pre-submit.yaml

name: Pre Submit
on:
  push:
    branches:
      - main
      - release-*
  pull_request:
    branches:
      - main
      - release-*
jobs:
  build-and-unit-test:
    runs-on: ubuntu-22.04
    steps:
    - name: Checkout code
      uses: actions/checkout@v3
      with:
        fetch-depth: 0

    - name: Set up Go
      uses: actions/setup-go@v4
      with:
        go-version-file: go.mod

    - name: Build
      run: make manager

    - name: Test
      run: make test

    - name: Test container build
      run: make container-build

    - name: TestMutations
      run: make test-mutation-ci

  e2e-k8s:
    runs-on: ubuntu-22.04
    timeout-minutes: 30
    env:
      IMAGE_REGISTRY: kind-registry:5000
      VERSION: 9.9.9-ci
      DEPLOY_NAMESPACE: k8s-test
      # see https://github.com/kubernetes-sigs/kind/tags
      KIND_VERSION: v0.18.0
      # see https://hub.docker.com/r/kindest/node/tags for available versions!
      K8S_VERSION: v1.26.3
    steps:
    - name: Checkout code
      uses: actions/checkout@v3
      with:
        fetch-depth: 0

    - name: Set up Go
      uses: actions/setup-go@v4
      with:
        go-version-file: go.mod

    - name: Configure insecure registry
      run: |
        #sudo cat /etc/docker/daemon.json

        # allow insecure registry but keep original config!
        sudo bash -c "cat <<EOF >/etc/docker/daemon.json
        {
          \"exec-opts\": [\"native.cgroupdriver=cgroupfs\"],
          \"cgroup-parent\": \"/actions_job\",
          \"insecure-registries\" : [\"${IMAGE_REGISTRY}\"]
        }
        EOF"

        #sudo cat /etc/docker/daemon.json
        sudo systemctl restart docker

        # same for podman
        sudo bash -c "cat <<EOF >/etc/containers/registries.conf
        [[registry]]
        location=\"${IMAGE_REGISTRY}\"
        insecure=true
        EOF"
        #sudo cat /etc/containers/registries.conf

    - name: Start kind cluster
      uses: container-tools/kind-action@v2
      with:
        version: ${{env.KIND_VERSION}}
        config: ./hack/kind-config.yaml
        node_image: kindest/node:${{env.K8S_VERSION}}
        kubectl_version: ${{env.K8S_VERSION}}
        registry: true

    - name: Cluster info
      run: |
        kubectl version -o=yaml
        kubectl cluster-info
        kubectl get nodes -o=wide

    - name: Install OLM
      run: |
        curl -sL https://github.com/operator-framework/operator-lifecycle-manager/releases/download/v0.22.0/install.sh | bash -s v0.22.0

    - name: Build NHC
      run: |
        export NHC_SKIP_TEST=true
        make container-build-k8s container-push
        # make operator-sdk findable...
        cp ./bin/operator-sdk /usr/local/bin/

    - name: Deploy NHC + SNR
      run: |
        kubectl create ns ${DEPLOY_NAMESPACE}
        
        # deploy SNR first
        operator-sdk run bundle -n ${DEPLOY_NAMESPACE} quay.io/medik8s/self-node-remediation-operator-bundle:latest
        
        # deploy NHC
        operator-sdk run bundle -n ${DEPLOY_NAMESPACE} --use-http ${IMAGE_REGISTRY}/node-healthcheck-operator-bundle:v${VERSION}
        
        # wait a bit for OLM creating CSV etc
        sleep 1m

    - name: OLM status
      if: ${{ always() }}
      run: |
        kubectl get -A OperatorGroup -o wide
        kubectl get -A CatalogSource -o wide
        kubectl get -A Subscription -o wide
        kubectl get -A ClusterServiceVersion -o wide
        kubectl get -A InstallPlan -o wide

    - name: Wait for deployments
      run: |
        kubectl wait deployment -n ${DEPLOY_NAMESPACE} self-node-remediation-controller-manager --for condition=Available=True --timeout=300s
        # give SNR some time to create the DS
        sleep 30s
        kubectl rollout status daemonset -n ${DEPLOY_NAMESPACE} self-node-remediation-ds --timeout 300s
        kubectl wait deployment -n ${DEPLOY_NAMESPACE} node-healthcheck-controller-manager --for condition=Available=True --timeout=300s

    - name: Deployment status
      if: ${{ always() }}
      run: |
        kubectl -n ${DEPLOY_NAMESPACE} get deployments,daemonsets,pods -o=wide

    - name: Run NHC e2e
      run: |
        echo "running e2e test"
        OPERATOR_NS=${DEPLOY_NAMESPACE} TEST_OPTS='-ginkgo.label-filter="!OCP-ONLY"' make test-e2e
        echo "finished e2e test"

    - name: Debug
      if: ${{ failure() }}
      run: |
        # debug NHC
        echo "Debugging NHC"
        kubectl describe deployment -n ${DEPLOY_NAMESPACE} node-healthcheck-controller-manager
        echo "\n\n"
        kubectl describe pod -n ${DEPLOY_NAMESPACE} --selector=app.kubernetes.io/name=node-healthcheck-operator,app.kubernetes.io/component=controller-manager
        echo "\n\n"
        kubectl logs -n ${DEPLOY_NAMESPACE} -c manager --selector=app.kubernetes.io/name=node-healthcheck-operator,app.kubernetes.io/component=controller-manager --tail -1
        echo "\n\n"
        
        echo "Debugging SNR operator"
        kubectl describe deployment -n ${DEPLOY_NAMESPACE} self-node-remediation-controller-manager
        echo "\n\n"
        kubectl describe pod -n ${DEPLOY_NAMESPACE} --selector=self-node-remediation-operator=,control-plane=controller-manager
        echo "\n\n"
        kubectl logs -n ${DEPLOY_NAMESPACE} -c manager --selector=self-node-remediation-operator=,control-plane=controller-manager --tail -1
        echo "\n\n"
        
        echo "Debugging SNR agents"
        kubectl describe daemonset -n ${DEPLOY_NAMESPACE} self-node-remediation-ds
        echo "\n\n"
        kubectl describe pod -n ${DEPLOY_NAMESPACE} --selector=app=self-node-remediation-agent,control-plane=controller-manager
        echo "\n\n"
        kubectl logs -n ${DEPLOY_NAMESPACE} --selector=app=self-node-remediation-agent,control-plane=controller-manager --tail -1