【云原生】云原生监控方案（一）：基于kube-Prometheus部署Prometheus集成监控

Prometheus社区提供了在k8s部署的三种方式：

部署Prometheus-Operator，自定义部署
通过kube-Prometheus部署通用监控环境（yaml方式，本文）
通过kube-prometheus-stack部署（helm方式，简化第二种方式）

后两者都是基于第一种Prometheus-Operator的扩展版，内置了监控规则和模板

本文先使用第二种方式快速部署

开始部署

版本信息

kube-prometheus stack	Kubernetes 1.27	Kubernetes 1.28	Kubernetes 1.29	Kubernetes 1.30	Kubernetes 1.31	Kubernetes 1.32	Kubernetes 1.33
`release-0.13`	✔	✔	x	x	x	x	x
`release-0.14`	x	x	✔	✔	✔	x	x
`release-0.15`	x	x	x	x	✔	✔	✔
`main`	x	x	x	x	✔	✔	✔

拉取代码

手动下载：下载地址

1 2	git clone https://github.com/prometheus-operator/kube-prometheus.git -b release-0.14 cd kube-prometheus/

部署kube-Prometheus

1 2	kubectl apply --server-side -f manifests/setup # 如果是1.22以前的版本，使用kubectl create -f manifests/setup

部署Prometheus、alertmanager、grafana、kube-state-metrics、Prometheus-adapter、blackbox-exporter、node-exporter

这是kube-Prometheus基于Prometheus-operator为通用k8s环境提供的一整套监控方案

这整套方案都配置了PrometheusRule告警规则

并且kube-Prometheus为grafana内置了多个dashboard

1	kubectl apply -f manifests/

镜像问题

镜像查询：https://docker.aityp.com/

打tag方式

crictl pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/registry.k8s.io/kube-state-metrics/kube-state-metrics:v2.13.0
crictl pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/registry.k8s.io/prometheus-adapter/prometheus-adapter:v0.12.0

# 打tag
ctr -n k8s.io i tag swr.cn-north-4.myhuaweicloud.com/ddn-k8s/registry.k8s.io/kube-state-metrics/kube-state-metrics:v2.13.0 registry.k8s.io/kube-state-metrics/kube-state-metrics:v2.13.0
ctr -n k8s.io i tag swr.cn-north-4.myhuaweicloud.com/ddn-k8s/registry.k8s.io/prometheus-adapter/prometheus-adapter:v0.12.0 registry.k8s.io/prometheus-adapter/prometheus-adapter:v0.12.0

yaml修改镜像地址

进kube-prometheus/manifests对应的yaml文件中修改即可

外部访问

这套方案应用了networkPolicy，取消即可

1	kubectl delete -f prometheus-networkPolicy.yaml -f grafana-networkPolicy.yaml -f alertmanager-networkPolicy.yaml

修改service

1	kubectl edit svc -n monitoring prometheus-k8s

1
2
3

type: ClusterIP
# 改为
type: NodePort

持久化存储

Prometheus

修改kube-prometheus/manifests/prometheus-prometheus.yaml文件

storage:
  volumeClaimTemplate:
    spec:
      storageClassName: nfs-storage # 实际sc名称
      resources:
        requests:
          storage: 50Gi # 可根据需要调整

Alertmanager

与Prometheus一样

修改kube-prometheus/manifests/alertmanager-alertmanager.yaml文件

storage:
  volumeClaimTemplate:
    spec:
      storageClassName: nfs-storage # 实际sc名称
      resources:
        requests:
          storage: 50Gi # 可根据需要调整

grafana

修改kube-prometheus/manifests/grafana-deployment.yaml文件

      volumes:
      - emptyDir: {}
        name: grafana-storage
# 改为
      volumes:
      - name: grafana-storage
        persistentVolumeClaim:
          claimName: grafana

底部新增

---
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: grafana
  namespace: monitoring
spec:
  storageClassName: nfs-storage # 实际sc名称
  accessModes:
    - ReadWriteOnce
  resources:
    requests:
      storage: 5Gi  # 可根据需要调整

应用更改

1
2
3

kubectl apply -f prometheus-prometheus.yaml
kubectl apply -f alertmanager-alertmanager.yaml
kubectl apply -f grafana-deployment.yaml

Elijah