Skip to content

Latest commit

 

History

History
70 lines (48 loc) · 2.91 KB

README.md

File metadata and controls

70 lines (48 loc) · 2.91 KB
Raw

Infrastructure as Code deployment through Terraform

This document walks you through deploying and configuring the gitinsights function through Terraform. Specifically, the templates here will deploy the following:

  • Storage account and container used to hold Terraform state. These are kept in a separate resource group from the service deployment.
  • Resource group for gitinsights function deployment
  • KeyVault for hosting Azure DevOps token. The token is automatically created as a secret in this vault
  • Consumption App Service Plan
  • Azure Function with Managed Identity that can consume secrets from KeyVault

Configure environment

Many of the steps here assume that your shell environment is configured with the appropriate environment variables. You can find the required variables inside the relevant directory. Environment variables can be set using a tool like direnv, or by running the following:

DOT_ENV=.env
export $(cat $DOT_ENV | grep -v '^\s*#' | xargs)

Initial Deployment

Setup Terraform Backend State

Terraform state should live in a remote container in order to be leveraged across a variety of machines (developer workstations, CI agents, etc...). The bootstrap module provisions and configures the state container:

Note: ensure that the required environment variables are properly configured

cd bootstrap/
terraform apply

# capture backend state configuration
ARM_ACCESS_KEY=$(terraform output backend-state-account-key)
ARM_ACCOUNT_NAME=$(terraform output backend-state-account-name)
ARM_CONTAINER_NAME=$(terraform output backend-state-container-name)

cd ..

Deploy Infrastructure

Now that the remote state container is configured it is possible to deploy the infrastructure:

Note: ensure that the required environment variables are properly configuredInstructions for setting ARM_ACCESS_KEY, ARM_ACCOUNT_NAME, ARM_CONTAINER_NAME are described above.

Also, provide the required terraform variables with setting profile_aliases (i.e. {"cosmo@kramerica.com": "Cosmo Kramer"}), project_name, org_name, repo_names, backlog_team_name, token.

cd git-insights
terraform init \
    -backend-config "storage_account_name=${ARM_ACCOUNT_NAME}" \
    -backend-config "container_name=${ARM_CONTAINER_NAME}"
terraform plan -var-file="config.tfvars"
terraform apply -var-file="config.tfvars"

# capture function name for deployment
FUNCTION_NAME=$(terraform output function-name)

cd ..

Build & Deploy Azure Function

Now you should be able to deploy the function using the command below:

Note: this command should be run from the project root. Instructions for setting FUNCTION_NAME are described above.

func azure functionapp publish "$FUNCTION_NAME" --python