Giter Site home page Giter Site logo

playground's Introduction

Kosli Playground

A playground for learning how to implement Kosli

Learning about Kosli

Feel free to skip this section if you are already familiar with Kosli or want to get started straight away.

Understanding Kosli Core Concepts

Exploring real examples

Setting up

  • Click the Actions tab at the top of your forked repo.
  • You will see a message saying Workflows aren’t being run on this forked repository.
  • Click the green button to enable Workflows on your forked repo.
  • Please follow the remaining instructions from the README in your forked repo. (This is so the links correctly take you to the files in your repo)

Log into Kosli at https://app.kosli.com using GitHub

Logging in using GitHub creates a Personal Kosli Organization whose name is your GitHub username. You cannot invite other people to your personal organization; it is intended only to try Kosli out as you are now doing. For real use you would create a Shared Kosli Organization (from the top-right dropdown next to your user-icon) and invite people to it.

At https://app.kosli.com create a Docker Environment

In this playground the CI pipelines fake their deployment step by doing a "docker compose up". Create a Kosli Environment to record what is running in this fake deployment.

  • Select Environments from left hand side menu
  • Click the blue [Add new environment] button at the top
  • Fill in the Name field as playground-prod
  • Check the Docker host radio button for the Type
  • Fill in the Description field, eg Learning about Kosli
  • Leave Exclude scaling events checked
  • Leave Compliance Calculation Require artifacts to have provenance set to Off
  • Click the blue [Create environment] button
  • Open a tab in your browser for the playground-prod Kosli Environment as we will often review how it changes

Set the .env file variables

  • Edit (and save) the .env file as follows:
    • DOCKER_ORG_NAME to your GitHub username in lowercase
    • REPO_NAME (if you changed it from playground)

Create a KOSLI_API_TOKEN and save it as a GitHub Action secret

(Note: In a Shared Organization you would do this under a Service account)

  • At https://app.kosli.com click your GitHub user icon at the top-right
  • In the dropdown select Profile
  • Click the blue [+ Add API Key] button
  • Choose a value for the API key expires in or leave it as Never
  • Fill in the Description field, eg playground CI
  • Click the blue [Add] button
  • You will see the api-key, something like p1Qv8TggcjOG_UX-WImP3Y6LAf2VXPNN_p9-JtFuHr0
  • Copy this api-key (Kosli stores a hashed version of this, so it will never be available from https://app.kosli.com again). There is a small copy button to the right of the api-key.
  • Create a GitHub Action secret (at the repo level), called KOSLI_API_TOKEN, set to the copied value

Understand the fake deployment in the CI pipeline

  • There is a fake deploy job which runs this command to bring up the container in the CI pipeline!
    docker compose up --wait
    After this command, the CI pipeline installs the Kosli CLI, and runs this command:
    kosli snapshot docker "${KOSLI_ENVIRONMENT_NAME}"
    The kosli snapshot docker command takes a snapshot of the Docker containers currently running (inside the CI pipeline!) and sends their image names and digests/fingerprints to the named Kosli Environment (playground-prod). This command does not need to set the --org, or --api-token flags because the KOSLI_ORG and KOSLI_API_TOKEN environment variables have been set at the top of the workflow yml file.

Make a change, run the CI workflow, review the Environment in Kosli

  • Edit the file code/alpha.rb so the return string from the '/' route is a new string
  • Commit (add+commit+push if not editing in GitHub)
  • Wait for the GitHub Action Workflow to complete
  • Refresh your playground-prod Environment at https://app.kosli.com and verify it shows the playground-alpha image running. The image tag should be the short-sha of your new HEAD commit
  • This playground-alpha Artifact currently has No provenance but is nevertheless showing as Compliant. This is because the Environment was set up with Require artifacts to have provenance=Off. We will provide provenance shortly.

Make another change, rerun the CI workflow, review the Environment in Kosli

  • Re-edit the file code/alpha.rb so the return string from the '/' route is a new string
  • Commit (add+commit+push if not editing in GitHub)
  • Wait for the GitHub Action Workflow to complete
  • Refresh your playground-prod Environment at https://app.kosli.com and in the [Log] view verify
    • the previous playground-alpha Artifact has exited
    • the new playground-alpha Artifact is running, and this Artifact still has No provenance

Name your Kosli Flow and Trail

  • Kosli attestations must be made against a Trail, living inside a Flow.
    • A Kosli Flow represents a business or software process for which you want to track changes and monitor compliance.
    • A Kosli Trail represents a single execution of a process represented by a Kosli Flow. Each trail must have a unique identifier of your choice, based on your process and domain. Example identifiers include git commits or pull request numbers.
  • At the top of the .github/workflows/main.yml file add two new env: variables for the Kosli Flow (named after your repo) and Kosli Trail (named after each git-commit), as follows:
env:
  KOSLI_FLOW: playground-alpha-ci
  KOSLI_TRAIL: ${{ github.sha }}

Attest the provenance of the Artifact in the CI pipeline

  • Most attestations need the Docker image digest/fingerprint. We will start by making this available to all jobs.
  • In .github/workflows/main.yml...
    • uncomment the following comments near the top of the build-image: job
    #    outputs:
    #      artifact_digest: ${{ steps.variables.outputs.artifact_digest }}
    • uncomment the following comments at the bottom of the build-image: job
    #    - name: Make image digest available to all jobs
    #      id: variables
    #      run: |
    #        DIGEST=$(echo ${{ steps.docker_build.outputs.digest }} | sed 's/.*://')
    #        echo "artifact_digest=${DIGEST}" >> ${GITHUB_OUTPUT}
    • add the following to the end of the build-image: job to install the Kosli CLI and attest the Artifact's digest/fingerprint.
        - name: Setup Kosli CLI
          uses: kosli-dev/setup-cli-action@v2
          with:
            version: ${{ env.KOSLI_CLI_VERSION }}
    
        - name: Attest image provenance to Kosli Trail
          run: 
            kosli attest artifact "${{ needs.setup.outputs.image_name }}" 
              --artifact-type=docker
              --name=alpha
  • Note that kosli attest commands do not need to specify the --org or --flow or --trail flags because there are environment variables called KOSLI_ORG, KOSLI_FLOW, and KOSLI_TRAIL.
    • In the kosli attest artifact command above, the Kosli CLI calculates the fingerprint. To do this the CLI needs to be told the name of the Docker image (${{needs.setup.outputs.image_name}}), and that this is a Docker image (--artifact-type=docker), and that the image has previously been pushed to its registry (which it has)
    • You can also provide the fingerprint directly using the --fingerprint flag or KOSLI_FINGERPRINT environment variable.
  • Commit (add+commit+push if not editing in GitHub)
  • Wait for the GitHub Action Workflow to complete
  • Refresh your playground-prod Environment at https://app.kosli.com
  • You will see a new Snapshot
  • This time the Artifact will have Provenance. You can see the Flow and Trail, and also the git commit short-sha, and git commit message.
  • In https://app.kosli.com, click Flows on the left hand side menu
  • Click the Flow named playground-alpha-ci
  • You should see a single Trail whose name is the repo's current HEAD commit
  • Click the Trail name to view it, and confirm this Trail has a single Artifact attestation

View a deployment diff

  • Re-edit the file code/alpha.rb so the return string from the '/' route is a new string
  • Commit (add+commit+push if not editing in GitHub)
  • Wait for the GitHub Action Workflow to complete
  • Refresh your playground-prod Environment at https://app.kosli.com
  • You will see a new Snapshot
  • Its Artifact will have Provenance
  • Click the > chevron to reveal more information in a drop-down
  • Click the link titled View diff in the entry called Previous to see the deployment-diff; this is a diff (spanning potentially many commits) between the newly deployed alpha Artifact, and the previously running alpha Artifact.

Attest unit-test evidence to Kosli

  • .github/workflows/main.yml has a unit-test: job. You will attest its results to Kosli
  • Add the following to the end of the unit-test: job to install the Kosli CLI, and attest the unit-test results
    - name: Setup Kosli CLI
      uses: kosli-dev/setup-cli-action@v2
      with:
        version: ${{ env.KOSLI_CLI_VERSION }}
          
    - name: Attest unit-test results to Kosli
      run:
        kosli attest junit --name=alpha.unit-test --results-dir=test/reports/junit
  • Commit (add+commit+push if not editing in GitHub)
  • Wait for the GitHub Action Workflow to complete
  • The kosli attest junit command reports the JUnit XML results files in the --results-dir. The alpha Artifact's tests are written using the Ruby MiniTest framework. Like most test frameworks, it is easy to output the test results in the JUnit XML format. See line 7 of test/test_base.rb
  • Refresh your playground-prod Environment at https://app.kosli.com and verify it shows the new playground-alpha image running. The image tag should be the short-sha of your new HEAD commit
  • Open the latest Trail in the playground-alpha-ci Flow and verify
    • Its commit-sha name matches the commit in the latest playground-prod Snapshot
    • There is a Trail Event for the attestation called alpha.unit-test
    • Click the > chevron in this Trail Event to open its drop-down
    • Browse the JUnit results in the JSON sent by the kosli attest junit command

playground's People

Contributors

jonjagger avatar tooky avatar

Watchers

 avatar SimonC avatar Sami Alajrami avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.