cellxgene-fargate

Host CZI's cellxgene on AWS Fargate.

Usage

1. Clone the repository and its submodules:

   git clone --recurse-submodules [email protected]:DataBiosphere/cellxgene-fargate.git
   

2. Create the .active symlink to the active deployment, for example

   (cd deployments && ln -snf sc/dev .active)
   

3. Create deployments/.active/environment.local.py with any local settings you might need, like AWS_DEFAULT_PROFILE. The format of that file is exactly the same as that of environment.py. Then run

   source environment
   

   Once you've sourced the environment, you can use the _select shell function to activate a different deployment, and _refresh to source the environment again after changes to any of the environment*.py files.

4. Create and populate a virtualenv with

   make virtualenv
   source .venv/bin/activate
   make envhook  # optional but recommended when using PyCharm
   make requirements.dev
   

   The project is configured and all development dependencies have been installed.

5. To create a Docker image with cellxgene inside run

   make docker_image
   

6. Test the image with

   make docker_run
   echo $CELLXGENE_VERSION
   

   The cellxgene version printed by the container should match the value of the environment variable.

7. Once per AWS account and region, an Amazon ECR image repository needs to be created:

   make docker_repository
   

8. Push the cellxgene Docker image to Amazon ECR with

   make docker_login
   make docker_push
   

9. Provision the AWS resources needed to run cellxgene as a Fargate container in Amazon ECS behind an EC2 application load balancer:

   _preauth  # if you're assuming an IAM role requiring MFA
   make terraform