搬瓦工中文网
本文介绍了Centos7.3 RabbitMQ分布式集群搭建示例，分享给大家，具体如下：
注意事项
centos 7.x 关闭firewall
三台机器：
| 1 2 3 | 172.17.250.97 rabbiMQ01 172.17.250.98 rabbiMQ03 172.17.250.99 rabbiMQ02 |
配置 hosts
| 1 2 3 | 172.17.250.97 fz-rabbitMQ01 172.17.250.99 fz-rabbitMQ02 172.17.250.98 fz-rabbitMQ03 |
| 1 | $ systemctl stop firewalld.service # 停止firewall |
不想关闭防火墙，就开放15672端口,设置之后可以通过网页方式管理MQ
安装安装iptables防火墙
| 1 | yum install iptables-services |
添加配置
| 1 2 | iptables -I INPUT -p tcp --dport 5672 -j ACCEPT iptables -I INPUT -p tcp --dport 15672 -j ACCEPT |
保存配置
| 1 | service iptables save |
重启
| 1 | systemctl restart iptables.service |
设置开机自启动
| 1 | systemctl enable iptables.service |
安装
安装 Erlang
RabbitMQ 安装需要依赖 Erlang 环境
| 1 2 3 4 | $ cd /usr/local/src $ wget http: //www .rabbitmq.com /releases/erlang/erlang-19 .0.4-1.el7.centos.x86_64.rpm $ yum install erlang-19.0.4-1.el7.centos.x86_64.rpm -y |
安装 RabbitMQ
| 1 2 3 | $ cd /usr/local/src $ wget http: //www .rabbitmq.com /releases/rabbitmq-server/v3 .6.10 /rabbitmq-server-3 .6.10-1.el7.noarch.rpm $ yum install rabbitmq-server-3.6.10-1.el7.noarch.rpm -y |
仅仅rabbitMQ01运行：
| 1 | rabbitmq-server -detached |
之后运行
| 1 | rabbitmqctl status |
查看rabbitmq是否正常启动。
创建集群：
首先搞个密钥认证：
| 1 2 3 4 5 | [root@fz-rabbitMQ01 src] # ssh-keygen [root@fz-rabbitMQ01 src] # ssh-copy-id -pXXXXX -i ~/.ssh/id_rsa.pub [email protected] [root@fz-rabbitMQ01 src] # ssh-copy-id -pXXXXX -i ~/.ssh/id_rsa.pub [email protected] |
读取rabbitMQ01节点的cookie, 并复制到其他节点（节点之间通过cookie确定相互是否可通信）。 cookie存放在/var/lib/rabbitmq/.erlang.cookie或者$HOME/.erlang.cookie中。
| 1 2 3 4 5 6 | [root@fz-rabbitMQ01 src] # scp -PXXXXX /var/lib/rabbitmq/.erlang.cookie 172.17.250.99:/var/lib/rabbitmq /etc/bashrc :行94: soruce: 未找到命令 .erlang.cookie 100% 20 0.0KB /s 00:00 [root@fz-rabbitMQ01 src] # scp -PXXXXX /var/lib/rabbitmq/.erlang.cookie 172.17.250.98:/var/lib/rabbitmq /etc/bashrc :行94: soruce: 未找到命令 .erlang.cookie |
逐个启动节点
| 1 | rabbitmq-server -detached |
建立集群
分别在rabbitMQ02和rabbitMQ03上面执行即可
| 1 2 3 4 5 6 7 | rabbitmqctl stop_app rabbitmqctl reset rabbitmqctl join_cluster rabbit@fz-rabbitMQ01 rabbitmqctl start_app |
最后通过rabbitmqctl cluster_status查看集群的状态信息：
| 1 2 3 4 5 6 7 8 9 10 11 | [root@fz-rabbitMQ01 src]# rabbitmqctl cluster_status Cluster status of node 'rabbit@fz-rabbitMQ01' [{nodes,[{disc,['rabbit@fz-rabbitMQ01','rabbit@fz-rabbitMQ02', 'rabbit@fz-rabbitMQ03']}]}, {running_nodes,['rabbit@fz-rabbitMQ03','rabbit@fz-rabbitMQ02', 'rabbit@fz-rabbitMQ01']}, {cluster_name,<<"[email protected]">>}, {partitions,[]}, {alarms,[{'rabbit@fz-rabbitMQ03',[]}, {'rabbit@fz-rabbitMQ02',[]}, {'rabbit@fz-rabbitMQ01',[]}]}] |
在访问web（http://xxx.xxx.xxx.xxx:15672）时, 如果在Overview中的Nodes部分看到“Node statistics not available”的信息，说明在该节点上web管理插件还未启用。
3个节点都需要运行：
| 1 2 3 4 5 6 7 8 9 10 | [root@fz-rabbitMQ01 src]# rabbitmq-plugins enable rabbitmq_management The following plugins have been enabled: amqp_client cowlib cowboy rabbitmq_web_dispatch rabbitmq_management_agent rabbitmq_management Applying plugin configuration to rabbit@fz-rabbitMQ01... started 6 plugins. |
遇到的问题
Error: unable to connect to node ‘[email protected]’: nodedown
解决方案：执行如下两条命令：
| 1 2 | # /sbin/service rabbitmq-server stop # /sbin/service rabbitmq-server start |
然后再次执行：即可
| 1 2 3 4 | rabbitmqctl stop_app rabbitmqctl reset rabbitmqctl join_cluster rabbit@fz-rabbitMQ01 rabbitmqctl start_app |
