一、Galera Cluster
Galera是一个MySQL(也支持MariaDB，Percona)的同步多主集群软件，Galera集群的复制功能基于Galeralibrary实现,为了让MySQL与Galera library通讯，特别针对MySQL开发了wsrep API。
二、Galera Cluster的功能和优势
|
功能： multi-master，即所有节点可以同时读写数据库 自动的节点成员控制，失效节点自动被清除 新节点加入数据自动复制 并行复制，行级 |
|
优势: 因为是multi-master，所以不存在延迟；更小的客户端延迟 同时具有读和写的扩展能力 节点间数据是同步的,而Master/Slave模式是异步的,不同slave上的binlog可能是不同的 |
三、实现Galera Cluster过程；简单实现
1、准备工作
①测试环境： CentOS 7 ；
| node1 | 172.16.116.231 |
| node2 | 172.16.116.232 |
| node3 | 172.16.116.233 |
//同步各节点的时间 # ntpdate ntp.sjtu.edu.cn
//主机互信
# vim /etc/hosts
172.16.116.231 node1
172.16.116.232 node2
172.16.116.233 node3
//名称服务器
# vim /etc/resolv.conf
nameserver 172.16.0.1
②安装包选择
galera-25.3.9-1.rhel7.el7.centos.x86_64.rpmMariaDB-5.5.46-centos7-x86_64-client.rpmMariaDB-5.5.46-centos7-x86_64-common.rpmMariaDB-Galera-5.5.46-centos7-x86_64-server.rpm
2、各节点安装上述的软件包
# yum install -y *rpm//安装完成，配置node1[root@node1 ~]# vim /etc/my.cnf.d/server.cnf [galera]# Mandatory settingswsrep_provider=/usr/lib64/galera/libgalera_smm.so wsrep_cluster_address=gcomm://172.16.116.231,172.16.116.232,172.16.116.233 //定义节点wsrep_cluster_name=Mycluster //集群名称 binlog_format=row //二进制格式为rowdefault_storage_engine=InnoDB //默认存储引擎为innodbinnodb_autoinc_lock_mode=2bind-address=0.0.0.0 //绑定ip //启动服务；要加上--wsrep-new-cluster 这个参数，否则为报错error[root@node1 ~]# /etc/init.d/mysql start --wsrep-new-cluster Starting MySQL.... SUCCESS!
3、启动后查看的结果
########将/etc/my.cnf.d/server.cnf远程复制给node2，node3，启动node2，node3[root@node2 ~]# /etc/init.d/mysql startStarting MySQL......SST in progress, setting sleep higher. SUCCESS! [root@node3 ~]# /etc/init.d/mysql startStarting MySQL......SST in progress, setting sleep higher. SUCCESS! #####回到node1节点，进入mysqlMariaDB [(none)]> show status like'wsrep_%';+------------------------------+-------------------------------------------------------------+| Variable_name | Value |+------------------------------+-------------------------------------------------------------+| wsrep_local_state_uuid | 198c2664-8f88-11e5-bf9c-0e81c9575644 || wsrep_protocol_version | 7 || wsrep_last_committed | 0 || wsrep_replicated | 0 || wsrep_replicated_bytes | 0 || wsrep_repl_keys | 0 || wsrep_repl_keys_bytes | 0 || wsrep_repl_data_bytes | 0 || wsrep_repl_other_bytes | 0 || wsrep_received | 10 || wsrep_received_bytes | 758 || wsrep_local_commits | 0 || wsrep_local_cert_failures | 0 || wsrep_local_replays | 0 || wsrep_local_send_queue | 0 || wsrep_local_send_queue_max | 1 || wsrep_local_send_queue_min | 0 || wsrep_local_send_queue_avg | 0.000000 || wsrep_local_recv_queue | 0 || wsrep_local_recv_queue_max | 1 || wsrep_local_recv_queue_min | 0 || wsrep_local_recv_queue_avg | 0.000000 || wsrep_local_cached_downto | 18446744073709551615 || wsrep_flow_control_paused_ns | 0 || wsrep_flow_control_paused | 0.000000 || wsrep_flow_control_sent | 0 || wsrep_flow_control_recv | 0 || wsrep_cert_deps_distance | 0.000000 || wsrep_apply_oooe | 0.000000 || wsrep_apply_oool | 0.000000 || wsrep_apply_window | 0.000000 || wsrep_commit_oooe | 0.000000 || wsrep_commit_oool | 0.000000 || wsrep_commit_window | 0.000000 || wsrep_local_state | 4 || wsrep_local_state_comment | Synced || wsrep_cert_index_size | 0 || wsrep_causal_reads | 0 || wsrep_cert_interval | 0.000000 || wsrep_incoming_addresses | 172.16.116.231:3306,172.16.116.232:3306,172.16.116.233:3306 || wsrep_evs_delayed | || wsrep_evs_evict_list | || wsrep_evs_repl_latency | 0/0/0/0/0 || wsrep_evs_state | OPERATIONAL || wsrep_gcomm_uuid | 198b3495-8f88-11e5-8b30-6a22054488b1 || wsrep_cluster_conf_id | 3 || wsrep_cluster_size | 3 || wsrep_cluster_state_uuid | 198c2664-8f88-11e5-bf9c-0e81c9575644 || wsrep_cluster_status | Primary || wsrep_connected | ON || wsrep_local_bf_aborts | 0 || wsrep_local_index | 0 || wsrep_provider_name | Galera || wsrep_provider_vendor | Codership Oy <info@codership.com> || wsrep_provider_version | 25.3.9(r3387) || wsrep_ready | ON || wsrep_thread_count | 2 |+------------------------------+-------------------------------------------------------------+
4、测试
///node1中添加1个数据库MariaDB [(none)]> create database node1;Query OK, 1 row affected (0.01 sec)///node2上查看[root@node2 ~]# mysqlMariaDB [(none)]> show databases;+--------------------+| Database |+--------------------+| information_schema || mysql || node1 || performance_schema || test |+--------------------+5 rows in set (0.00 sec)//node3上向node1数据库中添加1个表test3，并插入数据MariaDB [(none)]> use node1;Database changedMariaDB [node1]> create table test3(id int unsigned auto_increment not null primary key,name char(10));Query OK, 0 rows affected (0.02 sec)MariaDB [node1]> insert into test3 (name) value ('zhang'),('wang');Query OK, 2 rows affected (0.01 sec)Records: 2 Duplicates: 0 Warnings: 0 //回到node1上查看，并插入新的数据MariaDB [(none)]> select * from node1.test3;+----+-------+| id | name |+----+-------+| 3 | zhang || 6 | wang |+----+-------+2 rows in set (0.00 sec)MariaDB [(none)]> insert into node1.test3 (name) value ('li'),('chen');Query OK, 2 rows affected (0.01 sec)Records: 2 Duplicates: 0 Warnings: 0//在node2上查看MariaDB [(none)]> select * from node1.test3;+----+-------+| id | name |+----+-------+| 3 | zhang || 6 | wang || 7 | li || 10 | chen |+----+-------+4 rows in set (0.00 sec)
通过上述测试可以看看3个节点都可以读写数据，即时同步；但对于自动增长的字段插入新数据，序号不连续，解决方法，只能定义全局的计数器来实现~
至此，Galera Cluster的高可用mysql集群介绍完毕！
后续工作，可以加入keepalived+haproxy实现对Galera Cluster的进一步扩展，O(∩_∩)O~~~
本文出自 “bengbengtu” 博客，请务必保留此出处http://bengbengtu.blog.51cto.com/9505633/1715213