MYSQL MGR安装

#############MGR 必选参数#######################

server_id=1

gtid_mode=ON

enforce_gtid_consistency=ON

master_info_repository=TABLE

relay_log_info_repository=TABLE

log_slave_updates=ON

log_bin=binlog

binlog_format=ROW

binlog_checksum=NONE

transaction_write_set_extraction=XXHASH64

loose-group_replication_group_name="aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa"

loose-group_replication_start_on_boot=off

loose-group_replication_local_address= "10.10.203.147:24901"

loose-group_replication_group_seeds= "10.10.203.145:24901,10.10.203.146:24901,10.10.203.147:24901"

loose-group_replication_bootstrap_group= off

####  for performance_schema

performance_schema =on

performance_schema_consumer_events_stages_current=on

performance_schema_consumer_events_stages_history=on

performance_schema_consumer_events_stages_history_long=off

performance_schema_consumer_statements_digest=on

performance_schema_consumer_events_statements_current=on

performance_schema_consumer_events_statements_history=on

performance_schema_consumer_events_statements_history_long=off

performance_schema_consumer_events_waits_current=on

performance_schema_consumer_events_waits_history=on

performance_schema_consumer_events_waits_history_long=off

performance_schema_consumer_global_instrumentation=on

performance_schema_consumer_thread_instrumentation=on

#####MYSQL CMD#############################

SET SQL_LOG_BIN=0;

CREATE USER rpl_user@'%' IDENTIFIED BY '123456';

GRANT REPLICATION SLAVE ON *.* TO rpl_user@'%';

FLUSH PRIVILEGES;

SET SQL_LOG_BIN=1;

CHANGE MASTER TO MASTER_USER='rpl_user', MASTER_PASSWORD='123456' FOR CHANNEL 'group_replication_recovery';

INSTALL PLUGIN group_replication SONAME 'group_replication.so';  --第一次安装了之后就可以不用执行了

SET GLOBAL group_replication_bootstrap_group=ON;  --只有作为SEED节点的机器才使用，否则所有的节点都默认为SEED节点了，坑

set global group_replication_allow_local_disjoint_gtids_join=1; --只在slave节点执行

START GROUP_REPLICATION;

SET GLOBAL group_replication_bootstrap_group=OFF;

SELECT * FROM performance_schema.replication_group_members;

select * from performance_schema.replication_connection_status\G

CREATE DATABASE test;

use test;

CREATE TABLE t1 (c1 INT PRIMARY KEY, c2 TEXT NOT NULL);

INSERT INTO t1 VALUES (1, 'Luis');

SHOW BINLOG EVENTS;

show global status like '%group_replication_primary%';

MGR的限制

仅支持InnoDB表，并且每张表一定要有一个主键，用于做write set的冲突检测；

必须打开GTID特性，二进制日志格式必须设置为ROW，用于选主与write set

COMMIT可能会导致失败，类似于快照事务隔离级别的失败场景

目前一个MGR集群最多支持9个节点

不支持外键于save point特性，无法做全局间的约束检测与部分部分回滚

二进制日志不支持binlog event checksum

错误解决：

##2018-07-10T06:17:00.054571Z 41 [ERROR] Slave I/O for channel 'group_replication_recovery': error connecting to master 'rpl_user@mha-slave1:3307' - retry-time: 60  retries: 1, Error_code: 2005

/etc/hosts中的IP解析出问题，修改后正常

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/27799221/viewspace-2215656/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/27799221/viewspace-2215656/