Implement working Task that deploys to k8s using helm

[bobcatfish]

Expected Behavior

Once we have implemented a simple hellworld Task (#59) a user should be able to define a Task that can deploy at a k8s resource using helm.

This would mean that:

• Templating is in place that allows for the Task to use the image and registry values specified in the input image resource in the steps of the Task as arguments to the helm commands
• Any required service account needed for accessing to the cluster is available to the Task
• Any logic needed to support the cluster type (see Add cluster and cluster bindings #60) is implemented

To try this out manually, you'll need a k8s cluster has been setup that you can deploy to.

Should include:

• Expanding the docs about how to write a Task that uses an k8s cluster resource, and how to set up the service account correctly
• Add docs on how to debug this when it doesn't work
• Test coverage, unit and also integration if possible (see Setup PR testing for build-pipeline integration tests #16)

Actual Behavior

We have no logic currently for even running a task, but #59 will add basic logic. #62 will expand on this but shouldn't block this task.

Steps to Reproduce the Problem

1. Deploy the Pipeline CRD + related CRDs to your cluster (note: should be docs on how to do this post kubebuilder)
2. Create a Resource like this (you can use a public image such as the kritis one in this example or you can use any other image):

apiVersion: pipeline.knative.dev/v1beta1
kind: Resource
metadata:
  name: helm-resources
  namespace: default
spec:
  resources:
    - name: kritis-app-github-resource
      type: github
      params:
      - name: url
        value: https://github.com/grafeas/kritis
      - name: revision
        value: master
    - name: kritis-server
      type: image
      params:
        - name: url
          value: gcr.io/kritis-project

3. Create a Task like this:

apiVersion: pipeline.knative.dev/v1beta1
kind: Task
metadata:
  name: helm
  namespace: default
spec:
    inputs:
        resources:
            - name: workspace
              type: github
            - name: imageToDeploy
              type: image
        params:
            - name: pathToHelmCharts
              value: string
            - name: helmArgs
              value: string
        cluster:
            - name: clusterToDeployTo
    buildSpec:
        steps:
        - name: deploy
          image: kubernetes-helm
          args: ['deploy', '--path=${pathToHelmChart}', '--set image=${imageToDeploy.Url}.${imageToDeploy.Name}', '${helmArgs}', '--cluster=${clusterToDeployTo.Endpoint}']

(Note I am pretty much making up the helm command here, please replace it with something that actually works and makes sense!)

4. Create a TaskRun like this:

apiVersion: pipeline.knative.dev/v1beta1
kind: TaskRun
metadata:
  name: helm-run-0
spec:
    taskRef:
        name: helm
    resourceRef:
       name: helm-resources
    trigger:
        triggerRef:
            type: manual

5. Look at the logs from the running container (Note the best docs we have on this at the moment are some of the steps in the Build quick start) to see that the helm command was successful (or not successful)
6. Observe that the status of the TaskRun has been updated appropriately, e.g. see this example which contains:

• Outcomes of individual steps
• Started, Completed, and Successful conditions

7. Look at the cluster itself and verify that the image has been deployed to it

Additional Info

• Blocked by Implement working "hello world" Task + TaskRun #59 (helloworld task), Implement the github Resource type #58 (GitHub resource).
• Whether or not you want the cluster to actually be managed by helm (e.g. as in this tutorial) or just use helm to deploy is optional.
• Kritis helm charts, docs are here

[nader-ziada]

Should we consider this as a resource since it is the output of a Task?

[bobcatfish]

Should we consider this as a resource since it is the output of a Task?

By "this" do you mean the cluster than gets deployed to? Definitely!

Or do you mean the deployment itself? b/c that's an interesting idea that I don't think we'd thought of!

[nader-ziada]

I meant the deployment of the chart, a helm resource that knows how to deploy similar to how the image resource knows how to upload the image to a registry

[bobcatfish]

I meant the deployment of the chart, a helm resource that knows how to deploy similar to how the image resource knows how to upload the image to a registry

Oooo neat! I think that's a cool idea, and makes a lot of sense since deployments will probably be a big part of what I imagine folks would want to use Pipelines for. Would you be interested in throwing together an issue to look into designing a Deployment resource?

[nader-ziada]

Sure, I'll write up an issue for it

[bobcatfish]

Note: for milestone 1, we should go ahead and implement this before #68

[nader-ziada]

/assign @pivotal-nader-ziada

[imjasonh]

If it helps, Helm's founder contributed and maintains a container image that runs helm, with GKE integration, here.

[nader-ziada]

@bobcatfish can you please close this issue, thanks

addressed by PR: #160
Will address the cluster selection by another issue and pr

[bobcatfish]

Okay I created #177 to add support for clusters, closing this issue.