infrastructure

Grafana Cloud Infrastructure Monitoring

Docs: https://grafana.com/docs/grafana-cloud/monitor-infrastructure/

Kubernetes Monitoring (k8s-monitoring Helm Chart)

helm repo add grafana https://grafana.github.io/helm-charts
helm repo update

# values.yaml
cluster:
  name: production-us-east

externalServices:
  prometheus:
    host: https://prometheus-prod-xx.grafana.net
    basicAuth:
      username: "123456"
      password:
        secretName: grafana-cloud-secret
        secretKey: api-key
  loki:
    host: https://logs-prod-xx.grafana.net
    basicAuth:
      username: "234567"
      password:
        secretName: grafana-cloud-secret
        secretKey: api-key
  tempo:
    host: https://tempo-prod-xx.grafana.net:443
    basicAuth:
      username: "345678"
      password:
        secretName: grafana-cloud-secret
        secretKey: api-key

metrics:
  enabled: true
  cost:
    enabled: true    # Kubernetes cost monitoring
  podMonitors:
    enabled: true
  serviceMonitors:
    enabled: true
  kube-state-metrics:
    enabled: true
  node-exporter:
    enabled: true
  cadvisor:
    enabled: true

logs:
  pod_logs:
    enabled: true
  cluster_events:
    enabled: true

traces:
  enabled: true

profiles:
  enabled: false

receivers:
  grpc:
    enabled: true
    port: 4317
  http:
    enabled: true
    port: 4318

kubectl create secret generic grafana-cloud-secret \
  --from-literal=api-key=<your-api-key> \
  -n monitoring

helm install k8s-monitoring grafana/k8s-monitoring \
  -n monitoring --create-namespace \
  -f values.yaml

Key Kubernetes Metrics

# CPU usage by pod
sum(rate(container_cpu_usage_seconds_total{
  namespace="$namespace", container!=""}[5m])) by (pod)

# Memory usage by pod
sum(container_memory_working_set_bytes{
  namespace="$namespace", container!=""}) by (pod)

# Node CPU pressure
1 - avg(rate(node_cpu_seconds_total{mode="idle"}[5m])) by (instance)

# Pod restarts
increase(kube_pod_container_status_restarts_total[1h])

# Deployment readiness
kube_deployment_status_replicas_ready / kube_deployment_spec_replicas

# PVC usage
kubelet_volume_stats_used_bytes / kubelet_volume_stats_capacity_bytes

AWS CloudWatch Integration

# Alloy config for AWS CloudWatch scraping
prometheus.scrape "cloudwatch" {
  targets = [{__address__ = "cloudwatch-exporter:9106"}]
  forward_to = [prometheus.remote_write.cloud.receiver]
}

Or use the CloudWatch datasource directly:

# provisioning/datasources/cloudwatch.yaml
apiVersion: 1
datasources:
  - name: CloudWatch
    type: cloudwatch
    jsonData:
      defaultRegion: us-east-1
      authType: default    # uses EC2 instance role / ECS task role
      # Or explicit credentials:
      # authType: credentials
    secureJsonData:
      accessKey: AKIAIOSFODNN7EXAMPLE
      secretKey: wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY

Azure Monitor Integration

# provisioning/datasources/azure.yaml
apiVersion: 1
datasources:
  - name: Azure Monitor
    type: grafana-azure-monitor-datasource
    jsonData:
      cloudName: AzureCloud
      tenantId: your-tenant-id
      clientId: your-client-id
    secureJsonData:
      clientSecret: your-client-secret

GCP / Google Cloud Monitoring

# provisioning/datasources/google.yaml
apiVersion: 1
datasources:
  - name: Google Cloud Monitoring
    type: stackdriver
    jsonData:
      authenticationType: gce    # uses GCE metadata server
      # Or JWT:
      # authenticationType: jwt
    secureJsonData:
      privateKey: |
        { "type": "service_account", ... }

Node Exporter / Linux Host Monitoring

// Alloy config for Linux host metrics
prometheus.exporter.unix "host" {
  rootfs_path = "/"
  enable_collectors = ["cpu", "diskstats", "filesystem", "loadavg", "meminfo", "netdev", "stat", "time", "uname"]
}

prometheus.scrape "node" {
  targets    = prometheus.exporter.unix.host.targets
  forward_to = [prometheus.remote_write.cloud.receiver]
  scrape_interval = "60s"
}

Docker / Container Monitoring

// cAdvisor metrics via Alloy
prometheus.scrape "cadvisor" {
  targets = [{"__address__" = "localhost:8080"}]
  metrics_path = "/metrics"
  forward_to   = [prometheus.remote_write.cloud.receiver]
}

// Docker container logs
loki.source.docker "containers" {
  host       = "unix:///var/run/docker.sock"
  targets    = discovery.docker.containers.targets
  forward_to = [loki.write.cloud.receiver]
}

discovery.docker "containers" {
  host = "unix:///var/run/docker.sock"
}

Common Infrastructure Dashboards (Grafana Cloud)

Pre-built dashboards available from the integrations catalog:

• Kubernetes / Cluster (ID: 15520)
• Kubernetes / Namespace (ID: 15521)
• Kubernetes / Pod (ID: 15522)
• Node Exporter Full (ID: 1860)
• cAdvisor (ID: 14282)
• AWS EC2 (via CloudWatch integration)
• Azure VMs (via Azure Monitor integration)

Alerting for Infrastructure

# Common infrastructure alert rules
groups:
  - name: kubernetes-alerts
    rules:
      - alert: PodCrashLooping
        expr: rate(kube_pod_container_status_restarts_total[15m]) * 60 * 15 > 0
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "Pod {{ $labels.namespace }}/{{ $labels.pod }} crash looping"

      - alert: NodeMemoryPressure
        expr: (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes) < 0.1
        for: 5m
        labels:
          severity: critical
        annotations:
          summary: "Node {{ $labels.instance }} low memory (<10% free)"

      - alert: PersistentVolumeAlmostFull
        expr: kubelet_volume_stats_available_bytes / kubelet_volume_stats_capacity_bytes < 0.1
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "PVC {{ $labels.namespace }}/{{ $labels.persistentvolumeclaim }} almost full"