[opentelemetry-kube-stack]: Merge both collectors in a single one with leader election (#1735)

* opentelemetry-kube-stack: Merge both collectores in a single one with leader election

Signed-off-by: Jorge Turrado <[email protected]>

* update generted manifests

Signed-off-by: Jorge Turrado <[email protected]>

* update generted manifests

Signed-off-by: Jorge Turrado <[email protected]>

* update generted manifests

Signed-off-by: Jorge Turrado <[email protected]>

* bump operator

Signed-off-by: Jorge Turrado <[email protected]>

* extract leader election config

Signed-off-by: Jorge Turrado <[email protected]>

* extract leader election config

Signed-off-by: Jorge Turrado <[email protected]>

* extract leader election config

Signed-off-by: Jorge Turrado <[email protected]>

* extract leader election config

Signed-off-by: Jorge Turrado <[email protected]>

* extract leader election config

Signed-off-by: Jorge Turrado <[email protected]>

* Support a setup without leader election

Signed-off-by: Jorge Turrado <[email protected]>

* Support a setup without leader election

Signed-off-by: Jorge Turrado <[email protected]>

* remove not needed code from example

Signed-off-by: Jorge Turrado <[email protected]>

---------

Signed-off-by: Jorge Turrado <[email protected]>

Author: JorTurFer

charts/opentelemetry-kube-stack/Chart.lock

@@ -15,4 +15,4 @@ dependencies:
   repository: https://prometheus-community.github.io/helm-charts
   version: 4.37.3
 digest: sha256:6bdd281bcc9df8f34dc4d553974a4ceeb9b967e10933648eecbed4f7b32570a1
-generated: "2025-07-14T12:56:09.098151354+02:00"
+generated: "2025-07-15T08:46:49.093988+02:00"

charts/opentelemetry-kube-stack/Chart.yaml

@@ -1,6 +1,6 @@
 apiVersion: v2
 name: opentelemetry-kube-stack
-version: 0.6.3
+version: 0.7.0
 description: |
   OpenTelemetry Quickstart chart for Kubernetes.
   Installs an operator and collector for an easy way to get started with Kubernetes observability.

charts/opentelemetry-kube-stack/README.md

@@ -11,15 +11,15 @@ This Helm chart serves as a quickstart for OpenTelemetry in a Kubernetes environ
 
 ## Features
 
-This chart installs the OpenTelemetry Operator and two collector pools with the following features:
-* Daemonset collector
-  * Kubernetes infrastructure metrics
-  * Applications logs
-  * OTLP trace receiver
-  * Kubernetes resource enrichment
-* Standalone collector
-  * Kubernetes events
-  * Cluster metrics
+This chart installs the OpenTelemetry Operator and a daemonset collector pool with the following features:
+* Kubernetes infrastructure metrics
+* Applications logs
+* OTLP trace receiver
+* Kubernetes resource enrichment
+* Kubernetes events
+* Cluster metrics
+
+**Note**: This setup requires the usage of leader election extension, if this isn't posible for any reason, this extension can be avoided with [this alternative setup](/charts/opentelemetry-kube-stack/examples/no-leader-election-extension/README.md)
 
 ## Usage
 
@@ -139,6 +139,11 @@ Consult also the [Helm Documentation on CRDs](https://helm.sh/docs/chart_best_pr
 
 _See [helm upgrade](https://helm.sh/docs/helm/helm_upgrade/) for command documentation._
 
+### Upgrade from 0.6.x to 0.7.x
+
+Version 0.7.0 has unified the previous collectors (daemonset and deployment) in a single one. If you are using custom configurations for `cluster` collector, you will need to merge your `cluster` collector configuration with `daemon` collector and remove `collectors.cluster` section from your values file.
+If you are using helm, upgrade command is enough the prune old resources, but gitops approaches like 'ArgoCD' could require to select pruning options during sync process to get rid of removed resources.
+
 ## Configuration
 
 The following command will show all the configurable options with detailed comments.

charts/opentelemetry-kube-stack/examples/cloud-demo/rendered/bridge.yaml

@@ -5,7 +5,7 @@ kind: OpAMPBridge
 metadata:
   name: example
   labels:
-    helm.sh/chart: opentelemetry-kube-stack-0.6.3
+    helm.sh/chart: opentelemetry-kube-stack-0.7.0
     app.kubernetes.io/version: "0.127.0"
     app.kubernetes.io/managed-by: Helm
     release: "example"    

charts/opentelemetry-kube-stack/examples/cloud-demo/rendered/clusterrole.yaml

@@ -52,6 +52,18 @@ rules:
 - nonResourceURLs: ["/metrics", "/metrics/cadvisor"]
   verbs: ["get"]
 
+- verbs:
+  - get
+  - list
+  - watch
+  - create
+  - update
+  - patch
+  - delete
+  apiGroups:
+  - coordination.k8s.io
+  resources:
+  - leases
 - apiGroups:
   - ""
   resources:
@@ -135,21 +147,6 @@ rules:
 # Source: opentelemetry-kube-stack/templates/clusterrole.yaml
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRoleBinding
-metadata:
-  name: example-cluster-stats
-roleRef:
-  apiGroup: rbac.authorization.k8s.io
-  kind: ClusterRole
-  name: example-collector
-subjects:
-- kind: ServiceAccount
-  # quirk of the Operator
-  name: "example-cluster-stats-collector"
-  namespace: default
----
-# Source: opentelemetry-kube-stack/templates/clusterrole.yaml
-apiVersion: rbac.authorization.k8s.io/v1
-kind: ClusterRoleBinding
 metadata:
   name: example-daemon
 roleRef:

charts/opentelemetry-kube-stack/examples/cloud-demo/rendered/collector.yaml

@@ -2,190 +2,11 @@
 # Source: opentelemetry-kube-stack/templates/collector.yaml
 apiVersion: opentelemetry.io/v1beta1
 kind: OpenTelemetryCollector
-metadata:
-  name: example-cluster-stats
-  namespace: default
-  labels:
-    helm.sh/chart: opentelemetry-kube-stack-0.6.3
-    app.kubernetes.io/version: "0.127.0"
-    app.kubernetes.io/managed-by: Helm
-    release: "example"        
-    opentelemetry.io/opamp-reporting: "true"
-spec:
-  managementState: managed
-  mode: deployment
-  config:
-    exporters:
-      debug: {}
-      otlp:
-        endpoint: ingest.example.com:443
-        headers:
-          access-token: ${ACCESS_TOKEN}
-    processors:
-      batch:
-        send_batch_max_size: 1500
-        send_batch_size: 1000
-        timeout: 1s
-      k8sattributes:
-        extract:
-          labels:
-          - from: pod
-            key: app.kubernetes.io/name
-            tag_name: service.name
-          - from: pod
-            key: k8s-app
-            tag_name: service.name
-          - from: pod
-            key: app.kubernetes.io/instance
-            tag_name: k8s.app.instance
-          - from: pod
-            key: app.kubernetes.io/version
-            tag_name: service.version
-          - from: pod
-            key: app.kubernetes.io/component
-            tag_name: k8s.app.component
-          metadata:
-          - k8s.namespace.name
-          - k8s.pod.name
-          - k8s.pod.uid
-          - k8s.node.name
-          - k8s.pod.start_time
-          - k8s.deployment.name
-          - k8s.replicaset.name
-          - k8s.replicaset.uid
-          - k8s.daemonset.name
-          - k8s.daemonset.uid
-          - k8s.job.name
-          - k8s.job.uid
-          - k8s.container.name
-          - k8s.cronjob.name
-          - k8s.statefulset.name
-          - k8s.statefulset.uid
-          - container.image.tag
-          - container.image.name
-          - k8s.cluster.uid
-        passthrough: false
-        pod_association:
-        - sources:
-          - from: resource_attribute
-            name: k8s.pod.uid
-        - sources:
-          - from: resource_attribute
-            name: k8s.pod.name
-          - from: resource_attribute
-            name: k8s.namespace.name
-          - from: resource_attribute
-            name: k8s.node.name
-        - sources:
-          - from: resource_attribute
-            name: k8s.pod.ip
-        - sources:
-          - from: resource_attribute
-            name: k8s.pod.name
-          - from: resource_attribute
-            name: k8s.namespace.name
-        - sources:
-          - from: connection
-      resourcedetection/env:
-        detectors:
-        - env
-        override: false
-        timeout: 2s
-    receivers:
-      k8s_cluster:
-        allocatable_types_to_report:
-        - cpu
-        - memory
-        - storage
-        auth_type: serviceAccount
-        collection_interval: 10s
-        node_conditions_to_report:
-        - Ready
-        - MemoryPressure
-        - DiskPressure
-        - NetworkUnavailable
-      k8sobjects:
-        objects:
-        - exclude_watch_type:
-          - DELETED
-          group: events.k8s.io
-          mode: watch
-          name: events
-    service:
-      pipelines:
-        logs:
-          exporters:
-          - debug
-          processors:
-          - k8sattributes
-          - resourcedetection/env
-          - batch
-          receivers:
-          - k8sobjects
-        metrics:
-          exporters:
-          - debug
-          - otlp
-          processors:
-          - k8sattributes
-          - resourcedetection/env
-          - batch
-          receivers:
-          - k8s_cluster
-  replicas: 1
-  imagePullPolicy: IfNotPresent
-  upgradeStrategy: automatic
-  terminationGracePeriodSeconds: 30
-  resources:
-    limits:
-      cpu: 200m
-      memory: 500Mi
-    requests:
-      cpu: 100m
-      memory: 250Mi
-  securityContext:
-    {}
-  env:
-  - name: OTEL_K8S_NODE_NAME
-    valueFrom:
-      fieldRef:
-        fieldPath: spec.nodeName
-  - name: OTEL_K8S_NODE_IP
-    valueFrom:
-      fieldRef:
-        fieldPath: status.hostIP
-  - name: OTEL_K8S_NAMESPACE
-    valueFrom:
-      fieldRef:
-        apiVersion: v1
-        fieldPath: metadata.namespace
-  - name: OTEL_K8S_POD_NAME
-    valueFrom:
-      fieldRef:
-        apiVersion: v1
-        fieldPath: metadata.name
-  - name: OTEL_K8S_POD_IP
-    valueFrom:
-      fieldRef:
-        apiVersion: v1
-        fieldPath: status.podIP
-  - name: OTEL_RESOURCE_ATTRIBUTES
-    value: "k8s.cluster.name=demo"
-  
-  - name: ACCESS_TOKEN
-    valueFrom:
-      secretKeyRef:
-        key: access_token
-        name: otel-collector-secret
----
-# Source: opentelemetry-kube-stack/templates/collector.yaml
-apiVersion: opentelemetry.io/v1beta1
-kind: OpenTelemetryCollector
 metadata:
   name: example-daemon
   namespace: default
   labels:
-    helm.sh/chart: opentelemetry-kube-stack-0.6.3
+    helm.sh/chart: opentelemetry-kube-stack-0.7.0
     app.kubernetes.io/version: "0.127.0"
     app.kubernetes.io/managed-by: Helm
     release: "example"        
@@ -200,6 +21,15 @@ spec:
         endpoint: ingest.example.com:443
         headers:
           access-token: ${ACCESS_TOKEN}
+    extensions:
+      k8s_leader_elector/k8s_cluster:
+        auth_type: serviceAccount
+        lease_name: k8s.cluster.receiver.opentelemetry.io
+        lease_namespace: default
+      k8s_leader_elector/k8s_objects:
+        auth_type: serviceAccount
+        lease_name: k8s.objects.receiver.opentelemetry.io
+        lease_namespace: default
     processors:
       batch:
         send_batch_max_size: 1500
@@ -341,6 +171,27 @@ spec:
               system.memory.utilization:
                 enabled: true
           network: {}
+      k8s_cluster:
+        allocatable_types_to_report:
+        - cpu
+        - memory
+        - storage
+        auth_type: serviceAccount
+        collection_interval: 10s
+        k8s_leader_elector: k8s_leader_elector/k8s_cluster
+        node_conditions_to_report:
+        - Ready
+        - MemoryPressure
+        - DiskPressure
+        - NetworkUnavailable
+      k8sobjects:
+        k8s_leader_elector: k8s_leader_elector/k8s_objects
+        objects:
+        - exclude_watch_type:
+          - DELETED
+          group: events.k8s.io
+          mode: watch
+          name: events
       kubeletstats:
         auth_type: serviceAccount
         collection_interval: 15s
@@ -548,6 +399,9 @@ spec:
               ca_file: /var/run/secrets/kubernetes.io/serviceaccount/ca.crt
               insecure_skip_verify: true
     service:
+      extensions:
+      - k8s_leader_elector/k8s_objects
+      - k8s_leader_elector/k8s_cluster
       pipelines:
         logs:
           exporters:
@@ -560,6 +414,7 @@ spec:
           receivers:
           - otlp
           - filelog
+          - k8sobjects
         metrics:
           exporters:
           - debug
@@ -573,6 +428,7 @@ spec:
           - otlp
           - hostmetrics
           - kubeletstats
+          - k8s_cluster
         traces:
           exporters:
           - debug

charts/opentelemetry-kube-stack/examples/cloud-demo/rendered/hooks.yaml

@@ -62,4 +62,4 @@ spec:
           - -c
           - |
             kubectl delete instrumentations,opampbridges,opentelemetrycollectors \
-              -l helm.sh/chart=opentelemetry-kube-stack-0.6.3
+              -l helm.sh/chart=opentelemetry-kube-stack-0.7.0

charts/opentelemetry-kube-stack/examples/cloud-demo/rendered/instrumentation.yaml

@@ -5,7 +5,7 @@ kind: Instrumentation
 metadata:
   name: example
   labels:
-    helm.sh/chart: opentelemetry-kube-stack-0.6.3
+    helm.sh/chart: opentelemetry-kube-stack-0.7.0
     app.kubernetes.io/version: "0.127.0"
     app.kubernetes.io/managed-by: Helm
     release: "example"    

charts/opentelemetry-kube-stack/examples/no-leader-election-extension/README.md

@@ -0,0 +1,5 @@
+# No leader election extension example
+This example contains files to allow a user to use 2 different collectors to monitor the cluster metrics and k8s api info when leader election extension is not an option (because the collector compillation doesn't include it or because of RBAC limitations). 
+This can be done configuring a second collector to host the receivers and setting `disableLeaderElection: true` for `kubernetesEvents` and `clusterMetrics` presets.
+
+**Disclaimer**: This setup is functional but k8s API metrics and events receivers **ARE NOT in a High Availability** configuration as both run as part of a second collector deployed as `deployment` with single replica to avoid the necessity of using leader election in any way.