YNOTES笔记

创建用户(主库) ```sql mysql> CREATE USER 'repl'@'%' IDENTIFIED BY 'xxxxxx'; mysql> GRANT REPLICATION SLAVE ON *.* TO 'repl'@'%' ; mysql> flush privileges; ``` 创建备份文件(主库) `方式一:备份生成GTID` ```bash mysqldump -hxxxxx.mysql.rds.aliyuncs.com -uroot -pxxxxxx --databases db1 db2 db3 >db.sql ``` `方式一:备份生成binlog位置` ```bash mysqldump -hxxxxx.mysql.rds.aliyuncs.com --master-data=2 --set-gtid-purged=OFF -uroot -pxxxxxx --databases db1 db2 db3 >db.sql ``` 还原备份(从库) ```sql mysql> source db.sql ``` 创建同步(从库) `方式一:备份生成GTID方式` ```sql mysql> CHANGE MASTER TO MASTER_HOST='xxxxx.mysql.rds.aliyuncs.com', MASTER_USER='repl', MASTER_PASSWORD='xxxxxx', master_auto_position=1 FOR CHANNEL "test"; ``` `方式二:备份生成binlog位置方式` ```sql mysql> CHANGE MASTER TO MASTER_HOST='xxxxx.mysql.rds.aliyuncs.com', MASTER_USER='repl', MASTER_PASSWORD='xxxxxx', MASTER_LOG_FILE='mysql-bin.xxxx', MASTER_LOG_POS=xxxxx FOR CHANNEL "test"; ``` 启动同步(从库) ```bash mysql> START SLAVE FOR CHANNEL "test"; mysql> show slave status FOR CHANNEL "test"\G ```

##### 1.下载SymmetricDS [http://www.symmetricds.org/](https://www.symmetricds.org/) #### 2.安装配置 ##### 2.1解压安装包到E盘重命名为sym-38 ##### 2.2配置主从配置拷贝安装包下的samples下的corp-000.properties和store-001.properties到目录engines 配置sqlserver(主) ```properties engine.name=corp-000 db.driver=net.sourceforge.jtds.jdbc.Driver db.url=jdbc:jtds:sqlserver://172.16.13.3:1433/dingding_source;sendStringParametersAsUnicode=false; useCursors=true;bufferMaxMemory=10240;lobBuffer=5242880 db.user=sa db.password=xxxxxxxx registration.url= sync.url=http://172.16.13.3:8050/sync/corp-000 group.id=corp external.id=000 job.purge.period.time.ms=7200000 job.routing.period.time.ms=5000 job.push.period.time.ms=10000 job.pull.period.time.ms=10000 auto.registration=true initial.load.create.first=true ``` 配置mysql(从) ```properties engine.name=store-001 db.driver=com.mysql.jdbc.Driver db.url=jdbc:mysql://172.16.13.23/shudoon-dingding-source?tinyInt1isBit=false&useSSL=false db.user=root db.password=xxxxxx registration.url=http://172.16.13.3:8050/sync/corp-000 group.id=store external.id=001 job.routing.period.time.ms=5000 job.push.period.time.ms=10000 job.pull.period.time.ms=10000 ``` #### 3.初始化sym系统表 ```bat symadmin --engine corp-000 create-sym-tables ``` #### 4.重置sym系统表数据 ```bat dbimport --engine corp-000 E:\sym-38\samples\insert_sample.sql ``` #### 5.创建业务表 ```bat dbimport --engine store-001 --format XML E:\sym-38\samples\create_sample.xml ``` #### 6.启动master端 ```bat sym --engine corp-000 --port 8050 ``` #### 7.启动slave端 ```bat sym --engine store-001 --port 8010 ``` #### 8.初始化item表结构 ```bat symadmin --engine corp-000 send-schema item --node 001 ``` #### 8.初始化item表数据 ```bat symadmin --engine corp-000 reload-table item --node 001 ``` #### 9.新增表tCommunicationRecords同步主库执行 ```sql insert into sym_channel (channel_id, processing_order, max_batch_size, enabled, description) values('c_tCommunicationRecords', 3, 100000, 1, ''); insert into sym_router (router_id,source_node_group_id,target_node_group_id,router_type,create_time,last_update_time) values('corp_tCommunicationRecords_2_store', 'corp', 'store', 'default',current_timestamp, current_timestamp); insert into sym_trigger (trigger_id,source_table_name,channel_id,last_update_time,create_time) values('tCommunicationRecords','tCommunicationRecords','c_tCommunicationRecords',current_timestamp,current_timestamp); insert into sym_trigger_router (trigger_id,router_id,initial_load_order,last_update_time,create_time) values('tCommunicationRecords','corp_tCommunicationRecords_2_store', 200, current_timestamp, current_timestamp); ``` 命令行执行 ```bat symadmin --engine corp-000 send-schema tCommunicationRecords --node 001 symadmin --engine corp-000 reload-table tCommunicationRecords --node 001 ``` #### 常见问题: `1.处理sqlserver联合自增主键的表结构无法同步到mysql的问题？` 解决方法: 1.去掉建表语句中联合主键中非自增列，拷贝SQL语句手动在mysql建表。 2.修改sym_outgoing_batch表中batch_id对应数据的状态为OK ```sql update sym_outgoing_batch set status='OK' where batch_id='XXXXXX'; ``` 3.删除sym_data_event中对应batch_id和data_id的事件数据 ```sql delete from sym_data_event where batch_id='XXXXXX' and data_id='YYYYYY'; ```

### 一、环境准备系统: `CentOS7` 数据库: `ProxySQL:1.4.14` 服务器: `master`: `172.16.0.100/db1` `slave`: `172.16.0.101/db2` `slave/ProxySQL`: `172.16.0.102/db3` &emsp; &emsp; ### 二、准备工作 `a.数据库搭建MySQL主从` `b.从库开启`**`read_only=on`** &emsp; &emsp; ### 三、ProxySQL的安装 #### 安装依赖的软件包 `db3` ```bash yum -y install perl-DBD-MySQL perl-DBI perl-Time-HiRes perl-IO-Socket-SSL ``` &emsp; #### ProxySQL软件包的两个下载地址 GitHub官网：https://github.com/sysown/proxysql/releases percona官网：https://www.percona.com/downloads/proxysql/ &emsp; #### 安装ProxySQL `db3` ```bash yum install -y https://www.percona.com/downloads/proxysql/proxysql-1.4.14/binary/redhat/7/x86_64/proxysql-1.4.14-1.1.el7.x86_64.rpm ``` &emsp; #### 启动ProxySQL `db3` ```bash systemctl start proxysql ``` &emsp; #### 查看启动信息 `db3` ```bash netstat -tunlp |grep proxysql ``` ``` tcp 0 0 0.0.0.0:6032 0.0.0.0:* LISTEN 13331/proxysql tcp 0 0 0.0.0.0:6033 0.0.0.0:* LISTEN 13331/proxysql ``` `6032为管理端口，6033为对外服务的端口号` &emsp; #### 登录ProxySQL `db3` ```bash mysql -uadmin -padmin -h 127.0.0.1 -P 6032 ``` &emsp; #### 查看proxysql信息 `db3` ```sql mysql> show databases; +-----+---------------+-------------------------------------+ | seq | name | file | +-----+---------------+-------------------------------------+ | 0 | main | | | 2 | disk | /var/lib/proxysql/proxysql.db | | 3 | stats | | | 4 | monitor | | | 5 | stats_history | /var/lib/proxysql/proxysql_stats.db | +-----+---------------+-------------------------------------+ #可见有四个库：main、disk、stats和monitor。分别说明一下这四个库的作用。 #main：内存配置数据库，即MEMORY，表里存放后端db实例、用户验证、路由规则等信息。 mysql> show tables; +--------------------------------------------+ | tables | +--------------------------------------------+ | global_variables | | mysql_collations | | mysql_group_replication_hostgroups | | mysql_query_rules | | mysql_query_rules_fast_routing | | mysql_replication_hostgroups | | mysql_servers | | mysql_users | | proxysql_servers | | runtime_checksums_values | | runtime_global_variables | | runtime_mysql_group_replication_hostgroups | | runtime_mysql_query_rules | | runtime_mysql_query_rules_fast_routing | | runtime_mysql_replication_hostgroups | | runtime_mysql_servers | | runtime_mysql_users | | runtime_proxysql_servers | | runtime_scheduler | | scheduler | +--------------------------------------------+ #库下的主要表： #mysql_servers—后端可以连接MySQL服务器的列表。 #mysql_users—配置后端数据库的账号和监控的账号。 #mysql_query_rules—指定Query路由到后端不同服务器的规则列表。 #注：表名以runtime_开头的表示ProxySQL当前运行的配置内容，不能通过DML语句修改。只能修改对应的不以 runtime开头的表，然后“LOAD”使#其生效，“SAVE”使其存到硬盘以供下次重启加载。 #disk库—持久化磁盘的配置。 #stats库—统计信息的汇总。 #monitor库—一些监控的收集信息，包括数据库的健康状态等。 ``` &emsp; &emsp; ### 四、配置ProxySQL监控 #### Master上创建ProxySQL的监控账户和对外访问账户并赋予权限 `db1` ```sql mysql> create user 'monitor'@'172.16.0.%' identified by 'monitor'; mysql> grant all privileges on *.* to 'monitor'@'172.16.0.%' with grant option; mysql> create user 'proxysql'@'172.16.0.%' identified by 'proxysql'; mysql> grant all privileges on *.* to 'proxysql'@'172.16.0.%' with grant option; mysql> flush privileges; ``` &emsp; #### ProxySQL添加主从服务器列表 `db3` ```sql mysql> insert into mysql_servers(hostgroup_id,hostname,port) values(10,'172.16.0.100',3306); mysql> insert into mysql_servers(hostgroup_id,hostname,port) values(10,'172.16.0.101',3306); mysql> insert into mysql_servers(hostgroup_id,hostname,port) values(10,'172.16.0.102',3306); ``` &emsp; #### ProxySQL查看服务器列表 `db3` ```sql mysql> select * from mysql_servers; +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ | hostgroup_id | hostname | port | status | weight | compression | max_connections | max_replication_lag | use_ssl | max_latency_ms | comment | +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ | 10 | 172.16.0.100 | 3306 | ONLINE | 1 | 0 | 1000 | 0 | 0 | 0 | | | 20 | 172.16.0.101 | 3306 | ONLINE | 10 | 0 | 1000 | 0 | 0 | 0 | | | 20 | 172.16.0.102 | 3306 | ONLINE | 1 | 0 | 1000 | 0 | 0 | 0 | | +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ ``` &emsp; #### ProxySQL配置监控账号 `db3` ```sql mysql> set mysql-monitor_username='monitor'; mysql> set mysql-monitor_password='monitor'; mysql> load mysql variables to runtime; mysql> save mysql variables to disk; ``` &emsp; #### ProxySQL验证监控信息 `db3` ```sql mysql> select * from monitor.mysql_server_connect_log order by time_start_us desc limit 6; +--------------+------+------------------+-------------------------+---------------+ | hostname | port | time_start_us | connect_success_time_us | connect_error | +--------------+------+------------------+-------------------------+---------------+ | 172.16.0.100 | 3306 | 1565593172754235 | 2118 | NULL | | 172.16.0.101 | 3306 | 1565593172137991 | 2729 | NULL | | 172.16.0.102 | 3306 | 1565593171521858 | 773 | NULL | | 172.16.0.100 | 3306 | 1565593113076006 | 2163 | NULL | | 172.16.0.101 | 3306 | 1565593112298748 | 2377 | NULL | | 172.16.0.102 | 3306 | 1565593111521583 | 628 | NULL | +--------------+------+------------------+-------------------------+---------------+ ``` &emsp; &emsp; ### 五、配置ProxySQL主从分组信息 #### ProxySQL插入读写分离组 `db3` ```sql mysql> insert into mysql_replication_hostgroups(writer_hostgroup,reader_hostgroup,comment) values (10,20,'proxy'); mysql> load mysql servers to runtime; mysql> save mysql servers to disk; # writer_hostgroup是写入组的编号，reader_hostgroup是读取组的编号。实验中使用10作为写入组，20作为读取组编号。 mysql> select * from mysql_servers; +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ | hostgroup_id | hostname | port | status | weight | compression | max_connections | max_replication_lag | use_ssl | max_latency_ms | comment | +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ | 10 | 172.16.0.100 | 3306 | ONLINE | 1 | 0 | 1000 | 0 | 0 | 0 | | | 20 | 172.16.0.101 | 3306 | ONLINE | 10 | 0 | 1000 | 0 | 0 | 0 | | | 20 | 172.16.0.102 | 3306 | ONLINE | 1 | 0 | 1000 | 0 | 0 | 0 | | +--------------+--------------+------+--------+--------+-------------+-----------------+---------------------+---------+----------------+---------+ # ProxySQL会根据server的read_only的取值将服务器进行分组。read_only=0的server，master被分到编号为10的写组，read_only=1的server，slave则被分到编号为20的读组。 ``` &emsp; #### ProxySQL配置对外访问账号(`默认指定主库，并对该用户开启事务持久化保护`) `db3` ```sql insert into mysql_users(username,password,default_hostgroup) values ('proxysql','proxysql',10); update mysql_users set transaction_persistent=1 where username='proxysql'; #设置为1，避免发生脏读、幻读等现象 load mysql users to runtime; save mysql users to disk; ``` &emsp; #### ProxySQL验证登录的服务器 `db3` ```bash mysql -uproxysql -pproxysql -h 172.16.0.102 -P 6033 ``` ```sql mysql> select @@hostname; +------------+ | @@hostname | +------------+ | db1 | +------------+ # 验证登入的服务器默认为主库(db1) ``` &emsp; &emsp; ### 六、ProxySQL配置读写分离策略 #### ProxySQL配置读写分离 `db3` ```sql mysql> insert into mysql_query_rules(active,match_pattern,destination_hostgroup, apply) VALUES(1,'^SELECT.*FOR UPDATE$',10,1); mysql> insert into mysql_query_rules(active,match_pattern,destination_hostgroup, apply) VALUES(1,'^SELECT',20,1); mysql> load mysql query rules to runtime; mysql> save mysql query rules to disk; ``` &emsp; #### ProxySQL配置服务器的权重 `db3` ```sql mysql> update mysql_servers set weight=10 where hostname='172.16.0.101'; mysql> load mysql servers to runtime; mysql> save mysql servers to disk; # 172.16.0.101的权重被修改为10,172.16。0.102的默认权重为1，则请求比例为10:1 ``` &emsp; &emsp; ### 七、测试读写分离 #### 执行select查询 `db3` ```sql mysql -uproxysql -pproxysql -h 172.16.0.102 -P 6033 ``` ```sql mysql> select node_name from percona.example where node_id=1; ``` &emsp; #### 执行select for update查询 `db3` ```sql mysql> select node_name from percona.example where node_id=1 for update; ``` &emsp; #### 执行update语句 `db3` ```sql mysql> update percona.example set node_name='test' where node_id=1; ``` &emsp; #### proxySQL查看统计信息 `db3` ```bash mysql -uadmin -padmin -h 127.0.0.1 -P 6032 ``` ```sql mysql> select hostgroup,digest_text from stats_mysql_query_digest; +-----------+--------------------------------------------------------------------------------------------------+ | hostgroup | digest_text | +-----------+--------------------------------------------------------------------------------------------------+ | 10 | update percona.example set node_name=? where node_id=? | | 10 | select node_name,@@hostname from percona.example where node_id=? for update | | 20 | select node_name,@@hostname from percona.example where node_id=? | +-----------+--------------------------------------------------------------------------------------------------+ ``` `从上面的输出信息可以发现读写分离成功了` 参考: https://blog.51cto.com/sumongodb/2130453

### 一、环境准备系统: `CentOS7` 数据库: `Percona-XtraDB-Cluster-57` 服务器: `master1`: `172.16.0.100/db1` `master2`: `172.16.0.101/db2` `master2`: `172.16.0.102/db3` &emsp; &emsp; ### 二、准备工作 `a.删除mysql-community` ```bash yum remove mysql-community-client mysql-community-server -y #仅安装了mysql-community需要执行 ``` `b.关闭防火墙` `c.关闭Selinux` &emsp; &emsp; ### 三、安装PXC集群 `所有节点` 安装Percona库 ```bash yum install https://repo.percona.com/yum/percona-release-latest.noarch.rpm -y ``` 安装Percona-XtraDB-Cluster-57 ```bash yum install Percona-XtraDB-Cluster-57 -y ``` 启动MySQL ```bash systemctl start mysql.service ``` 获取初始密码 ```bash grep 'temporary password' /var/log/mysqld.log ``` ``` 2019-08-09T03:22:26.358453Z 1 [Note] A temporary password is generated for root@localhost: so*WrNqjm3(e ``` 修改root密码 ```sql mysql -u root -p mysql> ALTER USER 'root'@'localhost' IDENTIFIED BY 'rootPass'; mysql> exit ``` 停止MySQL ```bash systemctl stop mysql.service ``` &emsp; &emsp; ### 四、引导第一个节点修改配置 `Master1` ```bash vim /etc/percona-xtradb-cluster.conf.d/wsrep.cnf ``` ```ini [mysqld] wsrep_provider=/usr/lib64/galera3/libgalera_smm.so wsrep_cluster_address=gcomm:// #配置wsrep_cluster_address=gcomm:// 为了初始化集群，添加其他节点之后再修改回正常的配置，再重启节点即可 binlog_format=ROW default_storage_engine=InnoDB wsrep_slave_threads= 8 wsrep_log_conflicts innodb_autoinc_lock_mode=2 wsrep_cluster_name=pxc-cluster-test #集群名字 wsrep_node_name=db1 #节点名 wsrep_node_address=172.16.0.100 #当前节点IP pxc_strict_mode=ENFORCING wsrep_sst_method=xtrabackup-v2 #SST传输方法 wsrep_sst_auth=sstuser:passw0rd #SST账号 ``` bootstrap启动节点 ```bash systemctl start mysql@bootstrap.service ``` 查看集群信息 ```sql mysql> show status like 'wsrep%'; +----------------------------+--------------------------------------+ | Variable_name | Value | +----------------------------------+--------------------------------+ | wsrep_local_state_uuid | ee7be278-ba54-11e9-9621-8ee7979a72d4 | | ... | ... | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | ... | ... | | wsrep_incoming_addresses | 172.16.0.100:3306 | | wsrep_cluster_size | 1 | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | ... | ... | | wsrep_ready | ON | +----------------------------+--------------------------------------+ ``` `上面信息显示集群几点为1，节点状态为Synced` 创建SST用户 ```sql mysql CREATE USER 'sstuser'@'localhost' IDENTIFIED BY 'passw0rd'; mysql> GRANT RELOAD, LOCK TABLES, PROCESS, REPLICATION CLIENT ON *.* TO 'sstuser'@'localhost'; mysql> FLUSH PRIVILEGES; ``` &emsp; &emsp; ### 五、添加节点到集群 #### 添加Master2到集群修改配置 `Master2` ```bash vim /etc/percona-xtradb-cluster.conf.d/wsrep.cnf ``` ```bash [mysqld] wsrep_provider=/usr/lib64/galera3/libgalera_smm.so wsrep_cluster_address=gcomm://172.16.0.100,172.16.0.101,172.16.0.102 binlog_format=ROW default_storage_engine=InnoDB wsrep_slave_threads= 8 wsrep_log_conflicts innodb_autoinc_lock_mode=2 wsrep_cluster_name=pxc-cluster-test wsrep_node_name=db2 wsrep_node_address=172.16.0.101 pxc_strict_mode=ENFORCING wsrep_sst_method=xtrabackup-v2 wsrep_sst_auth=sstuser:passw0rd ``` 启动节点 ```bash systemctl start mysql ``` 查看集群状态 ```sql mysql> show status like 'wsrep%'; +----------------------------+--------------------------------------+ | Variable_name | Value | +----------------------------------+--------------------------------+ | wsrep_local_state_uuid | ee7be278-ba54-11e9-9621-8ee7979a72d4 | | ... | ... | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | ... | ... | | wsrep_incoming_addresses | 172.16.0.101:3306,172.16.0.100:3306 | | wsrep_cluster_size | 2 | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | ... | ... | | wsrep_ready | ON | +----------------------------+--------------------------------------+ ``` `上面信息显示集群现在为2个节点,172.16.0.101:3306,172.16.0.100:3306` #### 添加Master3到集群修改配置 `Master3` ```bash vim /etc/percona-xtradb-cluster.conf.d/wsrep.cnf ``` ```bash [mysqld] wsrep_provider=/usr/lib64/galera3/libgalera_smm.so wsrep_cluster_address=gcomm://172.16.0.100,172.16.0.101,172.16.0.102 binlog_format=ROW default_storage_engine=InnoDB wsrep_slave_threads= 8 wsrep_log_conflicts innodb_autoinc_lock_mode=2 wsrep_cluster_name=pxc-cluster-test wsrep_node_name=db3 wsrep_node_address=172.16.0.102 pxc_strict_mode=ENFORCING wsrep_sst_method=xtrabackup-v2 wsrep_sst_auth=sstuser:passw0rd ``` 启动节点 ```bash systemctl start mysql ``` 查看集群状态 ```sql mysql> show status like 'wsrep%'; +----------------------------+-------------------------------------------------------+ | Variable_name | Value | +----------------------------+-------------------------------------------------------+ | wsrep_local_state_uuid | ee7be278-ba54-11e9-9621-8ee7979a72d4 | | ... | ... | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | ... | ... | | wsrep_incoming_addresses | 172.16.0.102:3306,172.16.0.101:3306,172.16.0.100:3306 | | wsrep_cluster_size | 3 | | wsrep_cluster_status | Primary | | wsrep_connected | ON | | ... | ... | | wsrep_ready | ON | +----------------------------+-------------------------------------------------------+ ``` `上面信息显示集群现在为3个节点,172.16.0.102:3306,172.16.0.101:3306,172.16.0.100:3306` 重新添加Master1到集群停止MySQL `Master1` ```bash systemctl stop mysql@bootstrap.service ``` 修改配置 `Master1` ```sql vim /etc/percona-xtradb-cluster.conf.d/wsrep.cnf ``` ```ini ... #wsrep_cluster_address=gcomm:// 修改前 wsrep_cluster_address=gcomm://172.16.0.100,172.16.0.101,172.16.0.102 #修改后 ... ``` 启动数据库 `Master1` ```bash systemctl start mysql ``` &emsp; &emsp; ### 六、验证集群在Master2节点插入测试数据 `Master2` ```sql mysql> CREATE DATABASE percona; mysql> USE percona; mysql> CREATE TABLE example (node_id INT PRIMARY KEY, node_name VARCHAR(30)); mysql> INSERT INTO example VALUES (1, 'percona1'); ``` 所有节点查询数据 `所有节点` ```sql mysql> SELECT * FROM percona.example; +---------+-----------+ | node_id | node_name | +---------+-----------+ | 1 | percona1 | +---------+-----------+ ``` `测试发现所有数据都已经同步，说明集群搭建成功!` &emsp; &emsp; ### 七、从节点在线转换成PXC节点 #### **`说明:测试使用db1做主节点,db3做从节点`** #### 重置db3 `a.清理旧的数据并初始化数据库` ```bash systemctl stop mysql mv /var/lib/mysql/ /var/lib/mysql_bak/ mkdir /var/lib/mysql mv /etc/percona-xtradb-cluster.conf.d/wsrep.cnf /tmp #移除PXC相关配置 chown -R mysql.mysql /var/lib/mysql systemctl start mysql #启动并初始化数据库 ``` `b.获取初始密码` ```bash grep 'temporary password' /var/log/mysqld.log ``` ``` 2019-08-09T03:22:26.358453Z 1 [Note] A temporary password is generated for root@localhost: so*WrNqjm3(e ``` `c.修改root密码` ```sql mysql -u root -p mysql> ALTER USER 'root'@'localhost' IDENTIFIED BY 'rootPass'; mysql> exit ``` &emsp; #### 创建复制账号 `db1` ```sql mysql> create user 'bak'@'172.16.0.%' identified by '123456'; mysql> grant replication slave on *.* to 'bak'@'172.16.0.%'; mysql> flush privileges; ``` #### 备份数据库并发送到db3 `db1` ```bash mysqldump -hlocalhost -uroot -pxxxxxx --single-transaction -A --master-data=2 > all.sql scp all.sql db3:/root ``` #### 从节点执行数据库恢复 `db3` ```bash mysql -hlocalhost -uroot -prootPass </root/all.sql ``` #### 配置从节点同步主节点 `db3` ```sql mysql> CHANGE MASTER TO MASTER_HOST='172.16.0.100',MASTER_USER='bak',MASTER_PASSWORD='test123',MASTER_LOG_FILE='db1-bin.000004', MASTER_LOG_POS=88754; mysql> start slave; ``` #### 停止从节点获取同步位置 `db3` ```sql mysql> stop slave; #停止slave mysql> show slave status\G #查看同步位置 *************************** 1. row *************************** Slave_IO_State: Master_Host: 172.16.0.100 Master_User: bak Master_Port: 3306 Connect_Retry: 60 Master_Log_File: db1-bin.000004 Read_Master_Log_Pos: 190214 Relay_Log_File: db3-relay-bin.000003 Relay_Log_Pos: 101778 Relay_Master_Log_File: db1-bin.000004 Slave_IO_Running: No Slave_SQL_Running: No Replicate_Do_DB: Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: Last_Errno: 0 Last_Error: Skip_Counter: 0 Exec_Master_Log_Pos: 190214 ``` `上面的输出显示同步文件:db1-bin.000004，位置为:190214` #### 从节点重置slave配置(`下次启动直接使用PXC去同步`) `db3` ```sql mysql> reset slave all; ``` #### 从节点停止MySQL `db3` ```bash systemctl stop mysql ``` #### 主节点查看同步的Position对应Xid `db1` ```bash mysqlbinlog db1-bin.000004 |grep Xid|grep 190214 ``` ``` #190809 8:57:53 server id 1 end_log_pos 190214 CRC32 0x4cfb3e3b Xid = 675 ``` `确认Postion 190214对应的Xid为675` #### 主节点拷贝grastate.dat文件到从节点 `db1` ```bash scp grastate.dat db3:/var/lib/mysql ``` #### 从节点修改grastate.dat `db3` ```bash cat grastate.dat ``` ```ini # GALERA saved state version: 2.1 uuid: ee7be278-ba54-11e9-9621-8ee7979a72d4 seqno: 675 #-1修改为上面的Xid safe_to_bootstrap: 0 ``` ```bash chown mysql.mysql /var/lib/mysql/grastate.dat ``` #### 加入PXC相关配置 `db3` ```bash vim /etc/percona-xtradb-cluster.conf.d/wsrep.cnf ``` ```ini [mysqld] wsrep_provider=/usr/lib64/galera3/libgalera_smm.so wsrep_cluster_address=gcomm://172.16.0.100,172.16.0.101,172.16.0.102 binlog_format=ROW default_storage_engine=InnoDB wsrep_slave_threads= 8 wsrep_log_conflicts innodb_autoinc_lock_mode=2 wsrep_cluster_name=pxc-cluster-test wsrep_node_name=db3 wsrep_node_address=172.16.0.102 pxc_strict_mode=ENFORCING wsrep_sst_method=xtrabackup-v2 wsrep_sst_auth=sstuser:passw0rd ``` #### 启动数据库 `db3` ```bash systemctl start mysql ``` #### 查看集群信息 `db3` ```sql mysql> show status like 'wsrep%'; +----------------------------------+-------------------------------------------------------+ | Variable_name | Value | +----------------------------------+-------------------------------------------------------+ | wsrep_local_state_uuid | ee7be278-ba54-11e9-9621-8ee7979a72d4 | | wsrep_protocol_version | 9 | | wsrep_last_applied | 1431 | | wsrep_last_committed | 1431 | | 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 | 247 | | wsrep_received_bytes | 77425 | | 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 | 2 | | wsrep_local_recv_queue_min | 0 | | wsrep_local_recv_queue_avg | 0.004049 | | wsrep_local_cached_downto | 1191 | | wsrep_flow_control_paused_ns | 0 | | wsrep_flow_control_paused | 0.000000 | | wsrep_flow_control_sent | 0 | | wsrep_flow_control_recv | 0 | | wsrep_flow_control_interval | [ 173, 173 ] | | wsrep_flow_control_interval_low | 173 | | wsrep_flow_control_interval_high | 173 | | wsrep_flow_control_status | OFF | | wsrep_cert_deps_distance | 75.099585 | | wsrep_apply_oooe | 0.654762 | | wsrep_apply_oool | 0.009259 | | wsrep_apply_window | 5.033069 | | wsrep_commit_oooe | 0.000000 | | wsrep_commit_oool | 0.000000 | | wsrep_commit_window | 2.513228 | | wsrep_local_state | 4 | | wsrep_local_state_comment | Synced | | wsrep_cert_index_size | 116 | | wsrep_cert_bucket_count | 210 | | wsrep_gcache_pool_size | 88288 | | wsrep_causal_reads | 0 | | wsrep_cert_interval | 0.000000 | | wsrep_open_transactions | 0 | | wsrep_open_connections | 0 | | wsrep_ist_receive_status | | | wsrep_ist_receive_seqno_start | 0 | | wsrep_ist_receive_seqno_current | 0 | | wsrep_ist_receive_seqno_end | 0 | | wsrep_incoming_addresses | 172.16.0.101:3306,172.16.0.100:3306,172.16.0.102:3306 | +----------------------------------+-------------------------------------------------------+ ``` 参考：https://www.percona.com/doc/percona-xtradb-cluster/5.7/index.html#introduction 参考书籍: MySQL王者晋级之路/张甦

### 环境准备系统: `CentOS7` 软件: `MySQL5.7` `MHA 0.56` VIP: `172.16.0.222` 服务器: `node1(Master节点)`: `172.16.0.100` `node2(SLave节点/Master备)`: `172.16.0.101` `node3(SLave节点/MHA manager)`: `172.16.0.102` &emsp; ### 初始化工作 **`a.三台机器SSH免密登陆`** **`b.三台机器的主机名和设置hosts文件解析`** **`c.两台SLAVE节点配置成MASTER节点的MySQL主从`** **`d.MySQL的MHA配置要求`** ```bash vim /etc/my.cnf ``` ```ini [mysqld] datadir=/var/lib/mysql socket=/var/lib/mysql/mysql.sock symbolic-links=0 log-error=/var/log/mysqld.log pid-file=/var/run/mysqld/mysqld.pid bind-address = 0.0.0.0 #配置 server-id = 1 #节点的值唯一 log_bin = mysql-bin #MySQL Master或者备用Master binlog_ignore_db = mysql binlog_ignore_db = infomation_schema binlog_ignore_db = performance_schema #relay_log_purge = 0 #从库需要配置,执行完relay不删除 ``` `提示:MySQL Master或者备用Master需要打开binlog,MHA要求各个数据库节点的复制过滤规则(binlog-do-db, replicate-ignore-db)都一样` **`e. 开启半同步复制`** `所有节点都执行下面命令` ```sql mysql> install plugin rpl_semi_sync_master soname 'semisync_master.so'; #安装Master半同步复制插件 mysql> install plugin rpl_semi_sync_slave soname 'semisync_slave.so'; #安装Slave半同步复制插件 mysql> set global rpl_semi_sync_slave_enabled=on; #开启Slave半同步复制 mysql> set global rpl_semi_sync_master_enabled=on; #开启Master半同步复制 mysql> stop slave io_thread; mysql> start slave io_thread; mysql> show variables like '%semi%'; #查看半同步复制参数 mysql> show plugins; #查看加载的插件 ``` **`f. MySQL配置GTID复制(可选)`** 参考:https://ynotes.cn/blog/article_detail/203 传统复制和GTID复制转换 &emsp; ### 安装MHA 下载MHA Manager & MHA Node ```bash curl http://www.mysql.gr.jp/frame/modules/bwiki/index.php\?plugin\=attach\&pcmd\=open\&file\=mha4mysql-manager-0.56-0.el6.noarch.rpm\&refer\=matsunobu -o mha4mysql-manager-0.56-0.el6.noarch.rpm curl http://www.mysql.gr.jp/frame/modules/bwiki/index.php\?plugin\=attach\&pcmd\=open\&file\=mha4mysql-node-0.56-0.el6.noarch.rpm\&refer\=matsunobu -o mha4mysql-node-0.56-0.el6.noarch.rpm ``` 安装MHA Node `所有节点` ```bash yum localinstall -y mha4mysql-node-0.56-0.el6.noarch.rpm ``` 安装MHA Manager `node3(MHA manager节点)` ```bash yum localinstall -y mha4mysql-manager-0.56-0.el6.noarch.rpm ``` &emsp; ### 配置MHA管理节点 `node3(MHA manager节点)` 创建mha相关目录 ```bash mkdir -p /usr/local/mha mkdir -p /etc/mha ``` 创建mha.conf ```bash vim /etc/mha/mha.conf ``` ```ini [server default] manager_log=/var/log/mha/manager.log master_ip_failover_script=/etc/mha/scripts/master_ip_failover master_ip_online_change_script=/etc/mha/scripts/master_ip_failover # mysql user and password user=mha_admin password=123456 #ssh user ssh_user=root # working directory on the manager manager_workdir=/usr/local/mha # working directory on MySQL servers remote_workdir=/usr/local/mha repl_user=replica repl_password=123456 [server1] hostname=172.16.0.100 [server2] hostname=172.16.0.101 [server3] hostname=172.16.0.102 ``` 创建VIP切换脚本 ```bash vim /etc/mha/scripts/master_ip_failover ``` ```perl #!/usr/bin/env perl use strict; use warnings FATAL => 'all'; use Getopt::Long; my ( $command, $ssh_user, $orig_master_host, $orig_master_ip, $orig_master_port, $new_master_host, $new_master_ip, $new_master_port ); my $vip = '172.16.0.222/24'; # Virtual IP #设置VIP地址 my $key = "1"; my $ssh_start_vip = "/sbin/ifconfig eth1:$key $vip"; #针对网卡配置，如果使用的是eth0,则eth0:$key my $ssh_stop_vip = "/sbin/ifconfig eth1:$key down"; #针对网卡配置，如果使用的是eth0,则eth0:$key $ssh_user = "root"; GetOptions( 'command=s' => \$command, 'ssh_user=s' => \$ssh_user, 'orig_master_host=s' => \$orig_master_host, 'orig_master_ip=s' => \$orig_master_ip, 'orig_master_port=i' => \$orig_master_port, 'new_master_host=s' => \$new_master_host, 'new_master_ip=s' => \$new_master_ip, 'new_master_port=i' => \$new_master_port, ); exit &main(); sub main { print "\n\nIN SCRIPT TEST====$ssh_stop_vip==$ssh_start_vip===\n\n"; if ( $command eq "stop" || $command eq "stopssh" ) { # $orig_master_host, $orig_master_ip, $orig_master_port are passed. # If you manage master ip address at global catalog database, # invalidate orig_master_ip here. my $exit_code = 1; eval { print "Disabling the VIP on old master: $orig_master_host \n"; &stop_vip(); $exit_code = 0; }; if ($@) { warn "Got Error: $@\n"; exit $exit_code; } exit $exit_code; } elsif ( $command eq "start" ) { # all arguments are passed. # If you manage master ip address at global catalog database, # activate new_master_ip here. # You can also grant write access (create user, set read_only=0, etc) here. my $exit_code = 10; eval { print "Enabling the VIP - $vip on the new master - $new_master_host \n"; &start_vip(); $exit_code = 0; }; if ($@) { warn $@; exit $exit_code; } exit $exit_code; } elsif ( $command eq "status" ) { print "Checking the Status of the script.. OK \n"; `ssh $ssh_user\@cluster1 \" $ssh_start_vip \"`; exit 0; } else { &usage(); exit 1; } } # A simple system call that enable the VIP on the new master sub start_vip() { `ssh $ssh_user\@$new_master_host \" $ssh_start_vip \"`; } # A simple system call that disable the VIP on the old_master sub stop_vip() { `ssh $ssh_user\@$orig_master_host \" $ssh_stop_vip \"`; } sub usage { print "Usage: master_ip_failover --command=start|stop|stopssh|status --orig_master_host=host --orig_master_ip=ip --orig_master_port=port --new_master_host=host --new_master_ip=ip --new_master_port=port\n"; } ``` 创建复制用户(replica) `node1-2节点(Master节点/Master备)` ```sql mysql> GRANT REPLICATION SLAVE ON *.* TO 'replica'@'172.16.0.%' IDENTIFIED BY '123456'; mysql> FLUSH PRIVILEGES; ``` 创建管理用户(mha) `所有节点` ```sql mysql> GRANT ALL ON *.* TO 'mha'@'172.16.0.%' IDENTIFIED BY '123456'; mysql> FLUSH PRIVILEGES; ``` &emsp; ### Master配置VIP(第一次手动添加) `node1节点` ```bash /sbin/ifconfig eth1:1 172.16.0.222 ``` &emsp; ### 测试节点间的SSH登录 `node3(MHA manager节点)` ```bash masterha_check_ssh --conf=/etc/mha/mha.conf ``` ``` Mon Aug 5 03:17:00 2019 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping. Mon Aug 5 03:17:00 2019 - [info] Reading application default configuration from /etc/mha/mha.conf.. Mon Aug 5 03:17:00 2019 - [info] Reading server configuration from /etc/mha/mha.conf.. Mon Aug 5 03:17:00 2019 - [info] Starting SSH connection tests.. Mon Aug 5 03:17:01 2019 - [debug] Mon Aug 5 03:17:00 2019 - [debug] Connecting via SSH from root@172.16.0.100(172.16.0.100:22) to root@172.16.0.101(172.16.0.101:22).. Mon Aug 5 03:17:00 2019 - [debug] ok. Mon Aug 5 03:17:00 2019 - [debug] Connecting via SSH from root@172.16.0.100(172.16.0.100:22) to root@172.16.0.102(172.16.0.102:22).. Mon Aug 5 03:17:01 2019 - [debug] ok. Mon Aug 5 03:17:01 2019 - [debug] Mon Aug 5 03:17:00 2019 - [debug] Connecting via SSH from root@172.16.0.101(172.16.0.101:22) to root@172.16.0.100(172.16.0.100:22).. Mon Aug 5 03:17:01 2019 - [debug] ok. Mon Aug 5 03:17:01 2019 - [debug] Connecting via SSH from root@172.16.0.101(172.16.0.101:22) to root@172.16.0.102(172.16.0.102:22).. Mon Aug 5 03:17:01 2019 - [debug] ok. Mon Aug 5 03:17:02 2019 - [debug] Mon Aug 5 03:17:01 2019 - [debug] Connecting via SSH from root@172.16.0.102(172.16.0.102:22) to root@172.16.0.100(172.16.0.100:22).. Mon Aug 5 03:17:01 2019 - [debug] ok. Mon Aug 5 03:17:01 2019 - [debug] Connecting via SSH from root@172.16.0.102(172.16.0.102:22) to root@172.16.0.101(172.16.0.101:22).. Mon Aug 5 03:17:02 2019 - [debug] ok. Mon Aug 5 03:17:02 2019 - [info] All SSH connection tests passed successfully. ``` `上面显示ssh免密登陆配置成功` ### 检查mha配置 `node3(MHA manager节点)` ```bash masterha_check_repl --conf=/etc/mha/mha.conf ``` ``` Mon Aug 5 03:39:17 2019 - [warning] Global configuration file /etc/masterha_default.cnf not found. Skipping. Mon Aug 5 03:39:17 2019 - [info] Reading application default configuration from /etc/mha/mha.conf.. Mon Aug 5 03:39:17 2019 - [info] Reading server configuration from /etc/mha/mha.conf.. Mon Aug 5 03:39:17 2019 - [info] MHA::MasterMonitor version 0.56. Mon Aug 5 03:39:18 2019 - [info] GTID failover mode = 0 Mon Aug 5 03:39:18 2019 - [info] Dead Servers: Mon Aug 5 03:39:18 2019 - [info] Alive Servers: Mon Aug 5 03:39:18 2019 - [info] 172.16.0.100(172.16.0.100:3306) Mon Aug 5 03:39:18 2019 - [info] 172.16.0.101(172.16.0.101:3306) Mon Aug 5 03:39:18 2019 - [info] 172.16.0.102(172.16.0.102:3306) Mon Aug 5 03:39:18 2019 - [info] Alive Slaves: Mon Aug 5 03:39:18 2019 - [info] 172.16.0.101(172.16.0.101:3306) Version=5.7.27-log (oldest major version between slaves) log-bin:enabled Mon Aug 5 03:39:18 2019 - [info] Replicating from 172.16.0.100(172.16.0.100:3306) Mon Aug 5 03:39:18 2019 - [info] 172.16.0.102(172.16.0.102:3306) Version=5.7.27-log (oldest major version between slaves) log-bin:enabled Mon Aug 5 03:39:18 2019 - [info] Replicating from 172.16.0.100(172.16.0.100:3306) Mon Aug 5 03:39:18 2019 - [info] Current Alive Master: 172.16.0.100(172.16.0.100:3306) Mon Aug 5 03:39:18 2019 - [info] Checking slave configurations.. Mon Aug 5 03:39:18 2019 - [info] read_only=1 is not set on slave 172.16.0.101(172.16.0.101:3306). Mon Aug 5 03:39:18 2019 - [warning] relay_log_purge=0 is not set on slave 172.16.0.101(172.16.0.101:3306). Mon Aug 5 03:39:18 2019 - [info] read_only=1 is not set on slave 172.16.0.102(172.16.0.102:3306). Mon Aug 5 03:39:18 2019 - [warning] relay_log_purge=0 is not set on slave 172.16.0.102(172.16.0.102:3306). Mon Aug 5 03:39:18 2019 - [info] Checking replication filtering settings.. Mon Aug 5 03:39:18 2019 - [info] binlog_do_db= , binlog_ignore_db= infomation_schema,mysql,performance_schema Mon Aug 5 03:39:18 2019 - [info] Replication filtering check ok. Mon Aug 5 03:39:18 2019 - [info] GTID (with auto-pos) is not supported Mon Aug 5 03:39:18 2019 - [info] Starting SSH connection tests.. Mon Aug 5 03:39:20 2019 - [info] All SSH connection tests passed successfully. Mon Aug 5 03:39:20 2019 - [info] Checking MHA Node version.. Mon Aug 5 03:39:21 2019 - [info] Version check ok. Mon Aug 5 03:39:21 2019 - [info] Checking SSH publickey authentication settings on the current master.. Mon Aug 5 03:39:21 2019 - [info] HealthCheck: SSH to 172.16.0.100 is reachable. Mon Aug 5 03:39:21 2019 - [info] Master MHA Node version is 0.56. Mon Aug 5 03:39:21 2019 - [info] Checking recovery script configurations on 172.16.0.100(172.16.0.100:3306).. Mon Aug 5 03:39:21 2019 - [info] Executing command: save_binary_logs --command=test --start_pos=4 --binlog_dir=/var/lib/mysql,/var/log/mysql --output_file=/usr/local/mha/save_binary_logs_test --manager_version=0.56 --start_file=mysql-bin.000008 Mon Aug 5 03:39:21 2019 - [info] Connecting to root@172.16.0.100(172.16.0.100:22).. Creating /usr/local/mha if not exists.. ok. Checking output directory is accessible or not.. ok. Binlog found at /var/lib/mysql, up to mysql-bin.000008 Mon Aug 5 03:39:22 2019 - [info] Binlog setting check done. Mon Aug 5 03:39:22 2019 - [info] Checking SSH publickey authentication and checking recovery script configurations on all alive slave servers.. Mon Aug 5 03:39:22 2019 - [info] Executing command : apply_diff_relay_logs --command=test --slave_user='mha_admin' --slave_host=172.16.0.101 --slave_ip=172.16.0.101 --slave_port=3306 --workdir=/usr/local/mha --target_version=5.7.27-log --manager_version=0.56 --relay_log_info=/var/lib/mysql/relay-log.info --relay_dir=/var/lib/mysql/ --slave_pass=xxx Mon Aug 5 03:39:22 2019 - [info] Connecting to root@172.16.0.101(172.16.0.101:22).. Checking slave recovery environment settings.. Opening /var/lib/mysql/relay-log.info ... ok. Relay log found at /var/lib/mysql, up to db2-relay-bin.000002 Temporary relay log file is /var/lib/mysql/db2-relay-bin.000002 Testing mysql connection and privileges..mysql: [Warning] Using a password on the command line interface can be insecure. done. Testing mysqlbinlog output.. done. Cleaning up test file(s).. done. Mon Aug 5 03:39:22 2019 - [info] Executing command : apply_diff_relay_logs --command=test --slave_user='mha_admin' --slave_host=172.16.0.102 --slave_ip=172.16.0.102 --slave_port=3306 --workdir=/usr/local/mha --target_version=5.7.27-log --manager_version=0.56 --relay_log_info=/var/lib/mysql/relay-log.info --relay_dir=/var/lib/mysql/ --slave_pass=xxx Mon Aug 5 03:39:22 2019 - [info] Connecting to root@172.16.0.102(172.16.0.102:22).. Checking slave recovery environment settings.. Opening /var/lib/mysql/relay-log.info ... ok. Relay log found at /var/lib/mysql, up to db3-relay-bin.000002 Temporary relay log file is /var/lib/mysql/db3-relay-bin.000002 Testing mysql connection and privileges..mysql: [Warning] Using a password on the command line interface can be insecure. done. Testing mysqlbinlog output.. done. Cleaning up test file(s).. done. Mon Aug 5 03:39:22 2019 - [info] Slaves settings check done. Mon Aug 5 03:39:22 2019 - [info] 172.16.0.100(172.16.0.100:3306) (current master) +--172.16.0.101(172.16.0.101:3306) +--172.16.0.102(172.16.0.102:3306) Mon Aug 5 03:39:22 2019 - [info] Checking replication health on 172.16.0.101.. Mon Aug 5 03:39:22 2019 - [info] ok. Mon Aug 5 03:39:22 2019 - [info] Checking replication health on 172.16.0.102.. Mon Aug 5 03:39:22 2019 - [info] ok. Mon Aug 5 03:39:22 2019 - [info] Checking master_ip_failover_script status: Mon Aug 5 03:39:22 2019 - [info] /etc/mha/scripts/master_ip_failover --command=status --ssh_user=root --orig_master_host=172.16.0.100 --orig_master_ip=172.16.0.100 --orig_master_port=3306 IN SCRIPT TEST====/sbin/ifconfig eth1:1 down==/sbin/ifconfig eth1:1 172.16.0.222/24=== Checking the Status of the script.. OK ssh: Could not resolve hostname cluster1: Name or service not known Mon Aug 5 03:39:22 2019 - [info] OK. Mon Aug 5 03:39:22 2019 - [warning] shutdown_script is not defined. Mon Aug 5 03:39:22 2019 - [info] Got exit code 0 (Not master dead). MySQL Replication Health is OK. ``` `MySQL Replication Health is OK.显示配置正确!` ### 启动MHA Manager ```bash nohup masterha_manager --conf=/etc/mha/mha.conf & ``` ### 验证是否启动成功 ```bash masterha_check_status --conf=/etc/mha/mha.conf ``` ``` mha (pid:27057) is running(0:PING_OK), master:172.16.0.100 ``` ### 停止MHA Manager(`不执行`) ```bash masterha_stop --conf=/etc/mha/mha.conf ``` &emsp; ### 测试 #### 测试故障转移 **停止node1节点的MySQL** ```bash systemctl stop mysqld ``` **node2查看VIP** ```bash ip a ``` ``` ... 3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc pfifo_fast state UP group default qlen 1000 link/ether 5a:00:02:32:0f:46 brd ff:ff:ff:ff:ff:ff inet 172.16.0.101/16 brd 172.16.255.255 scope global eth1 valid_lft forever preferred_lft forever inet 172.16.0.222/24 brd 172.16.0.255 scope global eth1:1 valid_lft forever preferred_lft forever inet6 fe80::5800:2ff:fe32:f46/64 scope link valid_lft forever preferred_lft forever ``` `通过上面的信息可以看到VIP已经转移到node2` **查看是否为slave** ```sql mysql> show slave status\G #没有输出，说明node2已经为非slave节点 Empty set (0.00 sec) ``` **查看node3的slave信息** ```sql mysql> show slave status\G ``` ``` *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 172.16.0.101 Master_User: replica Master_Port: 3306 Connect_Retry: 60 ... ``` `通过上面的信息可以确定故障转移成功，master已经变成node2节点，node3节点已经同步到新的slave` &emsp; #### 旧master恢复的操作 `node1数据库恢复之后,首先设置为node2的slave，之后启动MHA Manager` 启动MySQL `node1` ```bash systemctl start mysqld ``` 配置node1为node2的slave _MHA节点查看同步信息_ `node3` ```bash grep "CHANGE MASTER TO" /var/log/mha/manager.log |tail -1 ``` ```bash Mon Aug 5 03:32:28 2019 - [info] All other slaves should start replication from here. Statement should be: CHANGE MASTER TO MASTER_HOST='172.16.0.101', MASTER_PORT=3306, MASTER_LOG_FILE='mysql-bin.000008', MASTER_LOG_POS=154, MASTER_USER='replica', MASTER_PASSWORD='xxx'; ``` _node1配置同步node2_ `node1` ```sql mysql> CHANGE MASTER TO MASTER_HOST='172.16.0.101', MASTER_PORT=3306, MASTER_LOG_FILE='mysql-bin.000008', MASTER_LOG_POS=154, MASTER_USER='replica', MASTER_PASSWORD='123456'; mysql> start slave; #启动slave mysql> show slave status\G ``` ``` *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 172.16.0.101 Master_User: replica Master_Port: 3306 Connect_Retry: 60 ... ``` 把 Master切回到原理的node1 `node3` ```bash masterha_master_switch --conf=/etc/mha/mha.conf --master_state=alive --new_master_host=172.16.0.100 --orig_master_is_new_slave ``` 检查mha配置 `node3` ```bash masterha_check_repl --conf=/etc/mha/mha.conf ``` ``` ... MySQL Replication Health is OK. ``` 启动MHA Manager `node3` ```bash nohup masterha_manager --conf=/etc/mha/mha.conf & ``` 参考: http://www.fblinux.com/?p=1018 https://joelhy.github.io/2015/02/06/mysql-mha/