如何在CentOS 7系统上配置高可用的MariaDB Galera Cluster多主群集

MariaDB Galera群集是多主同步群集。它只能在Linux系统上使用，并且只支持XtraDB/InnoDB存储引擎。
我们在本篇文章将配置一个HA（高可用性）多主数据库群集。

VMware® Workstation 12 Pro

CentOS 7最小化的系统镜像
CentOS-7-x86_64-Minimal-1503-01.iso


hostname ip address node name node address cluster name
contoso21.com 192.168.10.21 db1 192.168.10.21 galera1_cluster
contoso22.com 192.168.10.22 db2 192.168.10.22 galera1_cluster

contoso23.com 192.168.10.23 db3 192.168.10.23 galera1_cluster

第1步: 建立主机IP地址与主机名称的映射关系[root@contoso21 ~]# cat > /etc/hosts127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 contoso21.com::1 localhost localhost.localdomain localhost6 localhost6.localdomain6192.168.10.21 contoso21.com[root@contoso22 ~]# cat > /etc/hosts127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 contoso22.com::1 localhost localhost.localdomain localhost6 localhost6.localdomain6192.168.10.22 contoso22.com[root@contoso23 ~]# cat > /etc/hosts127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 contoso23.com::1 localhost localhost.localdomain localhost6 localhost6.localdomain6192.168.10.23 contoso23.com第2步: 添加MariaDB 软件包仓库在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：tee /etc/yum.repos.d/MariaDB.repo <<-'EOF'[mariadb]name = MariaDBbaseurl = http://yum.mariadb.org/10.1/centos7-amd64gpgkey=https://yum.mariadb.org/RPM-GPG-KEY-MariaDBgpgcheck=1EOF第3步: – 设置selinux的disabled模式 setenforce 0，并且关闭防火墙服务在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：cat > /etc/selinux/config# This file controls the state of SELinux on the system.# SELINUX= can take one of these three values:# enforcing - SELinux security policy is enforced.# permissive - SELinux prints warnings instead of enforcing.# disabled - No SELinux policy is loaded.SELINUX=disabled# SELINUXTYPE= can take one of three two values:# targeted - Targeted processes are protected,# minimum - Modification of targeted policy. Only selected processes are protected. # mls - Multi Level Security protection.SELINUXTYPE=targeted在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：systemctl stop firewalld && systemctl disable firewalldinit 6 ## reboot第4步: – 安装MariaDB Galera群集10.1版软件包在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：yum -y install socat deltarpm && yum install -y MariaDB-server MariaDB-client在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：mkdir -p /var/log/mariadbtouch /var/log/mariadb/mariadb-slow.logtouch /var/log/mariadb/queries.logchmod 666 /var/log/mariadb/mariadb-slow.logchmod 666 /var/log/mariadb/queries.log第5步: 配置MariaDB数据库的安全性 systemctl status mysql -l 命令可以查看服务的完整运行状态在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：/usr/bin/mysql_secure_installationNOTE: RUNNING ALL PARTS OF THIS SCRIPT IS RECOMMENDED FOR ALL MariaDB SERVERS IN PRODUCTION USE! PLEASE READ EACH STEP CAREFULLY!In order to log into MariaDB to secure it,we'll need the currentpassword for the root user. If you've just installed MariaDB,andyou haven't set the root password yet,the password will be blank,so you should just press enter here.Enter current password for root (enter for none): 刚刚安装的MariaDB没有密码,此处直接回车便可成功连接数据库.OK,successfully used password,moving on...Setting the root password ensures that nobody can log into the MariaDBroot user without the proper authorisation.Set root password? [Y/n] yNew password: 123456Re-enter new password: 123456Password updated successfully!Reloading privilege tables..... Success!By default,a MariaDB installation has an anonymous user,allowing anyoneto log into MariaDB without having to have a user account created forthem. This is intended only for testing,and to make the installationgo a bit smoother. You should remove them before moving into aproduction environment.Remove anonymous users? [Y/n] y... Success!Normally,root should only be allowed to connect from 'localhost'. Thisensures that someone cannot guess at the root password from the network.Disallow root login remotely? [Y/n] n... Success!By default,MariaDB comes with a database named 'test' that anyone canaccess. This is also intended only for testing,and should be removedbefore moving into a production environment.Remove test database and access to it? [Y/n] y- Dropping test database...... Success!- Removing privileges on test database...... Success!Reloading the privilege tables will ensure that all changes made so farwill take effect immediately.Reload privilege tables now? [Y/n] y... Success!Cleaning up...All done! If you've completed all of the above steps,your MariaDBinstallation should now be secure.Thanks for using MariaDB!第6步: 创建MariaDB Galera群集需要用到的多个用户Now,we have to createsome users that must be able to access the database. The‘sst_user’is the user which a database node will use for authenticating to another database node in the State Transfer Snapshot (SST) phase. Runthe following command on all nodes:在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令:mysql -uroot -p123456 -h127.0.0.1在db1 MariaDB [(none)]> 和 db2 MariaDB [(none)]> 还有 db2 MariaDB [(none)]>上分别执行如下SQL命令:USE mysql;SHOW DATABASES;SELECT host,user FROM user;DELETE FROM mysql.user WHERE user='';GRANT ALL ON *.* TO 'root'@'%' IDENTIFIED BY '123456';GRANT USAGE ON *.* to sst_user@'%' IDENTIFIED BY '123456';GRANT ALL PRIVILEGES on *.* to sst_user@'%';FLUSH PRIVILEGES;SELECT host,user FROM user;SHOW VARIABLES LIKE 'character%';SHOW VARIABLES LIKE '%collation%';QUIT建议你使用能够访问数据库的主机名称或者IP地址替代‘%’,因为‘%’ 的意思是允许用户root或者sst_user从任何主机上访问数据库，因此安全性较差.第7步: 修改MariaDB Galera群集所有节点的配置文件,设置MariaDB数据库的默认编码格式和表名称的小写格式，配置查询日志和慢查询日志,关闭所有节点的mysql服务在[root@contoso21 ~]# 和 [root@contoso22 ~]# 还有 [root@contoso23 ~]#上分别执行如下终端命令：service mysql stop[root@contoso21 ~]# cat > /etc/my.cnf.d/server.cnf## These groups are read by MariaDB server.# Use it for options that only the server (but not clients) should see## See the examples of server my.cnf files in /usr/share/mysql/## this is read by the standalone daemon and embedded servers[server]# this is only for the mysqld standalone daemon[mysqld]character-set-server=utf8lower_case_table_names=1general-log=ONgeneral-log-file=/var/log/mariadb/queries.loglog-output=fileslow-query-log=ONslow-query-log-file=/var/log/mariadb/mariadb-slow.loglong_query_time=1## * Galera-related settings#[galera]# Mandatory settingswsrep_on=ONwsrep_provider=/usr/lib64/galera/libgalera_smm.sowsrep_cluster_address='gcomm://'wsrep_cluster_name='galera1_cluster'wsrep_node_address='192.168.10.21'wsrep_node_name='db1'wsrep_sst_method=rsyncbinlog_format=rowdefault_storage_engine=InnoDBinnodb_autoinc_lock_mode=2bind-address=0.0.0.0#wsrep_on=ON#wsrep_provider=#wsrep_cluster_address=#binlog_format=row#default_storage_engine=InnoDB#innodb_autoinc_lock_mode=2## Allow server to accept connections on all interfaces.##bind-address=0.0.0.0## Optional setting#wsrep_slave_threads=1#innodb_flush_log_at_trx_commit=0# this is only for embedded server[embedded]# This group is only read by MariaDB servers,not by MySQL.# If you use the same .cnf file for MySQL and MariaDB,# you can put MariaDB-only options here[mariadb]# This group is only read by MariaDB-10.1 servers.# If you use the same .cnf file for MariaDB of different versions,# use this group for options that older servers don't understand[mariadb-10.1][root@contoso22 ~]# cat > /etc/my.cnf.d/server.cnf## These groups are read by MariaDB server.# Use it for options that only the server (but not clients) should see## See the examples of server my.cnf files in /usr/share/mysql/## this is read by the standalone daemon and embedded servers[server]# this is only for the mysqld standalone daemon[mysqld]character-set-server=utf8lower_case_table_names=1general-log=ONgeneral-log-file=/var/log/mariadb/queries.loglog-output=fileslow-query-log=ONslow-query-log-file=/var/log/mariadb/mariadb-slow.loglong_query_time=1## * Galera-related settings#[galera]# Mandatory settingswsrep_on=ONwsrep_provider=/usr/lib64/galera/libgalera_smm.sowsrep_cluster_address='gcomm://192.168.10.21'wsrep_cluster_name='galera1_cluster'wsrep_node_address='192.168.10.22'wsrep_node_name='db2'wsrep_sst_method=rsyncbinlog_format=rowdefault_storage_engine=InnoDBinnodb_autoinc_lock_mode=2bind-address=0.0.0.0#wsrep_on=ON#wsrep_provider=#wsrep_cluster_address=#binlog_format=row#default_storage_engine=InnoDB#innodb_autoinc_lock_mode=2## Allow server to accept connections on all interfaces.##bind-address=0.0.0.0## Optional setting#wsrep_slave_threads=1#innodb_flush_log_at_trx_commit=0# this is only for embedded server[embedded]# This group is only read by MariaDB servers,# use this group for options that older servers don't understand[mariadb-10.1][root@contoso23 ~]# cat > /etc/my.cnf.d/server.cnf## These groups are read by MariaDB server.# Use it for options that only the server (but not clients) should see## See the examples of server my.cnf files in /usr/share/mysql/## this is read by the standalone daemon and embedded servers[server]# this is only for the mysqld standalone daemon[mysqld]character-set-server=utf8lower_case_table_names=1general-log=ONgeneral-log-file=/var/log/mariadb/queries.loglog-output=fileslow-query-log=ONslow-query-log-file=/var/log/mariadb/mariadb-slow.loglong_query_time=1## * Galera-related settings#[galera]# Mandatory settingswsrep_on=ONwsrep_provider=/usr/lib64/galera/libgalera_smm.sowsrep_cluster_address='gcomm://192.168.10.21'wsrep_cluster_name='galera1_cluster'wsrep_node_address='192.168.10.23'wsrep_node_name='db3'wsrep_sst_method=rsyncbinlog_format=rowdefault_storage_engine=InnoDBinnodb_autoinc_lock_mode=2bind-address=0.0.0.0#wsrep_on=ON#wsrep_provider=#wsrep_cluster_address=#binlog_format=row#default_storage_engine=InnoDB#innodb_autoinc_lock_mode=2## Allow server to accept connections on all interfaces.##bind-address=0.0.0.0## Optional setting#wsrep_slave_threads=1#innodb_flush_log_at_trx_commit=0# this is only for embedded server[embedded]# This group is only read by MariaDB servers,# use this group for options that older servers don't understand[mariadb-10.1]第8步: 初始化MariaDB Galera群集中第一个节点使用专门的参数项‘ --wsrep-new-cluster’来启动MariaDB Gelera群集,只在db1节点（第1个节点）上如此操作,以便群集主要的节点被初始化.[root@contoso21 ~]# /etc/init.d/mysql start --wsrep-new-cluster[root@contoso21 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"wsrep_local_state_comment | Synced <-- cluster is syncedwsrep_incoming_addresses | 192.168.10.21:3306 <-- node db1 is a providerwsrep_cluster_size | 1 <-- cluster consists of 1 node wsrep_ready | ON <-- good :)[root@contoso21 ~]# /etc/init.d/mysql start --wsrep-new-clusterStarting mysql (via systemctl): [ OK ][root@contoso21 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"+------------------------------+---------------------------------------------+| Variable_name | Value |+------------------------------+---------------------------------------------+| wsrep_apply_oooe | 0.000000 || wsrep_apply_oool | 0.000000 || wsrep_apply_window | 0.000000 || wsrep_causal_reads | 0 || wsrep_cert_deps_distance | 0.000000 || wsrep_cert_index_size | 0 || wsrep_cert_interval | 0.000000 || wsrep_cluster_conf_id | 1 || wsrep_cluster_size | 1 || wsrep_cluster_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_cluster_status | Primary || wsrep_commit_oooe | 0.000000 || wsrep_commit_oool | 0.000000 || wsrep_commit_window | 0.000000 || wsrep_connected | ON || wsrep_evs_delayed | || wsrep_evs_evict_list | || wsrep_evs_repl_latency | 8.91e-07/6.818e-06/1.9412e-05/6.92271e-06/5 || wsrep_evs_state | OPERATIONAL || wsrep_flow_control_paused | 0.000000 || wsrep_flow_control_paused_ns | 0 || wsrep_flow_control_recv | 0 || wsrep_flow_control_sent | 0 || wsrep_gcomm_uuid | b47891c2-1ad6-11e6-b7ef-66d1299a9d50 || wsrep_incoming_addresses | 192.168.10.21:3306 || wsrep_last_committed | 0 || wsrep_local_bf_aborts | 0 || wsrep_local_cached_downto | 18446744073709551615 || wsrep_local_cert_failures | 0 || wsrep_local_commits | 0 || wsrep_local_index | 0 || wsrep_local_recv_queue | 0 || wsrep_local_recv_queue_avg | 0.000000 || wsrep_local_recv_queue_max | 1 || wsrep_local_recv_queue_min | 0 || wsrep_local_replays | 0 || wsrep_local_send_queue | 0 || wsrep_local_send_queue_avg | 0.000000 || wsrep_local_send_queue_max | 1 || wsrep_local_send_queue_min | 0 || wsrep_local_state | 4 || wsrep_local_state_comment | Synced || wsrep_local_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_protocol_version | 7 || wsrep_provider_name | Galera || wsrep_provider_vendor | Codership Oy <[e-mail]> || wsrep_provider_version | 25.3.15(r3578) || wsrep_ready | ON || wsrep_received | 2 || wsrep_received_bytes | 140 || wsrep_repl_data_bytes | 0 || wsrep_repl_keys | 0 || wsrep_repl_keys_bytes | 0 || wsrep_repl_other_bytes | 0 || wsrep_replicated | 0 || wsrep_replicated_bytes | 0 || wsrep_thread_count | 2 |+------------------------------+---------------------------------------------+第9步: 向已经启动的MariaDB Galera群集添加另外两个节点db2和db3检查和确认db2和db3两个节点的配置文件/etc/my.cnf.d/server.cnf内部的[galera] 下面配置的所有参数是否正确，db2和db3两个节点的服务就如同像启动普通服务一样去启动MariaDB Galera节点服务.[root@contoso22 ~]# service mysql start通过在db1和db2节点上执行如下命令，观察群集状态中已经变化参数值内容:[root@contoso22 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"至此，你会发现节点db2已被识别，群集的大小已经是2个节点，并且节点db2的IP地址已被列出：wsrep_local_state_comment | Syncedwsrep_incoming_addre sses | 192.168.10.21:3306,192.168.10.22:3306wsrep_cluster_size | 2wsrep_connected | ONwsrep_ready | ON[root@contoso22 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"+------------------------------+--------------------------------------------------+| Variable_name | Value |+------------------------------+--------------------------------------------------+| wsrep_apply_oooe | 0.000000 || wsrep_apply_oool | 0.000000 || wsrep_apply_window | 0.000000 || wsrep_causal_reads | 0 || wsrep_cert_deps_distance | 0.000000 || wsrep_cert_index_size | 0 || wsrep_cert_interval | 0.000000 || wsrep_cluster_conf_id | 2 || wsrep_cluster_size | 2 || wsrep_cluster_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_cluster_status | Primary || wsrep_commit_oooe | 0.000000 || wsrep_commit_oool | 0.000000 || wsrep_commit_window | 0.000000 || wsrep_connected | ON || wsrep_evs_delayed | || wsrep_evs_evict_list | || wsrep_evs_repl_latency | 0.000227382/0.000654379/0.00225715/0.000659168/7 || wsrep_evs_state | OPERATIONAL || wsrep_flow_control_paused | 0.000000 || wsrep_flow_control_paused_ns | 0 || wsrep_flow_control_recv | 0 || wsrep_flow_control_sent | 0 || wsrep_gcomm_uuid | 8dc1b4b3-1ad7-11e6-8bf0-9a0cfaca8d40 || wsrep_incoming_addresses | 192.168.10.22:3306,192.168.10.21:3306 || wsrep_last_committed | 0 || wsrep_local_bf_aborts | 0 || wsrep_local_cached_downto | 18446744073709551615 || wsrep_local_cert_failures | 0 || wsrep_local_commits | 0 || wsrep_local_index | 0 || wsrep_local_recv_queue | 0 || wsrep_local_recv_queue_avg | 0.000000 || wsrep_local_recv_queue_max | 1 || wsrep_local_recv_queue_min | 0 || wsrep_local_replays | 0 || wsrep_local_send_queue | 0 || wsrep_local_send_queue_avg | 0.333333 || wsrep_local_send_queue_max | 2 || wsrep_local_send_queue_min | 0 || wsrep_local_state | 4 || wsrep_local_state_comment | Synced || wsrep_local_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_protocol_version | 7 || wsrep_provider_name | Galera || wsrep_provider_vendor | Codership Oy <[e-mail]> || wsrep_provider_version | 25.3.15(r3578) || wsrep_ready | ON || wsrep_received | 3 || wsrep_received_bytes | 216 || wsrep_repl_data_bytes | 0 || wsrep_repl_keys | 0 || wsrep_repl_keys_bytes | 0 || wsrep_repl_other_bytes | 0 || wsrep_replicated | 0 || wsrep_replicated_bytes | 0 || wsrep_thread_count | 2 |+------------------------------+--------------------------------------------------+同样地在节点db3上执行如下2个命令:[root@contoso23 ~]# service mysql start # [root@contoso23 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"至此，你应该会发现节点db3已被识别，群集的大小已经是3个节点，并且节点db3的IP地址已被列出：wsrep_local_state_comment | Syncedwsrep_incoming_addresses | 192.168.10.22:3306,192.168.10.21:3306,192.168.10.23:3306wsrep_cluster_size | 3wsrep_connected | ONwsrep_ready | ON[root@contoso23 ~]# mysql -uroot -p123456 -e"show status like 'wsrep%'"+------------------------------+----------------------------------------------------------+| Variable_name | Value |+------------------------------+----------------------------------------------------------+| wsrep_apply_oooe | 0.000000 || wsrep_apply_oool | 0.000000 || wsrep_apply_window | 0.000000 || wsrep_causal_reads | 0 || wsrep_cert_deps_distance | 0.000000 || wsrep_cert_index_size | 0 || wsrep_cert_interval | 0.000000 || wsrep_cluster_conf_id | 3 || wsrep_cluster_size | 3 || wsrep_cluster_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_cluster_status | Primary || wsrep_commit_oooe | 0.000000 || wsrep_commit_oool | 0.000000 || wsrep_commit_window | 0.000000 || wsrep_connected | ON || wsrep_evs_delayed | || wsrep_evs_evict_list | || wsrep_evs_repl_latency | 0/0/0/0/0 || wsrep_evs_state | OPERATIONAL || wsrep_flow_control_paused | 0.000000 || wsrep_flow_control_paused_ns | 0 || wsrep_flow_control_recv | 0 || wsrep_flow_control_sent | 0 || wsrep_gcomm_uuid | 23817fc4-1ad8-11e6-8ec5-3340a442db9a || wsrep_incoming_addresses | 192.168.10.23:3306,192.168.10.22:3306,192.168.10.21:3306 || wsrep_last_committed | 0 || wsrep_local_bf_aborts | 0 || wsrep_local_cached_downto | 18446744073709551615 || wsrep_local_cert_failures | 0 || wsrep_local_commits | 0 || wsrep_local_index | 0 || wsrep_local_recv_queue | 0 || wsrep_local_recv_queue_avg | 0.000000 || wsrep_local_recv_queue_max | 1 || wsrep_local_recv_queue_min | 0 || wsrep_local_replays | 0 || wsrep_local_send_queue | 0 || wsrep_local_send_queue_avg | 0.000000 || wsrep_local_send_queue_max | 1 || wsrep_local_send_queue_min | 0 || wsrep_local_state | 4 || wsrep_local_state_comment | Synced || wsrep_local_state_uuid | b47976ee-1ad6-11e6-b848-c6770ef480cb || wsrep_protocol_version | 7 || wsrep_provider_name | Galera || wsrep_provider_vendor | Codership Oy <[e-mail]> || wsrep_provider_version | 25.3.15(r3578) || wsrep_ready | ON || wsrep_received | 3 || wsrep_received_bytes | 284 || wsrep_repl_data_bytes | 0 || wsrep_repl_keys | 0 || wsrep_repl_keys_bytes | 0 || wsrep_repl_other_bytes | 0 || wsrep_replicated | 0 || wsrep_replicated_bytes | 0 || wsrep_thread_count | 2 |+------------------------------+----------------------------------------------------------+第10步: 测试验证群集的复制功能现在群集正在运行,让我们测试一下它是否能正常工作.在db1节点（contoso21）上创建一个‘clustertest’ 数据库，通过如下命令:[root@contoso21 ~]# mysql -uroot -p123456 -e 'SELECT * FROM clustertest.mycluster;'[root@contoso21 ~]# mysql -uroot -p123456 -e 'CREATE DATABASE clustertest;'[root@contoso21 ~]# mysql -uroot -p123456 -e 'CREATE TABLE clustertest.mycluster ( id INT NOT NULL AUTO_INCREMENT,name VARCHAR(50),birth DATE,PRIMARY KEY(id));'[root@contoso21 ~]# mysql -uroot -p123456 -e 'INSERT INTO clustertest.mycluster (name,birth) VALUES ("Tim","1989-12-27");'[root@contoso21 ~]# mysql -uroot -p123456 -e 'SELECT * FROM clustertest.mycluster;'+----+------+------------+| id | name | birth |+----+------+------------+| 3 | Tim | 1989-12-27 |+----+------+------------+[root@contoso22 ~]# mysql -uroot -p123456 -e 'SELECT * FROM clustertest.mycluster;'+----+------+------------+| id | name | birth |+----+------+------------+| 3 | Tim | 1989-12-27 |+----+------+------------+[root@contoso23 ~]# mysql -uroot -p123456 -e 'SELECT * FROM clustertest.mycluster;'+----+------+------------+| id | name | birth |+----+------+------------+| 3 | Tim | 1989-12-27 |+----+------+------------+从以上输出的结果我们能断定群集是正常工作的，群集通过了节点db1实现了与所有节点之间的复制操作.当我们群集关闭群集时，按照逆向启动顺序去关闭节点先关闭db3节点,，然后关闭db2节点，最后关闭db1节点当我们再一次启动群集时，按照以下命令顺序启动群集：[root@contoso21 ~]# /etc/init.d/mysql start --wsrep-new-cluster ## 第1步，第1个节点必须如此启动[root@contoso22 ~]# service mysql start ## 第2步[root@contoso23 ~]# service mysql start ## 第3步[root@contoso21 ~]# systemctl status mysql -l[root@contoso22 ~]# systemctl status mysql -l[root@contoso23 ~]# systemctl status mysql -lKeepAlived + HaProxy的高可用性方案配置hostname ip address VIP loadbalance1.com 192.168.100.55 192.168.10.200loadbalance2.com 192.168.100.66 192.168.10.200[root@loadbalance1 ~]# ss -anop | column[root@loadbalance2 ~]# ss -anop | column使用vi命令删除全部被编辑文件的内容第1步：输入 G 光标跳到文件末尾第2步：输入 :1,.d 回车当我们分配一个IP地址192.168.10.200当作虚拟IP地址(VIP)来使用的时候.我们需要做的第一件事情是告诉操作系统的内核,我们计划将IP地址绑定到没有被定义的接口上.为了达到这个目的我们需要编辑配置文件/etc/sysctl.conf，并且在该文件末尾添加如下行:net.ipv4.ip_nonlocal_bind=1[root@loadbalance1 ~]# vi /etc/sysctl.confnet.ipv4.ip_nonlocal_bind=1[root@loadbalance2 ~]# vi /etc/sysctl.confnet.ipv4.ip_nonlocal_bind=1然后，我们需要执行如下命令在不重新启动系统的情况下能够使VIP设置生效:[root@loadbalance1 ~]# sysctl -p[root@loadbalance2 ~]# sysctl -p[root@loadbalance1 ~]# vi /etc/hosts192.168.100.55 loadbalance1.com loadbalance1127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4::1 localhost localhost.localdomain localhost6 localhost6.localdomain6[root@loadbalance2 ~]# vi /etc/hosts192.168.100.66 loadbalance2.com loadbalance2127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4::1 localhost localhost.localdomain localhost6 localhost6.localdomain6[root@loadbalance1 ~]# vi /etc/selinux/config# disabled - No SELinux policy is loaded.SELINUX=disabled[root@loadbalance2 ~]# vi /etc/selinux/config# disabled - No SELinux policy is loaded.SELINUX=disabled[root@loadbalance1 ~]# systemctl stop firewalld && systemctl disable firewalld[root@loadbalance2 ~]# systemctl stop firewalld && systemctl disable firewalld[root@loadbalance1 ~]# yum -y install keepalived haproxy[root@loadbalance2 ~]# yum -y install keepalived haproxy[root@loadbalance1 ~]# cat /etc/keepalived/keepalived.conf[root@loadbalance2~]# cat /etc/keepalived/keepalived.conf[root@loadbalance1 ~]# vi /etc/keepalived/keepalived.confglobal_defs {notification_email {[e-mail]}notification_email_from [e-mail]_server localhostsmtp_connect_timeout 30router_id loadbalance1.com}vrrp_script haproxy {script "killall -0 haproxy"interval 2weight 2}vrrp_instance 50 {virtual_router_id 50advert_int 1priority 101state MASTERinterface eno16777736virtual_ipaddress {192.168.10.200 dev eno16777736}track_script {haproxy}}[root@loadbalance2 ~]# vi /etc/keepalived/keepalived.conf global_defs {notification_email {[e-mail]}notification_email_from [e-mail]_server localhostsmtp_connect_timeout 30router_id loadbalance2.com}vrrp_script haproxy {script "killall -0 haproxy"interval 2weight 2}vrrp_instance 50 {virtual_router_id 50advert_int 1priority 99state SLAVEinterface eno16777736virtual_ipaddress {192.168.10.200 dev eno16777736}track_script {haproxy}}[root@loadbalance1 ~]# vi /etc/haproxy/haproxy.cfgglobal chroot /var/lib/haproxy daemon group haproxy log 192.168.100.55 local2 maxconn 4000 pidfile /var/run/haproxy.pid stats socket /var/lib/haproxy/stats user haproxy defaults log global retries 3 timeout queue 1m timeout connect 10s timeout client 1m timeout server 1m listen mariadb-cluster bind 192.168.10.200:3306 mode tcp balance roundrobin server mariadb1 192.168.10.21:3306 server mariadb2 192.168.10.22:3306 server mariadb3 192.168.10.23:3306[root@loadbalance2 ~]# vi /etc/haproxy/haproxy.cfgglobal chroot /var/lib/haproxy daemon group haproxy log 192.168.10.66 local2 maxconn 4000 pidfile /var/run/haproxy.pid stats socket /var/lib/haproxy/stats user haproxy defaults log global retries 3 timeout queue 1m timeout connect 10s timeout client 1m timeout server 1m listen mariadb-cluster bind 192.168.10.200:3306 mode tcp balance roundrobin server mariadb1 192.168.10.21:3306 server mariadb2 192.168.10.22:3306 server mariadb3 192.168.10.23:3306[root@loadbalance1 ~]# vi /etc/default/haproxyENABLED=1[root@loadbalance2 ~]# vi /etc/default/haproxyENABLED=1[root@loadbalance1 ~]# systemctl enable keepalived[root@loadbalance2 ~]# systemctl enable keepalived[root@loadbalance1 ~]# systemctl restart keepalived[root@loadbalance2 ~]# systemctl restart keepalived[root@loadbalance1 ~]# systemctl enable haproxy[root@loadbalance2 ~]# systemctl enable haproxy[root@loadbalance1 ~]# systemctl restart haproxy[root@loadbalance2 ~]# systemctl restart haproxyshow status like 'wsrep%';show variables like 'wsrep_node%';[root@loadbalance1 ~]# cat > /etc/yum.repos.d/MariaDB.repo[root@loadbalance2 ~]# cat > /etc/yum.repos.d/MariaDB.repo[mariadb]name = MariaDBbaseurl = http://yum.mariadb.org/10.1/centos7-amd64gpgkey=https://yum.mariadb.org/RPM-GPG-KEY-MariaDBgpgcheck=1 [root@loadbalance1 ~]# yum -y install MariaDB-client[root@loadbalance2 ~]# yum -y install MariaDB-client[root@loadbalance1 ~]# mysql -uroot -p123456 -h192.168.10.200 -e"show variables like 'wsrep_node%'"+-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.21 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db1 |+-----------------------------+-----------------+[root@loadbalance1 ~]# mysql -uroot -p123456 -h192.168.10.200 -e"show variables like 'wsrep_node%'"+-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10 .22 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db2 |+-----------------------------+-----------------+[root@loadbalance1 ~]# mysql -uroot -p123456 -h192.168.10.200 -e"show variables like 'wsrep_node%'"+-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.23 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db3 |+-----------------------------+-----------------+[root@loadbalance1 ~]# for i in `seq 1 6`domysql -uroot -p123456 -h192.168.10.200 -e"show variables like 'wsrep_node%'"done[root@loadbalance1 ~]# for i in `seq 1 6`> do> mysql -uroot -p123456 -h192.168.10.200 -e"show variables like 'wsrep_node%'"> done+-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.21 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db1 |+-----------------------------+-----------------++-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.22 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db2 |+-----------------------------+-----------------++-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.23 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db3 |+-----------------------------+-----------------++-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.21 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db1 |+-----------------------------+-----------------++-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.22 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db2 |+-----------------------------+-----------------++-----------------------------+-----------------+| Variable_name | Value |+-----------------------------+-----------------+| wsrep_node_address | 192.168.10.23 || wsrep_node_incoming_address | AUTO || wsrep_node_name | db3 |+-----------------------------+-----------------+配置haproxy日志功能[root@loadbalance1 ~]# vi /etc/rsyslog.conf# Provides UDP syslog reception$ModLoad imudp$UDPServerRun 514local2.* /var/log/haproxy.log[root@loadbalance1 ~]# vi /etc/sysconfig/rsyslogSYSLOGD_OPTIONS="-r -m 0"[root@loadbalance1 ~]# systemctl restart rsyslog[root@loadbalance1 ~]# tail -f /var/log/haproxy.log