Openstack云平台搭建课程八·控制节点集群

其他优化补充说明

哪些组件一般不通过haproxy做负载？

Memcached一般使用Openstack自带的集群管理方案，而不是加入haproxy。

memcached_servers = controller01:11211,controller2:11211,controller3:11211

RabbitMQ用haproxy负载还是用openstack自带的集群管理方案？

Openstack官方建议使用其自带的集群管理方案，但本人一直通过haproxy进行负载，暂未发现问题。

transport_url=rabbit://openstack:Openstack123@10.10.100.98:5672
#transport_url=rabbit://openstack:Openstack123@controller1:5672,controller2:5672,controller3:5672

适用于openstack集群的haproxy的全部配置如下，具体参数可以根据实际情况进行调整：

global
  log 127.0.0.1 local2
  chroot /var/lib/haproxy
  pidfile /var/run/haproxy.pid
  maxconn 51200
  user   haproxy
  group haproxy
  daemon
  spread-checks 3
  tune.bufsize 32768
  tune.maxrewrite 1024
  tune.ssl.default-dh-param 2048
  stats socket /var/lib/haproxy/stats

defaults
  mode http
  log global
  option httplog
  option dontlognull
  option http-server-close
  option forwardfor except 127.0.0.0/8
  option redispatch
  retries 3
  timeout http-request 10s
  timeout queue 1m
  timeout connect 10s
  timeout client 1m
  timeout server 1m
  timeout http-keep-alive 10s
  timeout check 10s
  maxconn 3000

listen cinder_api
  bind 10.10.100.93:8776
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:8776 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:8776 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:8776 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen nova_api
  bind 10.10.100.94:8773
  option tcpka
  option tcplog
  balance source
  server controller1 10.10.100.150:8773 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:8773 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:8773 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen nova_compute_api
  bind 10.10.100.94:8774
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:8774 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:8774 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:8774 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen nova_placement
  bind 10.10.100.94:8778
  option tcpka
  option tcplog
  balance source
  server controller1 10.10.100.150:8778 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:8778 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:8778 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen nova_vncproxy
  bind 10.10.100.94:6080
  option tcpka
  option tcplog
  balance source
  server controller1 10.10.100.150:6080 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:6080 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:6080 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen neutron_metadata_api
  bind 10.10.100.95:8775
  option tcpka
  option tcplog
  balance source
  server controller1 10.10.100.150:8775 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:8775 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:8775 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen neutron_api
  bind 10.10.100.95:9696
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:9696 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:9696 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:9696 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen glance_api
  bind 10.10.100.96:9292
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:9292 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:9292 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:9292 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen glance_registry
  bind 10.10.100.96:9191
  option tcpka
  option tcplog
  balance source
  server controller1 10.10.100.150:9191 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:9191 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:9191 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen keystone_admin
  bind 10.10.100.97:35357
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:35357 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:35357 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:35357 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen keystone_public
  bind 10.10.100.97:5000
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:5000 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller2 10.10.100.149:5000 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  server controller3 10.10.100.148:5000 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen dashboard
  bind 10.10.100.97:80
  option tcpka
  option httpchk
  option tcplog
  balance source
  server controller1 10.10.100.150:80 check inter 5s rise 2 fall 3
  server controller2 10.10.100.149:80 check inter 5s rise 2 fall 3
  server controller3 10.10.100.148:80 check inter 5s rise 2 fall 3

listen rabbitmq
  bind 10.10.100.98:5672
  timeout client 12h
  timeout server 12h
  mode tcp
  option tcpka
  option clitcpka
  balance roundrobin
  server controller1 10.10.100.150:5672 weight 9
  server controller2 10.10.100.149:5672 weight 9 backup
  server controller3 10.10.100.148:5672 weight 5 backup
  #Openstack通过VIP连接RabbitMQ时，如果采用如下方案，Nova-api会报错
  #server controller2 10.10.100.150:5672 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  #server controller2 10.10.100.149:5672 check inter 5s fastinter 1s downinter 2s rise 3 fall 3
  #server controller3 10.10.100.148:5672 check inter 5s fastinter 1s downinter 2s rise 3 fall 3

listen mariadb
  bind 10.10.100.99:3306
  timeout client 24h
  timeout server 24h
  #如果未开启集群监控脚本，请使用如下配置
  #balance source
  #mode tcp
  #server controller1 10.10.100.150:3306 check inter 1000 rise 2 fall 3
  #server controller2 10.10.100.149:3306 check inter 1000 rise 2 fall 3
  #server controller3 10.10.100.148:3306 check inter 1000 rise 2 fall 3
  #如果开启了集群监控脚本，请使用如下配置
  balance leastconn
  option httpchk
  mode tcp
  server node1 10.10.100.150:3306 check port 9200 inter 3s fastinter 1s rise 3 fall 3
  server node2 10.10.100.149:3306 check port 9200 inter 3s fastinter 1s rise 3 fall 3
  server node3 10.10.100.148:3306 check port 9200 inter 3s fastinter 1s rise 3 fall 3 backup

listen stats
  bind 0.0.0.0:8080
  mode http
  stats enable
  stats refresh 30s
  stats uri /stats
  stats realm openstack\ Haproxy
  stats auth admin:admin
  stats hide-version
  stats show-node
  stats show-legends

原创文章禁止转载：技术学堂 » Openstack云平台搭建课程八·控制节点集群