Add hpa behavior helm chart (#6393)

* modify hpa yaml

* add HPA behavior support to helm chart

* new helm version 0.33.0 2024-02-20

* update helm chart version

* modify HPA to handle empty policies field

* add tests

* remove trailing spaces

* make docs

operations/helm/charts/grafana-agent/CHANGELOG.md

@@ -10,6 +10,13 @@ internal API changes are not present.
 Unreleased
 ----------
 
+0.33.0 (2024-02-20)
+-------------------
+
+### Features
+
+- Add HPA behavior support for scaling up and down. (@wildum)
+
 ### Enhancements
 
 - Allow setting tlsConfig for serviceMonitor (@captncraig)

operations/helm/charts/grafana-agent/Chart.yaml

@@ -2,7 +2,7 @@ apiVersion: v2
 name: grafana-agent
 description: 'Grafana Agent'
 type: application
-version: 0.32.0
+version: 0.33.0
 appVersion: 'v0.39.2'
 
 dependencies:

operations/helm/charts/grafana-agent/README.md

@@ -1,6 +1,6 @@
 # Grafana Agent Helm chart
 
-![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![Version: 0.32.0](https://img.shields.io/badge/Version-0.32.0-informational?style=flat-square) ![AppVersion: v0.39.2](https://img.shields.io/badge/AppVersion-v0.39.2-informational?style=flat-square)
+![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![Version: 0.33.0](https://img.shields.io/badge/Version-0.33.0-informational?style=flat-square) ![AppVersion: v0.39.2](https://img.shields.io/badge/AppVersion-v0.39.2-informational?style=flat-square)
 
 Helm chart for deploying [Grafana Agent][] to Kubernetes.
 
@@ -74,6 +74,12 @@ use the older mode (called "static mode"), set the `agent.mode` value to
 | controller.autoscaling.enabled | bool | `false` | Creates a HorizontalPodAutoscaler for controller type deployment. |
 | controller.autoscaling.maxReplicas | int | `5` | The upper limit for the number of replicas to which the autoscaler can scale up. |
 | controller.autoscaling.minReplicas | int | `1` | The lower limit for the number of replicas to which the autoscaler can scale down. |
+| controller.autoscaling.scaleDown.policies | list | `[]` | List of policies to determine the scale-down behavior. |
+| controller.autoscaling.scaleDown.selectPolicy | string | `"Max"` | Determines which of the provided scaling-down policies to apply if multiple are specified. |
+| controller.autoscaling.scaleDown.stabilizationWindowSeconds | int | `300` | The duration that the autoscaling mechanism should look back on to make decisions about scaling down. |
+| controller.autoscaling.scaleUp.policies | list | `[]` | List of policies to determine the scale-up behavior. |
+| controller.autoscaling.scaleUp.selectPolicy | string | `"Max"` | Determines which of the provided scaling-up policies to apply if multiple are specified. |
+| controller.autoscaling.scaleUp.stabilizationWindowSeconds | int | `0` | The duration that the autoscaling mechanism should look back on to make decisions about scaling up. |
 | controller.autoscaling.targetCPUUtilizationPercentage | int | `0` | Average CPU utilization across all relevant pods, a percentage of the requested value of the resource for the pods. Setting `targetCPUUtilizationPercentage` to 0 will disable CPU scaling. |
 | controller.autoscaling.targetMemoryUtilizationPercentage | int | `80` | Average Memory utilization across all relevant pods, a percentage of the requested value of the resource for the pods. Setting `targetMemoryUtilizationPercentage` to 0 will disable Memory scaling. |
 | controller.dnsPolicy | string | `"ClusterFirst"` | Configures the DNS policy for the pod. https://kubernetes.io/docs/concepts/services-networking/dns-pod-service/#pod-s-dns-policy |

operations/helm/charts/grafana-agent/ci/create-deployment-autoscaling-values.yaml

@@ -3,6 +3,22 @@ controller:
   type: deployment
   autoscaling:
     enabled: true
+    scaleDown:
+      policies:
+        - type: Pods
+          value: 4
+          periodSeconds: 60
+      selectPolicy: Min
+      stabilizationWindowSeconds: 100
+    scaleUp:
+      policies:
+        - type: Pods
+          value: 4
+          periodSeconds: 60
+        - type: Percent
+          value: 100
+          periodSeconds: 15
+      stabilizationWindowSeconds: 80
 agent:
   resources:
     requests:

operations/helm/charts/grafana-agent/templates/hpa.yaml

@@ -25,6 +25,33 @@ spec:
   {{- with .Values.controller.autoscaling }}
   minReplicas: {{ .minReplicas }}
   maxReplicas: {{ .maxReplicas }}
+  behavior:
+    {{- with .scaleDown }}
+    scaleDown:
+      {{- if .policies }}
+      policies:
+        {{- range .policies }}
+        - type: {{ .type }}
+          value: {{ .value }}
+          periodSeconds: {{ .periodSeconds }}
+        {{- end }}
+      selectPolicy: {{ .selectPolicy }}
+      {{- end }}
+      stabilizationWindowSeconds: {{ .stabilizationWindowSeconds }}
+    {{- end }}
+    {{- with .scaleUp }}
+    scaleUp:
+      {{- if .policies }}
+      policies:
+        {{- range .policies }}
+        - type: {{ .type }}
+          value: {{ .value }}
+          periodSeconds: {{ .periodSeconds }}
+        {{- end }}
+      selectPolicy: {{ .selectPolicy }}
+      {{- end }}
+      stabilizationWindowSeconds: {{ .stabilizationWindowSeconds }}
+    {{- end }}
   metrics:
     # Changing the order of the metrics will cause ArgoCD to go into a sync loop
     # memory needs to be first.

operations/helm/charts/grafana-agent/values.yaml

@@ -209,6 +209,28 @@ controller:
     # -- Average Memory utilization across all relevant pods, a percentage of the requested value of the resource for the pods. Setting `targetMemoryUtilizationPercentage` to 0 will disable Memory scaling.
     targetMemoryUtilizationPercentage: 80
 
+    scaleDown:
+      # -- List of policies to determine the scale-down behavior.
+      policies: []
+        # - type: Pods
+        #   value: 4
+        #   periodSeconds: 60
+      # -- Determines which of the provided scaling-down policies to apply if multiple are specified.
+      selectPolicy: Max
+      # -- The duration that the autoscaling mechanism should look back on to make decisions about scaling down.
+      stabilizationWindowSeconds: 300
+
+    scaleUp:
+      # -- List of policies to determine the scale-up behavior.
+      policies: []
+        # - type: Pods
+        #   value: 4
+        #   periodSeconds: 60
+      # -- Determines which of the provided scaling-up policies to apply if multiple are specified.
+      selectPolicy: Max
+      # -- The duration that the autoscaling mechanism should look back on to make decisions about scaling up.
+      stabilizationWindowSeconds: 0
+
   # -- Affinity configuration for pods.
   affinity: {}
 

operations/helm/tests/create-deployment-autoscaling/grafana-agent/templates/hpa.yaml

@@ -17,6 +17,24 @@ spec:
     name: grafana-agent
   minReplicas: 1
   maxReplicas: 5
+  behavior:
+    scaleDown:
+      policies:
+        - type: Pods
+          value: 4
+          periodSeconds: 60
+      selectPolicy: Min
+      stabilizationWindowSeconds: 100
+    scaleUp:
+      policies:
+        - type: Pods
+          value: 4
+          periodSeconds: 60
+        - type: Percent
+          value: 100
+          periodSeconds: 15
+      selectPolicy: Max
+      stabilizationWindowSeconds: 80
   metrics:
     # Changing the order of the metrics will cause ArgoCD to go into a sync loop
     # memory needs to be first.

operations/helm/tests/create-statefulset-autoscaling/grafana-agent/templates/hpa.yaml

@@ -17,6 +17,11 @@ spec:
     name: grafana-agent
   minReplicas: 1
   maxReplicas: 5
+  behavior:
+    scaleDown:
+      stabilizationWindowSeconds: 300
+    scaleUp:
+      stabilizationWindowSeconds: 0
   metrics:
     # Changing the order of the metrics will cause ArgoCD to go into a sync loop
     # memory needs to be first.