Azure devops notes
Planning
- by Product Manager
- What to developed
- Why: business value
Agile is usually used in a workflow.
Azure boards is the module we use on the platform.
- create work items
- track progress
- discussion
Code
Azure Repos
- host code
- similar to Github
- create branch -> pull request by others
Azure Pipelines build
What we want to do is to release the new feature.
Process: code -> test -> package -> deploy
CI
- continuous intergration
- test -> package
pipeline
- write in yaml
- test -> package -> build image -> push to repo
- use task (pre-created script)
- Job: boundary of a set of steps
- run in defferent environments
- each runs on an agent
- which OS / version
- agent
- selected from pools
Build or CI pipeline
- test code changes
- produce an artifact for deployment
Azure Artifacts
use Docker
Docker images
stored in Container Registry
Azure Pipelines Stages
CD: continuous delivery
CI/CD
- fully automated process
- code changes to production environment
Stage
- logical boundery in the pipeline
- contains jobs
- run after one
deploy
- intermediate environment: dev, testing, prod
- avoid pipeline configuration code
- extract common configuration to a template
Azure Pipelines Environments
- link tickets in Azure Boards
Release Pipelines
separate pipeline for CD
1 CI/CD pipeline in yaml
Test Plans
- test cases
- steps / output
- automated tests
- test reports
Azure Architecture
SaaS / Azure DevOps Services + Configure own agents
Pipeline task run on separate machines: Agent
Service Connections
interact with other platforms
- push docker image to container repository
- deploy to remote service: aws, gcp
- external code repository
manage external connection: authenticate
Comparison
Azure Devops: devops; everything in 1 platform
- better traceability: linked data
Jenkins / argo: CI/CD processes only
Gitlab
- similar
aws
- more than devops
- cloud platform
- microsoft azure + azure devops
Reference
Azure devops notes
https://hexwhat.top/2024/08/12/azure-devops/