Skip to content

Monitoring ArgoCD

  • ArgoCD exposes Prometheus metrics from the argocd-server, argocd-application-controller, and argocd-repo-server.
  • Grafana dashboards provide visual insight into sync counts, health status, and repo performance.
  • Alerting rules can notify your team when applications go out of sync or become degraded.

What will we learn?

  • Which Prometheus metrics ArgoCD exposes
  • How to scrape ArgoCD metrics with Prometheus
  • How to import the community ArgoCD Grafana dashboard
  • Key metrics to watch and alert on

Prerequisites

01. ArgoCD Prometheus Metrics

ArgoCD exposes metrics on three endpoints:

Component Port Path
argocd-server 8083 /metrics
argocd-application-controller 8082 /metrics
argocd-repo-server 8084 /metrics 
# Port-forward the metrics endpoint
kubectl port-forward svc/argocd-server -n argocd 8083:8083 &
sleep 2
curl http://localhost:8083/metrics | grep "^argocd_" | head -20

02. Key Metrics

# Application metrics
argocd_app_info{name,namespace,project,health_status,sync_status,repo,dest_server,dest_namespace}
argocd_app_sync_total{name,namespace,phase}
argocd_app_health_status{name,namespace,health_status}
argocd_app_k8s_request_total{name,namespace,response_code,verb}

# Repository metrics
argocd_git_request_total{repo,request_type}
argocd_git_request_duration_seconds{repo,request_type}

# Cluster metrics
argocd_cluster_info{name,server}
argocd_cluster_api_resource_objects{name}
argocd_cluster_api_resources{name}

03. Deploy Prometheus and Grafana

# Add Prometheus Community Helm repo
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts || true
helm repo update || true

# Install kube-prometheus-stack (Prometheus + Grafana + Alertmanager)
helm install kube-prometheus-stack \
  prometheus-community/kube-prometheus-stack \
  -n monitoring \
  --create-namespace \
  --set grafana.adminPassword=admin123 \
  --wait --timeout 10m || true

# Verify
kubectl get pods -n monitoring

04. Configure Prometheus to Scrape ArgoCD

# Create a ServiceMonitor for ArgoCD metrics
cat <<'EOF' | kubectl apply -f -
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: argocd-metrics
  namespace: monitoring
  labels:
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      - argocd
  selector:
    matchLabels:
      app.kubernetes.io/name: argocd-metrics
  endpoints:
    - port: metrics
      interval: 30s
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: argocd-server-metrics
  namespace: monitoring
  labels:
    release: kube-prometheus-stack
spec:
  namespaceSelector:
    matchNames:
      - argocd
  selector:
    matchLabels:
      app.kubernetes.io/name: argocd-server-metrics
  endpoints:
    - port: metrics
      interval: 30s
EOF

05. Import ArgoCD Grafana Dashboard

# Port-forward Grafana
kubectl port-forward svc/kube-prometheus-stack-grafana -n monitoring 3000:80 &
sleep 3
# Open http://localhost:3000
# Default credentials: admin / admin123

# The community ArgoCD dashboard ID is: 14584
# In Grafana: Dashboards → Import → Enter ID 14584 → Load

06. Alert Rules

cat <<'EOF' | kubectl apply -f -
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: argocd-alerts
  namespace: monitoring
  labels:
    release: kube-prometheus-stack
spec:
  groups:
    - name: argocd.rules
      rules:
        - alert: ArgoCDAppOutOfSync
          expr: argocd_app_info{sync_status="OutOfSync"} == 1
          for: 5m
          labels:
            severity: warning
          annotations:
            summary: "ArgoCD application {{ $labels.name }} is OutOfSync"
            description: "Application {{ $labels.name }} has been OutOfSync for more than 5 minutes."

        - alert: ArgoCDAppDegraded
          expr: argocd_app_info{health_status="Degraded"} == 1
          for: 2m
          labels:
            severity: critical
          annotations:
            summary: "ArgoCD application {{ $labels.name }} is Degraded"
            description: "Application {{ $labels.name }} health is Degraded."

        - alert: ArgoCDSyncFailed
          expr: increase(argocd_app_sync_total{phase="Error"}[5m]) > 0
          labels:
            severity: warning
          annotations:
            summary: "ArgoCD sync failed for {{ $labels.name }}"
EOF

Practice

07. Hands-on

  1. Port-forward the ArgoCD metrics endpoint and list all argocd_app_info metrics:

??? success “Solution”

kubectl port-forward svc/argocd-metrics -n argocd 8082:8082 &
sleep 2
curl -s http://localhost:8082/metrics | grep "^argocd_app_info" | head -10
kill %1 2>/dev/null || true
  1. Install kube-prometheus-stack and verify Prometheus and Grafana are running:

??? success “Solution”

helm repo add prometheus-community https://prometheus-community.github.io/helm-charts || true
helm repo update || true
helm install kube-prometheus-stack \
  prometheus-community/kube-prometheus-stack \
  -n monitoring --create-namespace \
  --set grafana.adminPassword=admin123 || true
kubectl get pods -n monitoring
  1. Create a PrometheusRule that fires an alert when any ArgoCD application is OutOfSync for more than 5 minutes:

??? success “Solution”

cat <<'EOF' | kubectl apply -f -
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: argocd-outofsync-alert
  namespace: monitoring
  labels:
    release: kube-prometheus-stack
spec:
  groups:
    - name: argocd.rules
      rules:
        - alert: ArgoCDAppOutOfSync
          expr: argocd_app_info{sync_status="OutOfSync"} == 1
          for: 5m
          labels:
            severity: warning
          annotations:
            summary: "App {{ $labels.name }} is OutOfSync"
EOF
kubectl get prometheusrule -n monitoring

08. Summary

  • ArgoCD exposes Prometheus metrics on ports 8082 (app-controller), 8083 (server), and 8084 (repo-server)
  • argocd_app_info is the most useful metric - it contains health status, sync status, and labels for every app
  • ServiceMonitors configure Prometheus to scrape ArgoCD endpoints automatically
  • The community Grafana dashboard ID 14584 provides a comprehensive ArgoCD overview
  • Alert on OutOfSync (>5min) and Degraded health to catch issues before users notice them