Add management support for the multicluster-observability-operator

CONTRIBUTING.md

@@ -66,28 +66,53 @@ Then every time you want to test a new version, you can just:
 ```shell
 make oci
 # Delete the mcoa pod which will make the Deployment pull the new image
-oc -n open-cluster-management delete pod -l app=multicluster-observability-addon-manager
+oc -n open-cluster-management-observability delete pod -l app=multicluster-observability-addon-manager
 ```
 
-### Disabeling specific signals 
+### Enable specific Observability Capabilities
 
-The addon supports disabling signals using the resource `AddOnDeploymentConfig`. For instance, to disable the logging signal create the following resource on the hub cluster:
+The addon supports enabling observability capabilities using the resource `AddOnDeploymentConfig`. For instance, to enable platform and user workloads logging/tracing/instrumentation create the following resource on the hub cluster:
 
 ```yaml
 apiVersion: addon.open-cluster-management.io/v1alpha1
 kind: AddOnDeploymentConfig
 metadata:
   name: multicluster-observability-addon
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   customizedVariables:
-    - name: loggingDisabled
-      value: "true"
+  # Platform Observability
+  - name: platformLogsCollection
+    value: clusterlogforwarders.v1.logging.openshift.io
+  # User Workloads Observability
+  - name: userWorkloadLogsCollection
+    value: clusterlogforwarders.v1.logging.openshift.io
+  - name: userWorkloadTracesCollection
+    value: opentelemetrycollectors.v1beta1.opentelemetry.io
+  - name: userWorkloadInstrumentation
+    value: instrumentations.v1alpha1.opentelemetry.io
 ``` 
 
-Supported keys are `metricsDisabled`, `loggingDisabled` and `tracingDisabled`
+Supported keys are:
+- `platformLogsCollection`: Supports values `clusterlogforwarders.v1.logging.openshift.io`
+- `userWorkloadLogsCollection`: Supports values `clusterlogforwarders.v1.logging.openshift.io`
+- `userWorkloadTracesCollection`: Supports values `opentelemetrycollectors.v1beta1.opentelemetry.io`
+- `userWorkloadTracesInstrumentation`: Supports values `instrumentations.v1alpha1.opentelemetry.io`
 
-## Install the addon on a Spoke Cluster
+__Note__: Some keys can hold multiple values separated by semicolon to support multiple data collection capabilities in parallel, e.g:
+
+```yaml
+apiVersion: addon.open-cluster-management.io/v1alpha1
+kind: AddOnDeploymentConfig
+metadata:
+  name: multicluster-observability-addon
+  namespace: open-cluster-management-observability
+spec:
+  customizedVariables:
+  # User Workloads Observability with multiple collectors
+  - name: userWorkloadLogsCollection
+    value: clusterlogforwarders.v1.logging.openshift.io;opentelemetrycollectors.v1beta1.opentelemetry.io
+```
 
 The addon installation is managed by the addon-manager. This means that users
 don't need to explicetelly create resources to install the addon on spoke
@@ -118,7 +143,7 @@ This MCOA supports all outputs defined in [OpenShift Documentation](https://docs
 
 Note: the service account used by the `ClusterLogForwarder` deployed by MCOA is `openshift-logging/mcoa-logcollector`, this information is esential when using the AWS STS authentication.
 
-### Configuring Traces
+### Traces Collection
 
 Currently MCOA supports deploying a single instance of `OpenTelemetryCollector`
 templated with the stanza created in the hub cluster. The instance deployed in

demo/addon-config/charts/logging/templates/clusterlogforwarder.yaml

@@ -2,7 +2,7 @@ apiVersion: logging.openshift.io/v1
 kind: ClusterLogForwarder
 metadata:
   name: instance
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   outputs:
 {{- range $_, $dic := .Values.outputs }}
@@ -38,4 +38,4 @@ spec:
 {{- end }}
      outputRefs:
      - {{ $dic.name }}
-{{- end }}
+{{- end }}

demo/addon-config/charts/logging/templates/logging-auth-configmap.yaml

@@ -3,7 +3,7 @@ apiVersion: v1
 kind: ConfigMap
 metadata:
   name: logging-auth
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
   labels:
     mcoa.openshift.io/signal: logging
 data:
@@ -15,4 +15,4 @@ data:
 {{- end }}
 {{- end }}
 ---
-{{- end }}
+{{- end }}

demo/addon-config/charts/tracing/templates/otel-collector.yaml

@@ -2,7 +2,7 @@ apiVersion: opentelemetry.io/v1alpha1
 kind: OpenTelemetryCollector
 metadata:
   name: spoke-otelcol
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   config: |
     receivers:

demo/addon-config/charts/tracing/templates/tracing-auth-configmap.yaml

@@ -3,9 +3,9 @@ apiVersion: v1
 kind: ConfigMap
 metadata:
   name: tracing-auth
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
   labels:
     mcoa.openshift.io/signal: tracing
 data:
   otlp: mTLS
-{{- end }}
+{{- end }}

demo/addon-install/templates/managed-cluster-addon.yaml

@@ -11,7 +11,7 @@ spec:
   # Logging Auth ConfigMap
   - resource: configmaps
     name: logging-auth
-    namespace: open-cluster-management
+    namespace: open-cluster-management-observability
   # Logging URLs for Loki ConfigMap
 {{- range $_, $dic := $.Values.logging.outputs }}
 {{- if eq $dic.type "loki" }}
@@ -38,10 +38,10 @@ spec:
   # Tracing Auth ConfigMap
   - resource: configmaps
     name: tracing-auth
-    namespace: open-cluster-management
+    namespace: open-cluster-management-observability
   # Tracing ca-bundle configmap
   - resource: secrets
     name: otel-gateway
     namespace: observability
 {{- end }}
-{{- end }}
+{{- end }}

demo/mcoa-demo/templates/logging-static-auth.yaml

@@ -3,8 +3,8 @@ apiVersion: v1
 kind: Secret
 metadata:
   name: static-authentication
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 data:
   aws_access_key_id: {{ .Values.logging.aws.keyID | b64enc }}
   aws_secret_access_key: {{ .Values.logging.aws.keySecret | b64enc }} 
-{{- end }}
+{{- end }}

deploy/kustomization.yaml

@@ -3,7 +3,7 @@ images:
   newName: quay.io/rhobs/multicluster-observability-addon
   newTag: v0.0.1
 
-namespace: open-cluster-management
+namespace: open-cluster-management-observability
 
 resources:
 - resources/cluster_role_binding.yaml

deploy/resources/addondeploymentconfig.yaml

@@ -2,8 +2,19 @@ apiVersion: addon.open-cluster-management.io/v1alpha1
 kind: AddOnDeploymentConfig
 metadata:
   name: multicluster-observability-addon
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   customizedVariables:
-    - name: loggingSubscriptionChannel
-      value: stable-5.9
+  # Operator Subscription Channels
+  - name: openshiftLoggingChannel
+    value: stable-5.9
+  # Platform Observability
+  - name: platformLogsCollection
+    value: clusterlogforwarders.v1.logging.openshift.io
+  # User Workloads Observability
+  - name: userWorkloadLogsCollection
+    value: clusterlogforwarders.v1.logging.openshift.io
+  - name: userWorkloadTracesCollection
+    value: opentelemetrycollectors.v1beta1.opentelemetry.io
+  - name: userWorkloadInstrumentation
+    value: instrumentations.v1alpha1.opentelemetry.io

deploy/resources/cluster-management-addon.yaml

@@ -16,7 +16,7 @@ spec:
       resource: addondeploymentconfigs
       defaultConfig:
         name: multicluster-observability-addon
-        namespace: open-cluster-management
+        namespace: open-cluster-management-observability
     # Describes the default log forwarding outputs for each log type applied to all managed clusters.
     - group: logging.openshift.io
       resource: clusterlogforwarders
@@ -32,8 +32,8 @@ spec:
           - group: logging.openshift.io
             resource: clusterlogforwarders
             name: instance
-            namespace: open-cluster-management
+            namespace: open-cluster-management-observability
           - group: opentelemetry.io
             resource: opentelemetrycollectors
             name: instance
-            namespace: open-cluster-management
+            namespace: open-cluster-management-observability

hack/addon-install/templates/aws-secret-default.yaml

@@ -2,7 +2,7 @@ apiVersion: v1
 kind: Secret
 metadata:
   name: aws-credentials-default
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 type: Opaque
 data:
   aws_access_key_id: {{ .Values.awsCredentials.accessKeyID | b64enc }}

hack/addon-install/templates/clf-instance.yaml

@@ -2,7 +2,7 @@ apiVersion: logging.openshift.io/v1
 kind: ClusterLogForwarder
 metadata:
   name: instance
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   outputs:
    - cloudwatch:

hack/addon-install/templates/instance-default.yaml

@@ -2,7 +2,7 @@ apiVersion: logging.openshift.io/v1
 kind: ClusterLogForwarder
 metadata:
   name: instance-default
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   outputs:
    - cloudwatch:
@@ -17,4 +17,4 @@ spec:
      inputRefs:
      - infrastructure
      outputRefs:
-     - cw
+     - cw

hack/addon-install/templates/otelcol-instance.yaml

@@ -2,7 +2,7 @@ apiVersion: opentelemetry.io/v1beta1
 kind: OpenTelemetryCollector
 metadata:
   name: instance
-  namespace: open-cluster-management
+  namespace: open-cluster-management-observability
 spec:
   config:
     exporters:

internal/addon/helm/values.go

@@ -2,14 +2,12 @@ package helm
 
 import (
 	"context"
-	"strconv"
 
 	"github.com/rhobs/multicluster-observability-addon/internal/addon"
 	lhandlers "github.com/rhobs/multicluster-observability-addon/internal/logging/handlers"
 	lmanifests "github.com/rhobs/multicluster-observability-addon/internal/logging/manifests"
 	thandlers "github.com/rhobs/multicluster-observability-addon/internal/tracing/handlers"
 	tmanifests "github.com/rhobs/multicluster-observability-addon/internal/tracing/manifests"
-	"k8s.io/klog/v2"
 	"open-cluster-management.io/addon-framework/pkg/addonfactory"
 	addonutils "open-cluster-management.io/addon-framework/pkg/utils"
 	addonapiv1alpha1 "open-cluster-management.io/api/addon/v1alpha1"
@@ -25,36 +23,35 @@ type HelmChartValues struct {
 	Tracing tmanifests.TracingValues `json:"tracing"`
 }
 
-type Options struct {
-	LoggingDisabled bool
-	TracingDisabled bool
-}
-
 func GetValuesFunc(ctx context.Context, k8s client.Client) addonfactory.GetValuesFunc {
 	return func(
 		cluster *clusterv1.ManagedCluster,
-		addon *addonapiv1alpha1.ManagedClusterAddOn,
+		mcAddon *addonapiv1alpha1.ManagedClusterAddOn,
 	) (addonfactory.Values, error) {
 		// if hub cluster, then don't install anything
 		if isHubCluster(cluster) {
 			return addonfactory.JsonStructToValues(HelmChartValues{})
 		}
 
-		aodc, err := getAddOnDeploymentConfig(ctx, k8s, addon)
+		aodc, err := getAddOnDeploymentConfig(ctx, k8s, mcAddon)
 		if err != nil {
 			return nil, err
 		}
-		opts, err := buildOptions(aodc)
+		opts, err := addon.BuildOptions(aodc)
 		if err != nil {
 			return nil, err
 		}
 
+		if !opts.Platform.Enabled && !opts.UserWorkloads.Enabled {
+			return addonfactory.JsonStructToValues(HelmChartValues{})
+		}
+
 		userValues := HelmChartValues{
 			Enabled: true,
 		}
 
-		if !opts.LoggingDisabled {
-			loggingOpts, err := lhandlers.BuildOptions(ctx, k8s, addon, aodc)
+		if opts.Platform.Logs.CollectionEnabled || opts.UserWorkloads.Logs.CollectionEnabled {
+			loggingOpts, err := lhandlers.BuildOptions(ctx, k8s, mcAddon, opts.Platform.Logs, opts.UserWorkloads.Logs)
 			if err != nil {
 				return nil, err
 			}
@@ -66,9 +63,8 @@ func GetValuesFunc(ctx context.Context, k8s client.Client) addonfactory.GetValue
 			userValues.Logging = *logging
 		}
 
-		if !opts.TracingDisabled {
-			klog.Info("Tracing enabled")
-			tracingOpts, err := thandlers.BuildOptions(ctx, k8s, addon, aodc)
+		if opts.UserWorkloads.Traces.CollectionEnabled {
+			tracingOpts, err := thandlers.BuildOptions(ctx, k8s, mcAddon, opts.UserWorkloads.Traces)
 			if err != nil {
 				return nil, err
 			}
@@ -94,35 +90,6 @@ func getAddOnDeploymentConfig(ctx context.Context, k8s client.Client, mcAddon *a
 	return addOnDeployment, nil
 }
 
-func buildOptions(addOnDeployment *addonapiv1alpha1.AddOnDeploymentConfig) (Options, error) {
-	var opts Options
-	if addOnDeployment == nil {
-		return opts, nil
-	}
-
-	if addOnDeployment.Spec.CustomizedVariables == nil {
-		return opts, nil
-	}
-
-	for _, keyvalue := range addOnDeployment.Spec.CustomizedVariables {
-		switch keyvalue.Name {
-		case addon.AdcLoggingDisabledKey:
-			value, err := strconv.ParseBool(keyvalue.Value)
-			if err != nil {
-				return opts, err
-			}
-			opts.LoggingDisabled = value
-		case addon.AdcTracingisabledKey:
-			value, err := strconv.ParseBool(keyvalue.Value)
-			if err != nil {
-				return opts, err
-			}
-			opts.TracingDisabled = value
-		}
-	}
-	return opts, nil
-}
-
 func isHubCluster(cluster *clusterv1.ManagedCluster) bool {
 	val, ok := cluster.Labels[annotationLocalCluster]
 	if !ok {