Skip to content

Latest commit

 

History

History
125 lines (89 loc) · 3.83 KB

Day03.md

File metadata and controls

125 lines (89 loc) · 3.83 KB
Raw

Build & Testing of Applications (CI/CD)

📌 Resource → Civo DevOps BootCamp: [Session #3](https://youtu.be/ppdBp1GvGpw)
  • Continuous Integration part of the DevOps lifecycle

Where do testing & building fit in DevOps?

  • Testing is not just in one place
  • Continous Integration v/s Continuous Delivery
    • CI
      • getting your changes merged into the deployable state frequently & continuously
      • Extreme version of this is → commiting after each change made
      • getting your changes merged into the deployable state frequently & continuously
      • Extreme version of this is → commiting after each change made
    • CD
      • As soon as you merge the code into your main branch or deployable state, it goes straight into the production environment
      • There can be a few stages in b/w like performancing testing, reviews etc. but the main idea is above ☝🏻
      • A very few orgs have this as it might be dangerous to the application for some cases

Continuous Integration

  • Integration of small changes
  • Every time you commit → code is tested
  • New build with every commit to main branch
  • Tests done in CI & CD
    • CI
      • Unit Test
      • Integration Test
    • CD
      • Full system tests
      • Performance Tests
      • Conformance tests

What do we mean by “Testing”?

  • we actually mean Automated Testing
  • giving us the confidence that our application is working
    • “If I add this feature, how do we know the product still works”
  • We can write tests in every stage of the DevOps cycle
  • Types:
    • Unit Tests
      • we are testing small parts of the code like functions or some lines of code
      • really quick to run
    • Integration Tests
      • whether several different functions are able to work together efficiently or not
    • End-to-End Testing
    • Conformance Testing
  • Pyramid of Testing
    • Lower the rank of the test is, faster it would be ⇒ take less time to run

      1. Unit tests (in large numbers)

      2. Integration tests

        .

        .

        .

  • Faster the test, you should run them frequently in your development stage!

  • Good tests == honest feedback about your application

  • Test driven Development (19:43)

    • you start with a failing test
    • you try to reproduce an issue in your application by creating a test for each part of it
    • These test would run & if there is a failing test, you would know:
      • why its failing?
      • where its failing?
      • & what you need to do to fix it!

CI Pipelines

  • Pipeline → set of automated process that take your application from developing → to running tests → to deploying → to the production environment
  • ⇒ x ⇒ y ⇒ z ⇒ . . . : At each step, some processes would automatically be executed & this sort of pathway is called as a Pipeline
  • Tools:
    • Jenkins
    • circleCI
    • TravisCI
    • GitLab CI/CD
  • which tool to choose?
    • depends on the budget, your requirements etc . . .
    • it should be able to continously integrate your code
  • configuring CI Pipelines is in the Operations side of things in DevOps
  • In an application, a function might depend on another function to work & writing tests can help us detect any sort of errors that might break the code
    • Also gives us confidence
  • Every programming language has a separate testing framework to be used for that specific language
    • Example → Golang has go test etc.
  • Configuring the CI Pipeline & adding tests there is a separate task & is done differently!

A great feeling when all the test checks pass 🎉