Grafana Cloud Infrastructure Monitoring

Grafana Cloud基础设施监控

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

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

Kubernetes Monitoring (k8s-monitoring Helm Chart)

Kubernetes监控（k8s-monitoring Helm Chart）

bash

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

yaml

undefined

bash

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

yaml

undefined

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


```bash
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

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


```bash
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

关键Kubernetes指标

promql

undefined

promql

undefined

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

undefined

kubelet_volume_stats_used_bytes / kubelet_volume_stats_capacity_bytes

undefined

AWS CloudWatch Integration

AWS CloudWatch集成

yaml

undefined

yaml

undefined

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:
```yaml

prometheus.scrape "cloudwatch" { targets = [{address = "cloudwatch-exporter:9106"}] forward_to = [prometheus.remote_write.cloud.receiver] }


或者直接使用CloudWatch数据源：
```yaml

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

undefined

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

undefined

Azure Monitor Integration

Azure Monitor集成

yaml

undefined

yaml

undefined

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

undefined

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

undefined

GCP / Google Cloud Monitoring

GCP / Google Cloud监控

yaml

undefined

yaml

undefined

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", ... }

undefined

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", ... }

undefined

Node Exporter / Linux Host Monitoring

Node Exporter / Linux主机监控

alloy

// 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"
}

alloy

// 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

Docker / 容器监控

alloy

// 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"
}

alloy

// 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)

常见基础设施仪表盘（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)

集成目录中提供的预构建仪表盘：

Kubernetes / 集群（ID: 15520）
Kubernetes / 命名空间（ID: 15521）
Kubernetes / Pod（ID: 15522）
Node Exporter Full（ID: 1860）
cAdvisor（ID: 14282）
AWS EC2（通过CloudWatch集成）
Azure VMs（通过Azure Monitor集成）

Alerting for Infrastructure

基础设施告警

yaml

undefined

yaml

undefined

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"

undefined

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"

undefined

infrastructure

Original

Translation

Grafana Cloud Infrastructure Monitoring

Grafana Cloud基础设施监控

Kubernetes Monitoring (k8s-monitoring Helm Chart)

Kubernetes监控（k8s-monitoring Helm Chart）

values.yaml

values.yaml

Key Kubernetes Metrics

关键Kubernetes指标

CPU usage by pod

CPU usage by pod

Memory usage by pod

Memory usage by pod

Node CPU pressure

Node CPU pressure

Pod restarts

Pod restarts

Deployment readiness

Deployment readiness

PVC usage

PVC usage

AWS CloudWatch Integration

AWS CloudWatch集成

Alloy config for AWS CloudWatch scraping

Alloy config for AWS CloudWatch scraping

provisioning/datasources/cloudwatch.yaml

provisioning/datasources/cloudwatch.yaml

Or explicit credentials:

authType: credentials

Or explicit credentials:

authType: credentials

Azure Monitor Integration

Azure Monitor集成

provisioning/datasources/azure.yaml

provisioning/datasources/azure.yaml

GCP / Google Cloud Monitoring

GCP / Google Cloud监控

provisioning/datasources/google.yaml

provisioning/datasources/google.yaml

Or JWT:

authenticationType: jwt

Or JWT:

authenticationType: jwt

Node Exporter / Linux Host Monitoring

Node Exporter / Linux主机监控

Docker / Container Monitoring

Docker / 容器监控

Common Infrastructure Dashboards (Grafana Cloud)

常见基础设施仪表盘（Grafana Cloud）

Alerting for Infrastructure

基础设施告警

Common infrastructure alert rules

Common infrastructure alert rules