k8s中使用cloudnativepg部署postgresql集群,主从故障自动切换

部署sc,我这里部署的local-path-provisioner

将sc设置为默认

kubectl patch storageclass local-path -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

部署cloudnativepg

cloudnativepg-deploy.zip

部署postgresql

在k8s集群中用来部署postgresql的节点打上污点2台或以上,如果不指定哪个节点单独部署postgresql,直接执行下一步

kubectl taint node nodename  node-role.kubernetes.io/postgres=run:NoSchedule

在k8s集群中用来部署postgresql的节点打上标签2台或以上

kubectl label node nodename  node-role.kubernetes.io/postgres=run

部署postgresql,instances: 2表示部署一主一从，根据自己的环境更改,storage、shared_buffers、max_connections 根据自己的环境更改

apiVersion: postgresql.cnpg.io/v1
kind: Cluster
metadata:
  name: cluster-postgres
spec:
  instances: 2

  postgresql:
    parameters:
      shared_buffers: "256MB"
      max_connections: "4000"
  storage:
    size: 1000Gi
  imageName: ghcr.io/cloudnative-pg/postgresql:16.13
  imagePullPolicy: IfNotPresent
  affinity:
    enablePodAntiAffinity: true
    topologyKey: kubernetes.io/hostname
    podAntiAffinityType: required
    nodeSelector:
      node-role.kubernetes.io/postgres: run
    tolerations:
    - key: node-role.kubernetes.io/postgres
      operator: Equal
      value: run
      effect: NoSchedule