Azure Multi Stage Pipeline Automation

This repository explains the deployment scenario of the reference architecture published in azure architecture center. It contains the following building blocks

• dotnet web project with angular front end - Example Workload to demonstrate pipeline automation
• azure resource group deployment project - Infrastructure as Code to deploy the example workload
• azure-pipelines yaml - pipeline definition with the steps required to build and deploy the example workload

Pre-Requisites

• Please follow the steps to deploy the azure-logic-app first.
• This will orchestrate the provisioning of multistage pipelines within your azure-devops project
• Create a service connection from azure devops to your azure subscription
• Change the parameters file in the deployment project to match your resource naming convention
• Change the variables in azure-pipelines.yml to match your subscriptionID and azure resource manager connection name
• change the webAppName in the yml file for staging and prod environment which matches with your deployment parameters

Deployment

when you commit this repository to your azure-devops project, the following actions will occur

• code-pushed event will be detected in azure devops
• The code-pushed event will dispatch a webhook to the HTTP Trigger of the Logic App
• The logic-app interacts with azure-devops-api to provision the pipelines when the code-push event is detected in any main branch of any repository within the azure devops project
• The build is queued and the artefact is deployed the staging and production environment
• The staging and production environment consists of an appservice plan and an appservice
• This however can be customized to the type of workload the team ships. It can be AKS, AzureFunctions etc.

Multi-Stage Pipelines

The pipelines comprises of the following stages

• build
• deploy to staging
• deploy to prouction

The build steps correspond to the steps required to build and publish a dotnet-angular workload. These steps can be customized to the type of workload the team is shipping out.

once the azure devops project is configured and setup as suggested, the setup will be representative of the below Architecture