Pod使用率用prometheus怎么监控告警

可以通过以下步骤使用 Prometheus 监控 Pod 使用率并设置告警：

步骤1：在 Kubernetes 环境中安装 Prometheus。可以通过 Helm Chart 安装 Prometheus。

步骤2：启用 Prometheus 的 kubelet 属性监控。在 Prometheus 配置文件中添加以下内容：

- job_name: kubelet
  kubernetes_sd_configs:
  - role: node
  relabel_configs:
  - source_labels: [__meta_kubernetes_node_label_kubernetes_io_hostname]
    target_label: kubernetes_io_hostname
  - action: labelmap
    regex: __meta_kubernetes_node_label_(.+)
  - action: replace
    source_labels: [__meta_kubernetes_node_name]
    target_label: instance
  - job_name: 'kubelet'
    static_configs:
    - targets: ['<kubelet_ip>:10255']
      labels:
        node_exporter_port: 10255

步骤3：创建 Prometheus 监控指标。在 Prometheus 配置文件中添加以下内容：

- job_name: 'kubernetes-pods'
  scrape_interval: 30s
  metrics_path: /metrics
  kubernetes_sd_configs:
  - role: pod
  relabel_configs:
  - source_labels: [__meta_kubernetes_pod_container_name]
    regex: '.*'
    action: keep
  - source_labels: [__meta_kubernetes_namespace]
    target_label: kubernetes_namespace
  - source_labels: [__meta_kubernetes_pod_name]
    target_label: kubernetes_pod_name
  - source_labels: [__meta_kubernetes_pod_container_name]
    target_label: container_name
  - source_labels: [__meta_kubernetes_pod_container_port_name]
    target_label: container_port_name
  - source_labels: [__meta_kubernetes_pod_container_port_number]
    target_label: container_port_number
  - source_labels: [__meta_kubernetes_pod_label_app]
    target_label: app
  relabel_configs:
  - source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_scrape]
    action: keep
    regex: true
  - source_labels: [__address__, __meta_kubernetes_pod_annotation_prometheus_io_port]
    action: replace
    target_label: __metrics_path__
    regex: (.+)
  - source_labels: [__meta_kubernetes_namespace]
    target_label: kubernetes_namespace
  - source_labels: [__meta_kubernetes_pod_name]
    target_label: kubernetes_pod_name

步骤4：配置 Prometheus 告警规则。在 Prometheus 配置文件中添加以下内容：

groups:
- name: pod-usage-rules
  rules:
  - alert: HighPodCPUUsage
    expr: sum(rate(container_cpu_usage_seconds_total{namespace="<namespace>", pod="<pod_name>", container="<container_name>"}[5m])) by (namespace, pod, container) > 1
    for: 5m
    labels:
      severity: warning
    annotations:
      summary: "High CPU usage on pod"
      description: "{{ $labels.namespace }}/{{ $labels.pod }}/{{ $labels.container }} is using too much CPU."

步骤5：重新加载 Prometheus 配置文件以应用更改。

以上就是使用 Prometheus 监控 Pod 使用率并设置告警的步骤。需要根据实际情况进行修改和调整。