Add support for testing aws-hosted-cp

* Break KubeClient helpers into provider specific file. * Try to simplify the validation process for lots of different providers with different requirements. * Finish aws-hosted-cp test and add comments through test to make it easier to understand. * Use GinkgoHelper across e2e tests, populate hosted vars from AWSCluster. * No longer rely on local registry for images in test/e2e. * Support OS for awscli install. * Prepend hostname to collected log artifacts. * Support no cleanup of provider specs, differentiate ci cluster names. * Add docs on running tests, do not wait for all providers if configured. * Reinstantiate resource validation map on each instance of validation. * Enable the external-gc feature via annotation, featureGate bool. (Closes: #152) * Bump aws-*-cp templates to 0.1.3 * Bump cluster-api-provider-aws template to 0.1.2 * Improve test logging to log template name and validation phase. * Bump k0s version to v1.30.4+k0s.0, set CCM nodeSelector to null for aws-hosted-cp. (Closes: #290) * Break cleanup into seperate job so that it is unaffected by concurrency group cancellations. * Make dev-aws-nuke target less PHONY. Closes: #212 Signed-off-by: Kyle Squizzato <[email protected]>
Mirantis · Sep 16, 2024 · d590bfb · d590bfb
1 parent b05008b
commit d590bfb
Show file tree

Hide file tree

Showing 31 changed files with 960 additions and 423 deletions.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
@@ -26,7 +26,7 @@ env:
 
 jobs:
   build:
-    name: Build and Unit Test
+    name: Build and Test
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository

diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
@@ -1,12 +1,7 @@
 name: E2E Tests
 
-concurrency:
-  group: test-${{ github.head_ref || github.run_id }}
-  cancel-in-progress: true
-
 on:
-  pull_request_target:
-    types: [labeled]
+  pull_request:
     branches:
       - main
       - release-*
@@ -15,31 +10,91 @@ on:
       - '**.md'
 env:
   GO_VERSION: '1.22'
+  AWS_REGION: us-west-2
+  AWS_ACCESS_KEY_ID: ${{ secrets.CI_AWS_ACCESS_KEY_ID }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.CI_AWS_SECRET_ACCESS_KEY }}
 
 jobs:
   e2etest:
+    concurrency:
+      group: test-e2e-${{ github.head_ref || github.run_id }}
+      cancel-in-progress: true
     name: E2E Tests
     runs-on: ubuntu-latest
-    if: contains(github.event.pull_request.labels.*.name, 'test-e2e')
-    env:
-      AWS_REGION: us-west-2
-      AWS_ACCESS_KEY_ID: ${{ secrets.CI_AWS_ACCESS_KEY_ID }}
-      AWS_SECRET_ACCESS_KEY: ${{ secrets.CI_AWS_SECRET_ACCESS_KEY }}
+    outputs:
+      clustername: ${{ steps.vars.outputs.clustername }}
+      version: ${{ steps.vars.outputs.version }}
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
       - name: Setup Go
         uses: actions/setup-go@v5
         with:
-        go-version: ${{ env.GO_VERSION }}
+          go-version: ${{ env.GO_VERSION }}
+      - name: Set up Buildx
+        uses: docker/setup-buildx-action@v3
+      - name: Login to GHCR
+        uses: docker/[email protected]
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+      - name: Get outputs
+        id: vars
+        run: |
+          echo "version=$(git describe --tags --always)" >> $GITHUB_OUTPUT
+          echo "clustername=ci-$(date +%s)-e2e-test" >> $GITHUB_OUTPUT
+      - name: Build and push HMC controller image
+        uses: docker/build-push-action@v6
+        with:
+          build-args: |
+            LD_FLAGS=-s -w -X github.com/Mirantis/hmc/internal/build.Version=${{ steps.vars.outputs.version }}
+          context: .
+          platforms: linux/amd64,linux/arm64
+          tags: |
+            ghcr.io/mirantis/hmc/controller-ci:${{ steps.vars.outputs.version }}
+          push: true
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+      - name: Prepare and push HMC template charts
+        run: |
+          make hmc-chart-release
+          REGISTRY_REPO="oci://ghcr.io/mirantis/hmc/charts-ci" make helm-push
       - name: Setup kubectl
         uses: azure/setup-kubectl@v4
       - name: Run E2E tests
+        env:
+          MANAGED_CLUSTER_NAME: ${{ steps.vars.outputs.clustername }}
+          REGISTRY_REPO: 'oci://ghcr.io/mirantis/hmc/charts-ci'
+          IMG: 'ghcr.io/mirantis/hmc/controller-ci:${{ steps.vars.outputs.version }}'
         run: |
           make test-e2e
       - name: Archive test results
+        if: ${{ failure() }}
         uses: actions/upload-artifact@v4
         with:
-        name: test-logs
-        path: |
-          test/e2e/*.log
+          name: test-logs
+          path: |
+            test/e2e/*.log
+  cleanup:
+    name: Cleanup
+    needs: e2etest
+    runs-on: ubuntu-latest
+    if: ${{ always() }}
+    timeout-minutes: 15
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+      - name: Setup Go
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+      - name: AWS Test Resources
+        env:
+          CLUSTER_NAME: '${{ needs.e2etest.outputs.clustername }}'
+        run: |
+          make dev-aws-nuke
diff --git a/Makefile b/Makefile
@@ -2,6 +2,8 @@ NAMESPACE ?= hmc-system
 VERSION ?= $(shell git describe --tags --always)
 # Image URL to use all building/pushing image targets
 IMG ?= hmc/controller:latest
+IMG_REPO = $(shell echo $(IMG) | cut -d: -f1)
+IMG_TAG = $(shell echo $(IMG) | cut -d: -f2)
 # ENVTEST_K8S_VERSION refers to the version of kubebuilder assets to be downloaded by envtest binary.
 ENVTEST_K8S_VERSION = 1.29.0
 
@@ -103,10 +105,11 @@ tidy:
 test: generate-all fmt vet envtest tidy external-crd ## Run tests.
 	KUBEBUILDER_ASSETS="$(shell $(ENVTEST) use $(ENVTEST_K8S_VERSION) --bin-dir $(LOCALBIN) -p path)" go test $$(go list ./... | grep -v /e2e) -coverprofile cover.out
 
-# Utilize Kind or modify the e2e tests to load the image locally, enabling compatibility with other vendors.
-.PHONY: test-e2e # Run the e2e tests against a Kind k8s instance that is spun up.
+# Utilize Kind or modify the e2e tests to load the image locally, enabling
+# compatibility with other vendors.
+.PHONY: test-e2e # Run the e2e tests using a Kind k8s instance as the management cluster.
 test-e2e: cli-install
-	KIND_CLUSTER_NAME="hmc-test" KIND_VERSION=$(KIND_VERSION) go test ./test/e2e/ -v -ginkgo.v -timeout=2h
+	 KIND_CLUSTER_NAME="hmc-test" KIND_VERSION=$(KIND_VERSION) go test ./test/e2e/ -v -ginkgo.v -timeout=2h
 
 .PHONY: lint
 lint: golangci-lint ## Run golangci-lint linter & yamllint
@@ -240,6 +243,13 @@ hmc-deploy: helm
 
 .PHONY: dev-deploy
 dev-deploy: ## Deploy HMC helm chart to the K8s cluster specified in ~/.kube/config.
+	@$(YQ) eval -i '.image.repository = "$(IMG_REPO)"' config/dev/hmc_values.yaml
+	@$(YQ) eval -i '.image.tag = "$(IMG_TAG)"' config/dev/hmc_values.yaml
+	@if [ "$(REGISTRY_REPO)" = "oci://127.0.0.1:$(REGISTRY_PORT)/charts" ]; then \
+		$(YQ) eval -i '.controller.defaultRegistryURL = "oci://$(REGISTRY_NAME):5000/charts"' config/dev/hmc_values.yaml; \
+	else \
+		$(YQ) eval -i '.controller.defaultRegistryURL = "$(REGISTRY_REPO)"' config/dev/hmc_values.yaml; \
+	fi; \
 	$(MAKE) hmc-deploy HMC_VALUES=config/dev/hmc_values.yaml
 	$(KUBECTL) rollout restart -n $(NAMESPACE) deployment/hmc-controller-manager
 
@@ -317,15 +327,16 @@ dev-mcluster-delete: envsubst
 .PHONY: dev-creds-apply
 dev-creds-apply: dev-$(DEV_PROVIDER)-creds
 
-.PHONY: envsubst awscli dev-aws-nuke
-dev-aws-nuke: ## Warning: Destructive! Nuke all AWS resources deployed by 'DEV_PROVIDER=aws dev-provider-apply', prefix with CLUSTER_NAME to nuke a specific cluster.
+.PHONY: dev-aws-nuke
+dev-aws-nuke: envsubst awscli yq cloud-nuke ## Warning: Destructive! Nuke all AWS resources deployed by 'DEV_PROVIDER=aws dev-provider-apply', prefix with CLUSTER_NAME to nuke a specific cluster.
+	@CLUSTER_NAME=$(CLUSTER_NAME) YQ=$(YQ) AWSCLI=$(AWSCLI) bash -c "./scripts/aws-nuke-ccm.sh elb"
 	@CLUSTER_NAME=$(CLUSTER_NAME) $(ENVSUBST) < config/dev/cloud_nuke.yaml.tpl > config/dev/cloud_nuke.yaml
-	DISABLE_TELEMETRY=true $(CLOUDNUKE) aws --region $$AWS_REGION --force --config config/dev/cloud_nuke.yaml --resource-type vpc,eip,nat-gateway,ec2-subnet,elb,elbv2,internet-gateway,network-interface,security-group
+	DISABLE_TELEMETRY=true $(CLOUDNUKE) aws --region $$AWS_REGION --force --config config/dev/cloud_nuke.yaml --resource-type vpc,eip,nat-gateway,ec2,ec2-subnet,elb,elbv2,ebs,internet-gateway,network-interface,security-group
 	@rm config/dev/cloud_nuke.yaml
-	@CLUSTER_NAME=$(CLUSTER_NAME) YQ=$(YQ) AWSCLI=$(AWSCLI) bash -c ./scripts/aws-nuke-ccm.sh
+	@CLUSTER_NAME=$(CLUSTER_NAME) YQ=$(YQ) AWSCLI=$(AWSCLI) bash -c "./scripts/aws-nuke-ccm.sh ebs"
 
 .PHONY: cli-install
-cli-install: clusterawsadm clusterctl cloud-nuke yq awscli ## Install the necessary CLI tools for deployment, development and testing.
+cli-install: clusterawsadm clusterctl cloud-nuke envsubst yq awscli ## Install the necessary CLI tools for deployment, development and testing.
 
 ##@ Dependencies
 
@@ -450,9 +461,21 @@ $(ENVSUBST): | $(LOCALBIN)
 .PHONY: awscli
 awscli: $(AWSCLI)
 $(AWSCLI): | $(LOCALBIN)
-	curl "https://awscli.amazonaws.com/awscli-exe-$(OS)-$(shell uname -m)-$(AWSCLI_VERSION).zip" -o "/tmp/awscliv2.zip"
-	unzip /tmp/awscliv2.zip -d /tmp
-	/tmp/aws/install -i $(LOCALBIN)/aws-cli -b $(LOCALBIN) --update
+	@if [ $(OS) == "linux" ]; then \
+		curl "https://awscli.amazonaws.com/awscli-exe-linux-$(shell uname -m)-$(AWSCLI_VERSION).zip" -o "/tmp/awscliv2.zip"; \
+		unzip /tmp/awscliv2.zip -d /tmp; \
+		/tmp/aws/install -i $(LOCALBIN)/aws-cli -b $(LOCALBIN) --update; \
+	fi; \
+	if [ $(OS) == "darwin" ]; then \
+			curl "https://awscli.amazonaws.com/AWSCLIV2.pkg" -o "AWSCLIV2.pkg"; \
+			installer -pkg AWSCLIV2.pkg -target $(LOCALBIN) -applyChoiceChangesXML choices.xml; \
+			rm AWSCLIV2.pkg; \
+	fi; \
+	if [ $(OS) == "windows" ]; then \
+		echo "Installing to $(LOCALBIN) on Windows is not yet implemented"; \
+		exit 1; \
+	fi; \
+
 
 # go-install-tool will 'go install' any package with custom target and name of binary, if it doesn't exist
 # $1 - target path with name of binary (ideally with version)

diff --git a/docs/aws/hosted-control-plane.md b/docs/aws/hosted-control-plane.md
@@ -19,7 +19,12 @@ reused with a management cluster.
 If you deployed your AWS Kubernetes cluster using Cluster API Provider AWS (CAPA)
 you can obtain all the necessary data with the commands below or use the
 template found below in the
-[HMC ManagedCluster manifest generation](#hmc-managed-cluster-manifest-generation) section.
+[HMC ManagedCluster manifest
+generation](#hmc-managed-cluster-manifest-generation) section.
+
+If using the `aws-standalone-cp` template to deploy a hosted cluster it is
+recommended to use a `t3.large` or larger instance type as the `hmc-controller`
+and other provider controllers will need a large amount of resources to run.
 
 **VPC ID**
 
@@ -89,7 +94,7 @@ Grab the following `ManagedCluster` manifest template and save it to a file name
 apiVersion: hmc.mirantis.com/v1alpha1
 kind: ManagedCluster
 metadata:
-  name: aws-hosted-cp
+  name: aws-hosted
 spec:
   template: aws-hosted-cp
   config:
@@ -109,3 +114,24 @@ Then run the following command to create the `managedcluster.yaml`:
 ```
 kubectl get awscluster cluster -o go-template="$(cat managedcluster.yaml.tpl)" > managedcluster.yaml
 ```
+## Deployment Tips
+* Ensure HMC templates and the controller image are somewhere public and
+  fetchable.
+* For installing the HMC charts and templates from a custom repository, load
+  the `kubeconfig` from the cluster and run the commands:
+
+```
+KUBECONFIG=kubeconfig IMG="ghcr.io/mirantis/hmc/controller-ci:v0.0.1-179-ga5bdf29" REGISTRY_REPO="oci://ghcr.io/mirantis/hmc/charts-ci" make dev-apply
+KUBECONFIG=kubeconfig make dev-templates
+```
+* The infrastructure will need to manually be marked `Ready` to get the
+  `MachineDeployment` to scale up.  You can patch the `AWSCluster` kind using
+  the command:
+
+```
+KUBECONFIG=kubeconfig kubectl patch AWSCluster <hosted-cluster-name> --type=merge --subresource status --patch 'status: {ready: true}' -n hmc-system
+```
+
+For additional information on why this is required [click here](https://docs.k0smotron.io/stable/capi-aws/#:~:text=As%20we%20are%20using%20self%2Dmanaged%20infrastructure%20we%20need%20to%20manually%20mark%20the%20infrastructure%20ready.%20This%20can%20be%20accomplished%20using%20the%20following%20command).
+
+
diff --git a/docs/dev.md b/docs/dev.md
@@ -107,3 +107,34 @@ export KUBECONFIG=~/.kube/config
 kubectl --kubeconfig ~/.kube/config get secret -n hmc-system <managedcluster-name>-kubeconfig -o=jsonpath={.data.value} | base64 -d > kubeconfig
 ```
 
+## Running E2E tests locally
+E2E tests can be ran locally via the `make test-e2e` target.  In order to have
+CI properly deploy a non-local registry will need to be used and the Helm charts
+and hmc-controller image will need to exist on the registry, for example, using
+GHCR:
+
+```
+IMG="ghcr.io/mirantis/hmc/controller-ci:v0.0.1-179-ga5bdf29" \
+    REGISTRY_REPO="oci://ghcr.io/mirantis/hmc/charts-ci" \
+    make test-e2e
+```
+
+Optionally, the `NO_CLEANUP=1` env var can be used to disable `After` nodes from
+running within some specs, this will allow users to debug tests by re-running
+them without the need to wait a while for an infrastructure deployment to occur.
+For subsequent runs the `MANAGED_CLUSTER_NAME=<cluster name>` env var should be
+passed to tell the test what cluster name to use so that it does not try to
+generate a new name and deploy a new cluster.
+
+Tests that run locally use autogenerated names like `12345678-e2e-test` while
+tests that run in CI use names such as `ci-1234567890-e2e-test`.  You can always
+pass `MANAGED_CLUSTER_NAME=` from the get-go to customize the name used by the
+test.
+
+### Nuke created resources
+In CI we run `make dev-aws-nuke` to cleanup test resources, you can do so
+manually with:
+
+```
+CLUSTER_NAME=example-e2e-test make dev-aws-nuke
+```
diff --git a/scripts/aws-nuke-ccm.sh b/scripts/aws-nuke-ccm.sh
@@ -33,28 +33,30 @@ if [ -z $AWSCLI ]; then
     exit 1
 fi
 
-echo "Checking for ELB with 'kubernetes.io/cluster/$CLUSTER_NAME' tag"
-for LOADBALANCER in $($AWSCLI elb describe-load-balancers --output yaml | $YQ '.LoadBalancerDescriptions[].LoadBalancerName');
-do
-    echo "Checking ELB: $LOADBALANCER for 'kubernetes.io/cluster/$CLUSTER_NAME tag"
-    DESCRIBE_TAGS=$($AWSCLI elb describe-tags \
-        --load-balancer-names $LOADBALANCER \
-        --output yaml | $YQ '.TagDescriptions[].Tags.[]' | grep 'kubernetes.io/cluster/$CLUSTER_NAME')
-    if [ ! -z "${DESCRIBE_TAGS}" ]; then
-        echo "Deleting ELB: $LOADBALANCER"
-        $AWSCLI elb delete-load-balancer --load-balancer-name $LOADBALANCER
-    fi
-done
+if [ "$1" == "elb" ]; then
+    echo "Checking for ELB with '$CLUSTER_NAME' tag"
+    for LOADBALANCER in $($AWSCLI elb describe-load-balancers --output yaml | $YQ '.LoadBalancerDescriptions[].LoadBalancerName');
+    do
+        echo "Checking ELB: $LOADBALANCER for tag"
+        DESCRIBE_TAGS=$($AWSCLI elb describe-tags --load-balancer-names $LOADBALANCER --output yaml | $YQ '.TagDescriptions[]' | grep $CLUSTER_NAME)
+        if [ ! -z "${DESCRIBE_TAGS}" ]; then
+            echo "Deleting ELB: $LOADBALANCER"
+            $AWSCLI elb delete-load-balancer --load-balancer-name $LOADBALANCER
+        fi
+    done
+fi
 
-echo "Checking for EBS Volumes with $CLUSTER_NAME within the 'kubernetes.io/created-for/pvc/name' tag"
-for VOLUME in $($AWSCLI  ec2 describe-volumes --output yaml | $YQ '.Volumes[].VolumeId');
-do
-    echo "Checking EBS Volume: $VOLUME for $CLUSTER_NAME claim"
-    DESCRIBE_VOLUMES=$($AWSCLI ec2 describe-volumes \
-        --volume-id $VOLUME \
-        --output yaml | $YQ '.Volumes | to_entries[] | .value.Tags[] | select(.Key == "kubernetes.io/created-for/pvc/name")' | grep $CLUSTER_NAME)
-    if [ ! -z "${DESCRIBE_VOLUMES}" ]; then
-        echo "Deleting EBS Volume: $VOLUME"
-        $AWSCLI ec2 delete-volume --volume-id $VOLUME
-    fi
-done
+if [ "$1" == "ebs" ]; then
+    echo "Checking for EBS Volumes with '$CLUSTER_NAME' within the 'kubernetes.io/created-for/pvc/name' tag"
+    for VOLUME in $($AWSCLI  ec2 describe-volumes --output yaml | $YQ '.Volumes[].VolumeId');
+    do
+        echo "Checking EBS Volume: $VOLUME for $CLUSTER_NAME claim"
+        DESCRIBE_VOLUMES=$($AWSCLI ec2 describe-volumes \
+            --volume-id $VOLUME \
+            --output yaml | $YQ '.Volumes | to_entries[] | .value.Tags[] | select(.Key == "kubernetes.io/created-for/pvc/name")' | grep $CLUSTER_NAME)
+        if [ ! -z "${DESCRIBE_VOLUMES}" ]; then
+            echo "Deleting EBS Volume: $VOLUME"
+            $AWSCLI ec2 delete-volume --volume-id $VOLUME
+        fi
+    done
+fi
diff --git a/templates/cluster/aws-hosted-cp/Chart.yaml b/templates/cluster/aws-hosted-cp/Chart.yaml
@@ -1,18 +1,18 @@
 apiVersion: v2
 name: aws-hosted-cp
-description: | 
+description: |
   An HMC template to deploy a k8s cluster on AWS with control plane components
   within the management cluster.
 type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.1.2
+version: 0.1.3
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "1.30.2+k0s.0"
+appVersion: '1.30.4+k0s.0'
 annotations:
   hmc.mirantis.com/infrastructure-providers: aws
   hmc.mirantis.com/controlplane-providers: k0smotron

diff --git a/templates/cluster/aws-hosted-cp/templates/awscluster.yaml b/templates/cluster/aws-hosted-cp/templates/awscluster.yaml
@@ -4,6 +4,7 @@ metadata:
   name: {{ include "cluster.name" . }}
   annotations:
     cluster.x-k8s.io/managed-by: k0smotron
+    aws.cluster.x-k8s.io/external-resource-gc: "true"
   finalizers:
   - hmc.mirantis.com/cleanup
 spec:

diff --git a/templates/cluster/aws-hosted-cp/templates/k0smotroncontrolplane.yaml b/templates/cluster/aws-hosted-cp/templates/k0smotroncontrolplane.yaml
@@ -46,7 +46,8 @@ spec:
                 - --cluster-name={{ include "cluster.name" . }}
               # Removing the default `node-role.kubernetes.io/control-plane` node selector
               # TODO: it does not work
-              # nodeSelector: ""
+              nodeSelector:
+                node-role.kubernetes.io/control-plane: null
           - name: aws-ebs-csi-driver
             namespace: kube-system
             chartname: aws-ebs-csi-driver/aws-ebs-csi-driver