Skip to main content

Containerization

DOI

Language Language

The containerization library is part of the of the TOS2CA Project. For more information, visit the TOS2CA website at https://nasa-jpl.github.io/tos2ca-documentation/.

This python library contains examples of how build images with the TOS2CA Anomaly Detection repo. The project uses this with AWS Fargate services, though it could be run locally on a single server if needed.

Requirements​

  • Will need an NASA Earthdata login to use any tools DAAC tools/applications (and have credentials in a .netrc file)
  • Should have access to the us-west-2 AWS region to access any NASA DAAC data over S3
  • Access to additional AWS services required by the tos2ca-anomaly-detection repo

How to run the tos2ca anomaly detection drivers​

The next steps are all in the script SetUp_Docker_tos2ca_jobs.sh if you choose but here is how to setup everything

  1. Into the top directory, make sure to have your Earthdata login .netrc file
  2. Change to the directory and run the commands to build the images:
    • Example: docker build -t data_driver . -f Dockerfile_tos2ca_data_driver
  3. Once an image is built, a container can be run like:
    • Example: docker run -it -e JOBID=469 -e CHUNKID=1 data_driver
    • Note that each container will have enviornmental variables that you will need to pass to it.
  4. Optionally, you can then tag image:
    • docker tag <image ID> <AWS URI>/<repo name>:<tag>
  5. Optionally, you can also then push the image to a repository, like AWS ECR:
    • aws ecr get-login-password --region <region> | docker login --username AWS --password-stdin <AWS URI>
    • docker push <AWS URI>/<repo name>:<tag>