000 -e "MONGO_URL=mongodb://username:password@host/database" -e "ROOT_URL=http://localhost:3000" -e "AWS_ACCESS_KEY_ID=myaccesskey" -e "AWS_SECRET_ACCESS_KEY=mysecretkey" orlade/cortex

If manually hosting an instance of MongoDB:

1. Run it in the background in a named Docker container like # docker run --name mydb -d mongo
2. Add --link mydb:db to the docker run command above
3. Also in the run command, replace the MONGO_URL variable above with mongodb://db