carlomazzaferro / docker-ci

Custom docker files for ci pipelines

Docker + CI

Set of docker files that are useful for running ci/cd workflow on CircleCI and Concourse. Come pre-packaged with a set of tools that may be useful for folks working with Terrafom, AWS CLI, Node, Python and docker (as in, you can run docker commands from within these containers. See: https://circleci.com/docs/2.0/circleci-images/#pre-installed-tools) for the entire set of available pre-packaged tools.

These mostly use slim base images to improve images size.

Images

CircleCI

Node + Terraform + DinD + awscli

Details:

• File: circleci/Dockerfile.circle
• Pull: docker pull carlomazzaferro/circle-ci-cm

Specs:

• Node 11
• Terraform 0.11.11
• awscli
• All the tools from CircleCI base images

Ideal for:

• Running tests on fronted servers
• Packaging frontend docker images
• Deploying images with the terraform CLI

Size: 998MB

Python + DinD + awscli

• File: circleci/Dockerfile.circle.awscli
• Pull: carlomazzaferro/circle-ci-awscli
• All the tools from CircleCI base images

Size: 620MB

Terraform + DinD

• File: circleci/Dockerfile.circle.terraform
• Pull: carlomazzaferro/circle-ci-terraform
• Some of the tools from CircleCI base images. See Dockerfile definition for more info.

Size: 367 MB

Ideal for:

• Deploying images with the terraform CLI using a pretty small image

Concourse

Alpine Python + Scientific libs (numpy, sagemaker) + awscli + Terraform

• File: concourse/Dockerfile.concourse.python
• Pull: carlomazzaferro/concourse-tf-python

Size: 350 MB

Ideal for:

• Provision terraform infra and do Sagemaker-related operations such as deploying models, until when terraform supports it fully

Python + TensorFlow + awscli + Terraform

• File: concourse/Dockerfile.concourse.tensorflow
• Pull: carlomazzaferro/concourse-tf-python-tensorflow

Size: 1.4 GB

Ideal for:

• Provision terraform infra and do Sagemaker-related operations such as training TensorFlow models