Containerized WebApi

The goal of this readme is to show how to build a container WebApi image that can be used to deploy the application into a container registry and that can be used later to run the service.

Build and List Runtime Image

Use docker to build the WebApi images. (Note: Make sure the run docker at the root branch)

cd ./$(git rev-parse --show-cdup)/src
docker build -t carbon_aware:v1 -f CarbonAware.WebApi/src/Dockerfile .

List carbon_aware image

docker image ls carbon_aware
REPOSITORY     TAG       IMAGE ID       CREATED             SIZE
carbon_aware   v1        6293e2528bf2   About an hour ago   230MB

Run WebApi Image

1. Run the image using docker run with host port 8000 mapped to the WebApi port 8080 and configure environment variable settings for WattTime provider.

   docker run --rm -p 8000:8080 \
   > -e DataSources__EmissionsDataSource="WattTime" \
   > -e DataSources__ForecastDataSource="WattTime" \
   > -e DataSources__Configurations__WattTime__Type="WattTime" \
   > -e DataSources__Configurations__WattTime__password="username" \
   > carbon_aware:v1
   

   or the ElectricityMaps provider

   docker run --rm -p 8000:8080 \
   > -e DataSources__EmissionsDataSource="ElectricityMaps" \
   > -e DataSources__ForecastDataSource="ElectricityMaps" \
   > -e DataSources__Configurations__ElectricityMaps__Type="ElectricityMaps" \
   > -e DataSources__Configurations__ElectricityMaps__APITokenHeader="auth-token" \
   > -e DataSources__Configurations__ElectricityMaps__APIToken="<YOUR_ELECTRICITYMAPS_TOKEN>" \
   > carbon_aware:v1
   

   or the ElectricityMapsFree provider

   docker run --rm -p 8000:8080 \
   > -e DataSources__EmissionsDataSource="ElectricityMapsFree" \
   > -e DataSources__Configurations__ElectricityMapsFree__Type="ElectricityMapsFree" \
   > -e DataSources__Configurations__ElectricityMapsFree__token="<YOUR_CO2SIGNAL_TOKEN>" \
   > carbon_aware:v1
   

2. Verify that the WebApi is responding to requests using an HTTP client tool (e.g. postman, curl)

   curl -v -s -X 'POST' http://localhost:8000/emissions/forecasts/batch  -H 'accept: */*' -H 'Content-Type: application/json' -d '[
       {
           "requestedAt": "2021-11-01T00:00:00Z",
           "dataStartAt": "2021-11-01T00:05:00Z",
           "dataEndAt": "2021-11-01T23:55:00Z",
           "windowSize": 5,
           "location": "eastus"
       }
   ]'
   ...
   > POST /emissions/forecasts/batch HTTP/1.1
   > Host: localhost:8000
   ...
   < HTTP/1.1 200 OK
   < Content-Type: application/json; charset=utf-8
   ...
   <
   [{"generatedAt":"2021-11-01T00:00:00+00:00","optimalDataPoint":{
       ...
   }}]
   

Upload Image to a Container Registry

For easy image consumption, upload it to a well-known container registry, self-hosted or managed. The following are examples of using docker hub or Azure Container Registry

Docker Hub

Sign in to Docker Hub and create a private (or public) repository (e.g #your-username#/my-private-repo)

1. Build and Push

   docker login --username=your-username
   cd ./$(git rev-parse --show-cdup)/src
   docker build -t #your-username#/my-private-repo/carbon_aware:v1 -f CarbonAware.WebApi/src/Dockerfile .
   docker push #your-username#/my-private-repo/carbon_aware:v1
   

2. Pull

   docker login --username=your-username
   docker pull #your-username#/my-private-repo/carbon_aware:v1
   

Azure Container Registry

1. Build and Push image Assuming the container registry is already created, use the user's credentials push the image using docker (it can be done also using Azure CLI)

   docker login <myacrname>.azurecr.io -u username -p <CopiedKeyFromAzurePortal>
   cd ./$(git rev-parse --show-cdup)/src
   docker build -t <myacrname>.azurecr.io/carbon_aware:v1 -f arbonAware.WebApi/src/Dockerfile .
   docker push <myacrname>.azurecr.io/carbon_aware:v1
   

2. Pull image

   docker login <myacrname>.azurecr.io -u username -p <CopiedKeyFromAzurePortal>
   docker pull <myacrname>.azurecr.io/carbon_aware:v1
   

Pipeline Integration (Github Action)

To automate an image deployment from a GitHub CI/CD pipeline, the following link provides detailed information on how to build a workflow with all the necesarily tools in order to push an image to a container registry of user's preference (i.e. Docker Hub).

Github Workflows