Observability Dashboards for Workbench Components¶

This document explains how to enable Grafana dashboards for various components in the Workbench, allowing you to monitor key metrics. The components covered include:

• Jenkins
• ArgoCD
• Argo Rollouts
• KubeVela

By default, metric collection is not enabled for these components and needs to be turned on manually. The methods for each component are described below.

Jenkins¶

You can enable metrics by updating the Helm chart parameters. Follow these steps:

1. Go to Container Management , click Clusters, and find the cluster where Jenkins is deployed.
2. Click Helm Apps , locate Jenkins in its namespace, and click Update

3. Choose to update via YAML, and modify the following fields:

   jenkins:
     Master:
       metrics:
         enabled: true
         serviceMonitor:
           enabled: true
           additionalLabels:
             operator.insight.io/managed-by: insight
   

4. Click Save , and wait for about 3 minutes.

5. Go to Observability -> Dashboards, and search for the keyword jenkins to open the dashboard.

ArgoCD¶

ArgoCD metrics can also be enabled by updating Helm parameters. The update steps are the same as described in Jenkins setup. Search for argocd in the dashboards. Modify the following YAML:

argo-cd:
  controller:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  dex:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  redis:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  server:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  repoServer:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  notifications:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight
  applicationSet:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight

Argo Rollouts¶

Argo Rollouts metrics can be enabled similarly by updating Helm parameters, following the same steps as in Jenkins setup. Search with the keyword argo-rollouts. Modify the following YAML:

argo-rollouts:
  controller:
    metrics:
      enabled: true
      serviceMonitor:
        enabled: true
        additionalLabels:
          operator.insight.io/managed-by: insight

KubeVela¶

KubeVela does not support enabling metrics via Helm parameters. You will need to manually deploy the following resources:

1. Deploy the Service:

   apiVersion: v1
   kind: Service
   metadata:
     name: kubevela-controller-service
     namespace: < your kubevela namespace > # Replace with your namespace
     labels:
       component: kubevela-controller
   spec:
     ports:
       - name: http
         protocol: TCP
         port: 9443
         targetPort: 9443
       - name: metrics
         protocol: TCP
         port: 8080
         targetPort: 8080
     selector: < kubevela core pod label selector> # Replace with the actual selector
   

2. Deploy the ServiceMonitor:

   apiVersion: monitoring.coreos.com/v1
   kind: ServiceMonitor
   metadata:
     name: amamba-kubevela
     namespace: < your kubevela namespace > # Replace with your namespace
     labels:
       operator.insight.io/managed-by: insight
   spec:
     endpoints:
       - honorLabels: true
         interval: 10s
         path: /metrics
         port: metrics
         scheme: http
     namespaceSelector:
       matchNames:
         - < your kubevela namespace > # Replace with your namespace
     selector:
       matchLabels:
         component: kubevela-controller
   

Comments