feat(chart): initial helm chart for the operator (without the collector)

.github/workflows/verify.yaml

@@ -50,6 +50,11 @@ jobs:
         run: |
           make lint
 
+      - name: install Helm unittest plugin
+        shell: bash
+        run: |
+          helm plugin install https://github.com/helm-unittest/helm-unittest.git
+
       - name: test
         run: |
           make test

CONTRIBUTING.md

@@ -6,6 +6,8 @@ Contributing
 - Docker
 - kubectl version v1.11.3+.
 - Access to a Kubernetes v1.11.3+ cluster.
+- [helm](https://helm.sh/docs/intro/install/)
+- [helm unittest plug-in](https://github.com/helm-unittest/helm-unittest/tree/main)
 
 ## Deploying to a Local Cluster for Testing Purposes
 
@@ -54,31 +56,42 @@ In this case your test cluster needs to be configured to have pull access for th
 
 After that, you can deploy the operator to your cluster:
 
-* Install the CRDs into the cluster: `make install`
-* Deploy the locally built image `dash0-operator-controller:latest` to the cluster: `make deploy`
-* Alternatively, deploy the image pushed to the remote registry with the image specified by `IMG`: `make deploy IMG=<some-registry>/dash0-operator:tag`
+* Deploy the locally built image `dash0-operator-controller:latest` to the cluster: `make deploy-via-helm`
+  (or `make deploy-via-kustomize`)
+* Alternatively, deploy the image pushed to the remote registry with the image specified by `IMG`:
+  `make deploy-via-helm IMG=<some-registry>/dash0-operator:tag`
+  (or `make deploy-via-kustomize IMG=<some-registry>/dash0-operator:tag`)
+* No matter if you deploy via helm or kustomize, the custom resource definition will automatically be installed when
+  deploying the operator. However, you can also do that separately via kustomize if required via `make install`.
 
 **NOTE**: If you encounter RBAC errors, you may need to grant yourself cluster-admin privileges or be logged in as
 admin.
 
-**Delete the APIs(CRDs) from the cluster:**
+**Undeploy the controller from the cluster:**
 
 ```sh
-make uninstall
+make undeploy-via-helm
 ```
 
-**UnDeploy the controller from the cluster:**
+or
 
 ```sh
-make undeploy
+make undeploy-via-kustomize
 ```
 
+When undeploying the controllor, the same tool (helm vs. kustomiz) should be used as when deploying it.
+
+This will also remove the custom resource definition. However, the custom resource definition can also be removed
+separately via `make uninstall` without removing the operator.
+
 ## Run Tests
 
 ```
 make test
 ```
 
+This will run the go unit tests as well as the helm chart tests.
+
 ### End-to-End Tests
 
 The end-to-end tests currently only support Kubernetes via Docker Desktop on Mac.
@@ -96,4 +109,3 @@ More information can be found via the [Kubebuilder Documentation](https://book.k
 ## Contributing
 
 No contribution guidelines are available at this point.
-

Makefile

@@ -50,8 +50,11 @@ endif
 # This is useful for CI or a project to utilize a specific version of the operator-sdk toolkit.
 OPERATOR_SDK_VERSION ?= v1.34.1
 
-# Image URL to use all building/pushing image targets
-IMG ?= dash0-operator-controller:latest
+# image repository and tag to use for building/pushing the operator image
+IMG_REPOSITORY ?= dash0-operator-controller
+IMG_TAG ?= latest
+IMG ?= $(IMG_REPOSITORY):$(IMG_TAG)
+
 # ENVTEST_K8S_VERSION refers to the version of kubebuilder assets to be downloaded by envtest binary.
 ENVTEST_K8S_VERSION = 1.28.3
 
@@ -114,6 +117,7 @@ vet: ## Run go vet against code.
 .PHONY: test
 test: manifests generate fmt vet envtest ## 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
+	cd helm-chart/dash0-operator && helm unittest -f 'tests/**/*.yaml' .
 
 # Invoking ginkgo via go run makes sure we use the version from go.mod and not a version installed globally, which
 # would be used when simply running `ginkgo -v test/e2e`. An alternative would be to invoke ginkgo via go test, that
@@ -193,15 +197,24 @@ uninstall: manifests kustomize ## Uninstall CRDs from the K8s cluster specified
 	sleep 1
 	$(KUSTOMIZE) build config/crd | $(KUBECTL) patch CustomResourceDefinition dash0s.operator.dash0.com -p '{"metadata":{"finalizers":null}}' --type=merge
 
-.PHONY: deploy
-deploy: manifests kustomize ## Deploy controller to the K8s cluster specified in ~/.kube/config.
+.PHONY: deploy-via-kustomize
+deploy-via-kustomize: manifests kustomize ## Deploy the controller via kustomize to the K8s cluster specified in ~/.kube/config.
 	cd config/manager && $(KUSTOMIZE) edit set image controller=${IMG}
 	$(KUSTOMIZE) build config/default | $(KUBECTL) apply -f -
 
-.PHONY: undeploy
-undeploy: ## Undeploy controller from the K8s cluster specified in ~/.kube/config. Call with ignore-not-found=true to ignore resource not found errors during deletion.
+.PHONY: undeploy-via-kustomize
+undeploy-via-kustomize: ## Undeploy the controller via kustomize from the K8s cluster specified in ~/.kube/config. Call with ignore-not-found=true to ignore resource not found errors during deletion.
 	$(KUSTOMIZE) build config/default | $(KUBECTL) delete --ignore-not-found=$(ignore-not-found) --wait=false -f -
 
+.PHONY: deploy-via-helm
+deploy-via-helm: ## Deploy the controller via helm to the K8s cluster specified in ~/.kube/config.
+	helm install --namespace dash0-operator-system --create-namespace --set operator.image=${IMG} --set operator.imagePullPolicy=Never dash0-operator helm-chart/dash0-operator
+
+.PHONY: undeploy-via-helm
+undeploy-via-helm: ## Undeploy the controller via helm from the K8s cluster specified in ~/.kube/config. Call with ignore-not-found=true to ignore resource not found errors during deletion.
+	helm uninstall --namespace dash0-operator-system dash0-operator
+	$(KUBECTL) delete ns dash0-operator-system
+
 ##@ Build Dependencies
 
 ## Location to install dependencies to

README.md

@@ -1,14 +1,15 @@
 # Dash0 Kubernetes Operator
 
-The Dash0 Kubernetes Operator makes observability easy for every Kubernetes setup.
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+
+The Dash0 Kubernetes Operator makes observability for Kubernetes _easy_.
 Simply install the operator into your cluster to get OpenTelemetry data flowing from your applications and
 infrastructure to Dash0.
 
 ## Description
 
-The Dash0 Kubernetes operator installs an OpenTelemetry collector into your cluster that sends data to your Dash0
-ingress endpoint, with authentication already configured out of the box. Additionally, it will enable gathering
-OpenTelemetry data from applications deployed to the cluster for a selection of supported runtimes.
+The Dash0 Kubernetes operator enables gathering OpenTelemetry data from your workloads for a selection of supported
+runtimes. 
 
 The Dash0 Kubernetes operator is currently in beta.
 
@@ -18,5 +19,5 @@ Supported runtimes:
 
 ## Getting Started
 
-TODO Describe installation via Helm etc.
-
+The preferred method of installation is via the operator's
+[Helm chart](https://github.com/dash0hq/dash0-operator/helm-chart/dash0-operator/README.md).

cmd/main.go

@@ -13,6 +13,7 @@ import (
 	// to ensure that exec-entrypoint and run can make use of them.
 	_ "k8s.io/client-go/plugin/pkg/client/auth"
 
+	corev1 "k8s.io/api/core/v1"
 	"k8s.io/apimachinery/pkg/runtime"
 	utilruntime "k8s.io/apimachinery/pkg/util/runtime"
 	"k8s.io/client-go/kubernetes"
@@ -138,34 +139,53 @@ func startOperatorManager(
 		return fmt.Errorf("unable to create the clientset client")
 	}
 
-	operatorVersion, isSet := os.LookupEnv("DASH0_OPERATOR_VERSION")
+	operatorImage, isSet := os.LookupEnv("DASH0_OPERATOR_IMAGE")
 	if !isSet {
-		operatorVersion = "unknown"
+		return fmt.Errorf("cannot start Dash0 operator, the mandatory environment variable " +
+			"\"DASH0_OPERATOR_IMAGE\" is missing")
 	}
-	initContainerImageVersion, isSet := os.LookupEnv("DASH0_INIT_CONTAINER_IMAGE_VERSION")
+
+	initContainerImage, isSet := os.LookupEnv("DASH0_INIT_CONTAINER_IMAGE")
 	if !isSet {
 		return fmt.Errorf("cannot start Dash0 operator, the mandatory environment variable " +
-			"\"DASH0_INIT_CONTAINER_IMAGE_VERSION\" is missing")
+			"\"DASH0_INIT_CONTAINER_IMAGE\" is missing")
+	}
+
+	initContainerImagePullPolicyRaw := os.Getenv("DASH0_INIT_CONTAINER_IMAGE_PULL_POLICY")
+	var initContainerImagePullPolicy corev1.PullPolicy
+	if initContainerImagePullPolicyRaw == "" {
+		if initContainerImagePullPolicyRaw == string(corev1.PullAlways) ||
+			initContainerImagePullPolicyRaw == string(corev1.PullIfNotPresent) ||
+			initContainerImagePullPolicyRaw == string(corev1.PullNever) {
+			initContainerImagePullPolicy = corev1.PullPolicy(initContainerImagePullPolicyRaw)
+		} else {
+			setupLog.Info(
+				fmt.Sprintf(
+					"Ignoring unknown pull policy for init container image: %s.", initContainerImagePullPolicyRaw))
+		}
 	}
 	setupLog.Info(
 		"version information",
-		"operator version",
-		operatorVersion,
-		"init container image version",
-		initContainerImageVersion,
+		"operator image and version",
+		operatorImage,
+		"init container image and version",
+		initContainerImage,
+		"init container image pull policy override",
+		initContainerImagePullPolicy,
 	)
 
-	versions := util.Versions{
-		OperatorVersion:           operatorVersion,
-		InitContainerImageVersion: initContainerImageVersion,
+	images := util.Images{
+		OperatorImage:                operatorImage,
+		InitContainerImage:           initContainerImage,
+		InitContainerImagePullPolicy: initContainerImagePullPolicy,
 	}
 
 	if err = (&controller.Dash0Reconciler{
 		Client:    mgr.GetClient(),
 		ClientSet: clientSet,
 		Scheme:    mgr.GetScheme(),
 		Recorder:  mgr.GetEventRecorderFor("dash0-controller"),
-		Versions:  versions,
+		Images:    images,
 	}).SetupWithManager(mgr); err != nil {
 		return fmt.Errorf("unable to set up the Dash0 reconciler: %w", err)
 	}
@@ -174,7 +194,7 @@ func startOperatorManager(
 	if os.Getenv("ENABLE_WEBHOOKS") != "false" {
 		if err = (&dash0webhook.Handler{
 			Recorder: mgr.GetEventRecorderFor("dash0-webhook"),
-			Versions: versions,
+			Images:   images,
 		}).SetupWebhookWithManager(mgr); err != nil {
 			return fmt.Errorf("unable to create the Dash0 webhook: %w", err)
 		}

config/manager/manager.yaml

@@ -48,10 +48,10 @@ spec:
         image: dash0-operator-controller:latest
         name: manager
         env:
-          - name: DASH0_OPERATOR_VERSION
-            value: 1.0.0
-          - name: DASH0_INIT_CONTAINER_IMAGE_VERSION
-            value: 1.0.0
+          - name: DASH0_OPERATOR_IMAGE
+            value: dash0-operator-controller:1.0.0
+          - name: DASH0_INIT_CONTAINER_IMAGE
+            value: dash0-instrumentation:1.0.0
 
         # Note: Use "imagePullPolicy: Never" when only building the image locally without pushing them anywhere. Omit
         # the attribute otherwise to use the default pull policy.

helm-chart/dash0-operator/.helmignore

@@ -0,0 +1,8 @@
+.DS_Store
+.gitignore
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+tests/

helm-chart/dash0-operator/Chart.yaml

@@ -0,0 +1,22 @@
+apiVersion: v2
+name: dash0-operator
+version: "1.0.0"
+description: The Dash0 Kubernetes Operator makes observability easy for every Kubernetes setup, simply install the operator into your cluster to get OpenTelemetry data flowing from your applications and infrastructure to Dash0.
+type: application
+keywords:
+  - OpenTelemetry
+  - Observability
+  - Monitoring
+  - Dash0
+  - Node.js
+  # - Python
+  # - Java
+  # - .NET
+home: https://www.dash0.com/
+sources:
+  - https://github.com/dash0hq/dash0-operator
+maintainers:
+  - name: Bastian Krol
+    email: [email protected]
+icon: icon/logo.svg
+appVersion: "1.0.0"