kubesphere+k8s+helm安装RabbitMQ集群+镜像模式-Tianch编程笔记

1.安装k8s多节点集群

1.下载Rabbitmq的chart包

添加bitnamid仓库

helm repo add bitnami https://charts.bitnami.com/bitnami

查询chart
```
helm search repo bitnami
```
创建rabbitmq的工作目录
```
mkdir -p /opt/rabbitmq
```
使用helm拉取rabbitmq
```
helm pull bitnami/rabbitmq
```

解压

tar -xvf 拉取的文件名-版本变更会导致文件名变化

2.编辑配置文件

设置管理员密码

配置文件方式

auth:
  ## @param auth.username RabbitMQ application username
  ## ref: https://github.com/bitnami/bitnami-docker-rabbitmq#environment-variables
  ##
  username: admin

  ## @param auth.password RabbitMQ application password
  ## ref: https://github.com/bitnami/bitnami-docker-rabbitmq#environment-variables
  ##
  password: "77589910"
  ## @param auth.existingPasswordSecret Existing secret with RabbitMQ credentials (must contain a value for `rabbitmq-password` key)
  ## e.g:
  ## existingPasswordSecret: name-of-existing-secret
  ##
  existingPasswordSecret: ""

  ## @param auth.erlangCookie Erlang cookie to determine whether different nodes are allowed to communicate with each other
  ## ref: https://github.com/bitnami/bitnami-docker-rabbitmq#environment-variables
  ##
  erlangCookie: "secretcookie123"
  ## @param auth.existingErlangSecret Existing secret with RabbitMQ Erlang cookie (must contain a value for `rabbitmq-erlang-cookie` key)
  ## e.g:
  ## existingErlangSecret: name-of-existing-secret
  ##
  existingErlangSecret: ""

![image-20220123203354564](# ubesphere+k8s+helm安装RabbitMQ集群+镜像模式/img/image-20220123203354564.png)

安装时set方式指定

--set auth.username=admin,auth.password=77589910,auth.erlangCookie=secretcookie123

设置启动副本数
```
replicaCount: 3
```

设置持久化存储

persistence:
  ## @param persistence.enabled Enable RabbitMQ data persistence using PVC
  ##
  enabled: true

  ## @param persistence.storageClass PVC Storage Class for RabbitMQ data volume
  ## If defined, storageClassName: <storageClass>
  ## If set to "-", storageClassName: "", which disables dynamic provisioning
  ## If undefined (the default) or set to null, no storageClassName spec is
  ##   set, choosing the default provisioner.  (gp2 on AWS, standard on
  ##   GKE, AWS & OpenStack)
  ##
  storageClass: "local"
  ## @param persistence.selector Selector to match an existing Persistent Volume
  ## selector:
  ##   matchLabels:
  ##     app: my-app
  ##
  selector: {}
  ## @param persistence.accessMode PVC Access Mode for RabbitMQ data volume
  ##
  accessMode: ReadWriteOnce

  ## @param persistence.existingClaim Provide an existing PersistentVolumeClaims
  ## The value is evaluated as a template
  ## So, for example, the name can depend on .Release or .Chart
  ##
  existingClaim: ""

  ## @param persistence.size PVC Storage Request for RabbitMQ data volume
  ## If you change this value, you might have to adjust `rabbitmq.diskFreeLimit` as well
  ##
  size: 60Gi

  ## @param persistence.volumes Additional volumes without creating PVC
  ##  - name: volume_name
  ##    emptyDir: {}
  ##
  volumes: []
  ## @param persistence.annotations Persistence annotations. Evaluated as a template
  ## Example:
  ## annotations:
  ##   example.io/disk-volume-type: SSD
  ##
  annotations: {}

设置集群访问方式
```
service:
  ## @param service.type Kubernetes Service type
  ##
  type: NodePort
```
在实操中我发现,如果service.type设置成ClusterIP,等待集群启动完成后,再手动添加NodePort的访问方式,手动删除一个容器后,容器在自动恢复时会报错,然后就是不停的重启

设置集群意外宕机强制重启

当rabbitmq启用持久化存储时，若rabbitmq所有pod同时宕机，将无法重新启动，因此有必要提前开启clustering.forceBoot

clustering:
  ## @param clustering.enabled Enable RabbitMQ clustering
  ##
  enabled: true
  ## @param clustering.addressType Switch clustering mode. Either `ip` or `hostname`
  ##
  addressType: hostname
  ## @param clustering.rebalance Rebalance master for queues in cluster when new replica is created
  ## ref: https://www.rabbitmq.com/rabbitmq-queues.8.html#rebalance
  ##
  rebalance: false

  ## @param clustering.forceBoot Force boot of an unexpectedly shut down cluster (in an unexpected order).
  ## forceBoot executes 'rabbitmqctl force_boot' to force boot cluster shut down unexpectedly in an unknown order
  ## ref: https://www.rabbitmq.com/rabbitmqctl.8.html#force_boot
  ##
  forceBoot: true

  ## @param clustering.partitionHandling Switch Partition Handling Strategy. Either `autoheal` or `pause-minority` or `pause-if-all-down` or `ignore`
  ## ref: https://www.rabbitmq.com/partitions.html#automatic-handling
  ##
  partitionHandling: autoheal

指定时区

extraEnvVars:
  - name: TZ
    value: "Asia/Shanghai"

3.部署RabbitMQ

在kubesphere后台创建命名空间demo-project

使用kubectl命令创建的在kubesphere后台可能会不显示
安装
- 使用配置文件中配置的管理员账号密码
```
helm install rabbitmq -n demo-project .
```
  helm install后的rabbitmq只是一个名称,起什么名称都可以
  
  命令最后有一个点
- 通过set指定管理员账号密码–后期upgrade时也须指定上参数
```
helm install rabbitmq -n demo-project . \
--set auth.username=admin,auth.password=77589910,auth.erlangCookie=secretcookie123
```

查看安装状态

查看安装进度
```
kubectl get pod -n demo-project -w
```
查看svc
```
kubectl get svc -n demo-project
```

查看集群状态

进入pod

kubectl exec -it -n demo-project rabbitmq-0 -- bash

查看集群状态
```
rabbitmqctl cluster_status
```
查看当前策略
```
rabbitmqctl list_policies
```

设置集群名称

rabbitmqctl set_cluster_name [cluster_name]

4.配置镜像模式实现高可用

使用命令

rabbitmqctl set_policy ha-all "^" '{"ha-mode":"all" , "ha-sync-mode":"automatic"}'

使用控制台

5.清理RabbitMQ集群

卸载

helm uninstall rabbitmq -n demo-project

删除pvc

kubectl delete pvc -n demo-project data-rabbitmq-0 data-rabbitmq-1 data-rabbitmq-2

6.参考

https://blog.csdn.net/qq_14999375/article/details/119085363