MySQL5.7.24 開啓Gtid+MGR組複製
MySQL5.7.24 Gtid+MGRmysql
一.服務器環境說明：
因爲服務器的資源有限，本次測試採用的是1臺機器測試：
一臺硬件物理機器：16G內存，一塊120G的ssd盤 系統是CentOS6.9 x86_64最小化安裝
系統內核：linux
[root@localhost ~]# uname -a Linux localhost.localdomain 2.6.32-696.el6.x86_64 #1 SMP Tue Mar 21 19:29:05 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
關閉selinux,關閉iptables sql
服務部署規劃說明：
物理機器上以不一樣的mysql配置文件my.cnf來初始化mysql ，而且以三個不一樣的端口和添加MGR的配置參數來開啓3個MySQL實例
mysql的版本爲mysql5.7.24 二進制安裝bootstrap
二.配置步驟以下：
第一，物理機器上開啓三個mysql5.7.24 實例：服務器
初始化mysql網絡
下載軟件： wget https://cdn.mysql.com//Downloads/MySQL-5.7/mysql-5.7.24-linux-glibc2.12-x86_64.tar.gz tar xf mysql-5.7.24-linux-glibc2.12-x86_64.tar.gz -C /usr/local mv /usr/local/mysql-5.7.24-linux-glibc2.12 /usr/local/mysql useradd mysql -s /sbin/nologin -M chown -R mysql.mysql /usr/local/mysql yum install numactl -y
建立MySQL初始化數據存放路徑：app
mkdir -p /home/mysql3306/{data,binlog,logs}
mkdir -p /home/mysql3307/{data,binlog,logs}
mkdir -p /home/mysql3308/{data,binlog,logs}
chown -R mysql.mysql /home/{mysql3306,mysql3307,mysql3308}
準備初始化mysql的my.cnf的配置文件：less
[root@localhost ~]# ll /opt/my330* -rw-r--r-- 1 root root 9227 12月 26 18:11 /opt/my3306.cnf -rw-r--r-- 1 root root 9294 12月 27 18:09 /opt/my3307.cnf -rw-r--r-- 1 root root 9294 12月 27 18:10 /opt/my3308.cnf [root@localhost ~]#
mysql5.7.17開始引入mysql的MGR特性簡單介紹MySQL5.7開啓Gtid+MGR特性的參數說明：dom
這次演示環境以mysql3306實例做爲第一個MGR的啓動實例，
MySQL5.7開啓Gtid+MGR特性my3306cnf配置文件必須開啓參數以下：socket
server_id =1 socket =/tmp/mysql3306.sock ##:for binlog binlog_format =row log_bin =/home/mysql3306/binlog/mysql-bin binlog_checksum=NONE log_slave_updates=ON ##:for gtid #gtid_executed_compression_period =1000 gtid_mode =on enforce_gtid_consistency =on ####for group_replication master_info_repository=TABLE relay_log_info_repository=TABLE binlog_checksum=NONE log_slave_updates=ON transaction_write_set_extraction=XXHASH64 loose-group_replication_group_name="3db33b36-0e51-409f-a61d-c99756e90155" ##格式要和server-uuid一致，可是不能和機器上mysql實例的uuid重複 loose-group_replication_start_on_boot=off ##禁止在重啓mysql服務時開啓組複製 loose-group_replication_local_address= "192.168.1.233:33061" #####端口隨便指定一個，可是不要和機上其餘服務的端口衝突就行，告訴插件本機使用網絡地址192.168.1.233和端口33061與組中的其餘成員進行內部通訊。 loose-group_replication_group_seeds= "192.168.1.233:33061,192.168.1.233:33071,192.168.1.233:33081" loose-group_replication_ip_whitelist="192.168.1.233/24" ##此處必須設置ip白名單，不然在開啓MGR時，會報錯。 loose-group_replication_bootstrap_group= off ## ##loose-group_replication_single_primary_mode=off ##在配置MGR的mutl-master模式時開啓 ##loose-group_replication_enforce_update_everywhere_checks=on ##在配置MGR的mutl-master模式時開啓 my3307.cnf配置文件和my3306.cnf中不一樣的是把loose-group_replication_local_address修改成 "192.168.1.233:33071" my3308.cnf配置文件和my3306.cnf中不一樣的是把loose-group_replication_local_address修改成 "192.168.1.233:33081"
啓動mysql實例：
/usr/local/mysql/bin/mysqld --defaults-file=/opt/my3306.cnf --initialize /usr/local/mysql/bin/mysqld --defaults-file=/opt/my3307.cnf --initialize /usr/local/mysql/bin/mysqld --defaults-file=/opt/my3308.cnf --initialize
爲了在配置MGR的過程當中防止出錯，在啓動mysql以前把auto.cnf 修改成3個不相同的uuid
[root@localhost ~]# cat /home/mysql3306/data/auto.cnf [auto] server-uuid=1ec3ac79-08ed-11e9-8da8-bcaec502b368 [root@localhost ~]# cat /home/mysql3307/data/auto.cnf [auto] server-uuid=288e7bbe-08f3-11e9-a605-bcaec502b311 [root@localhost ~]# cat /home/mysql3308/data/auto.cnf [auto] server-uuid=883c9421-08f5-11e9-8d47-bcaec502b333
啓動mysql服務：
/usr/local/mysql/bin/mysqld --defaults-file=/opt/my3306.cnf & /usr/local/mysql/bin/mysqld --defaults-file=/opt/my3307.cnf & /usr/local/mysql/bin/mysqld --defaults-file=/opt/my3308.cnf &
第二.開啓MGR配置過程(默認是single-master模式)：
mysql3306做爲第一個節點開啓MGR:
SET SQL_LOG_BIN=0; CREATE USER rpl_user@'%'; GRANT REPLICATION SLAVE ON *.* TO rpl_user@'%' IDENTIFIED BY 'Zykjwujianwei'; FLUSH PRIVILEGES; SET SQL_LOG_BIN=1; CHANGE MASTER TO MASTER_USER='rpl_user', MASTER_PASSWORD='Zykjwujianwei' FOR CHANNEL 'group_replication_recovery'; install PLUGIN group_replication SONAME 'group_replication.so'; set global group_replication_bootstrap_group=ON; start group_replication; select * from performance_schema.replication_group_members;
注意：只有在第一個開啓MGR的mysql3306節點上才執行這個參數set global group_replication_bootstrap_group=ON;
root@localhost [(none)]>select * from performance_schema.replication_group_members; +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | group_replication_applier | 1ec3ac79-08ed-11e9-8da8-bcaec502b368 | localhost.localdomain | 3306 | ONLINE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ 3 rows in set (0.00 sec) root@localhost [(none)]>
查看當前的leader：
select *from performance_schema.replication_group_members where member_id =(select variable_value from performance_schema.global_status WHERE VARIABLE_NAME= 'group_replication_primary_member'); select variable_value from performance_schema.global_status WHERE VARIABLE_NAME= 'group_replication_primary_member'；
節點mysql3307操做：
SET SQL_LOG_BIN=0; CREATE USER rpl_user@'%'; GRANT REPLICATION SLAVE ON *.* TO rpl_user@'%' IDENTIFIED BY 'Zykjwujianwei'; FLUSH PRIVILEGES; SET SQL_LOG_BIN=1; CHANGE MASTER TO MASTER_USER='rpl_user', MASTER_PASSWORD='Zykjwujianwei' FOR CHANNEL 'group_replication_recovery'; install PLUGIN group_replication SONAME 'group_replication.so'; start group_replication; select * from performance_schema.replication_group_members; root@localhost [(none)]>select * from performance_schema.replication_group_members; +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | group_replication_applier | 1ec3ac79-08ed-11e9-8da8-bcaec502b368 | localhost.localdomain | 3306 | ONLINE | | group_replication_applier | 288e7bbe-08f3-11e9-a605-bcaec502b311 | localhost.localdomain | 3307 | ONLINE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+
節點mysql3308操做：
SET SQL_LOG_BIN=0; CREATE USER rpl_user@'%'; GRANT REPLICATION SLAVE ON *.* TO rpl_user@'%' IDENTIFIED BY 'Zykjwujianwei'; FLUSH PRIVILEGES; SET SQL_LOG_BIN=1; CHANGE MASTER TO MASTER_USER='rpl_user', MASTER_PASSWORD='Zykjwujianwei' FOR CHANNEL 'group_replication_recovery'; install PLUGIN group_replication SONAME 'group_replication.so'; start group_replication; select * from performance_schema.replication_group_members; root@localhost [(none)]>select * from performance_schema.replication_group_members; +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | group_replication_applier | 1ec3ac79-08ed-11e9-8da8-bcaec502b368 | localhost.localdomain | 3306 | ONLINE | | group_replication_applier | 288e7bbe-08f3-11e9-a605-bcaec502b311 | localhost.localdomain | 3307 | ONLINE | | group_replication_applier | 883c9421-08f5-11e9-8d47-bcaec502b333 | localhost.localdomain | 3308 | ONLINE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ 3 rows in set (0.00 sec)
第三.測試效果：
root@localhost [(none)]>select @@port;create database test01;show databases; +--------+ | @@port | +--------+ | 3306 | +--------+ 1 row in set (0.00 sec) Query OK, 1 row affected (0.00 sec) +--------------------+ | Database | +--------------------+ | information_schema | | mysql | | performance_schema | | sys | | test01 | +--------------------+ 5 rows in set (0.00 sec) root@localhost [(none)]>select @@port;create database test02;show databases; +--------+ | @@port | +--------+ | 3307 | +--------+ 1 row in set (0.00 sec) ERROR 1290 (HY000): The MySQL server is running with the --super-read-only option so it cannot execute this statement +--------------------+ | Database | +--------------------+ | information_schema | | mysql | | performance_schema | | sys | | test01 | +--------------------+ 5 rows in set (0.00 sec) root@localhost [(none)]> root@localhost [(none)]>select @@port;create database test02;show databases; +--------+ | @@port | +--------+ | 3308 | +--------+ 1 row in set (0.00 sec) ERROR 1290 (HY000): The MySQL server is running with the --super-read-only option so it cannot execute this statement +--------------------+ | Database | +--------------------+ | information_schema | | mysql | | performance_schema | | sys | | test01 | +--------------------+ 5 rows in set (0.00 sec) root@localhost [(none)]>
測試說明：
發現mysql3306上建立的test01同步到了其餘的2個mysql。然而在MGR組中mysql3307 和mysql3308是隻讀，不能寫入的。
關掉mysql3306服務會發生什麼？？ 直接會從mysql3307和mysql3308實例中選擇一個做爲mysql主庫，另一個做爲新主庫的slave庫
關掉mysql3306服務一段時間後，有從新開啓mysql3306，而且start group_replication;從新加入到MGR組會發生什麼？？
此時mysql3306不會切換爲master庫，而是成爲了宕機後新主庫的slave庫。
mysql的mult-master多主寫入的模式：
MGR默認的模式也就是我們演示的single-master模式（單主寫入模式）。那如何開啓mysql的mult-master多主寫入的模式呢？？下面簡單介紹下：
首先要先關閉只讀的從庫mysql3307和mysql3308實例最後關閉mysql3306
其次在各自的my.cnf配置文件中開啓以下參數：
loose-group_replication_single_primary_mode=off ##在配置MGR的mutl-master模式時開啓 loose-group_replication_enforce_update_everywhere_checks=on ##在配置MGR的mutl-master模式時開啓
最後啓動mysql實例，按照配置MGR-single-master的順序步驟來進行配置MGR的multi-master模式
特別提示：
multi-master模式下，3個節點mysql實例都是可讀寫的。可是在第一個啓動的mysql實例上須要執行set global group_replication_bootstrap_group=ON;（注意：後面的節點不須要執行這個sql）
mysql5.7 開啓MGR multi-master模式後，查看實例的狀態發現不存在所謂的leader
(root@localhost:mysql.sock)[(none)]>select * from performance_schema.replication_group_members; +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | CHANNEL_NAME | MEMBER_ID | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ | group_replication_applier | 862addac-10c5-11e9-9af5-bcaec502b317 | localhost.localdomain | 3307 | ONLINE | | group_replication_applier | a36b8f24-10c3-11e9-bf76-bcaec502b318 | localhost.localdomain | 3308 | ONLINE | | group_replication_applier | da3809c4-10bc-11e9-bb9e-bcaec502b368 | localhost.localdomain | 3306 | ONLINE | +---------------------------+--------------------------------------+-----------------------+-------------+--------------+ 3 rows in set (0.00 sec) (root@localhost:mysql.sock)[(none)]>select *from performance_schema.replication_group_members where member_id =(select variable_value from performance_schema.global_status WHERE VARIABLE_NAME= 'group_replication_primary_member'); Empty set (0.00 sec) (root@localhost:mysql.sock)[(none)]>select variable_value from performance_schema.global_status WHERE VARIABLE_NAME= 'group_replication_primary_member'; +----------------+ | variable_value | +----------------+ | | +----------------+ 1 row in set (0.00 sec)
my.cnf配置文件演示：
mysql3307完整的配置文件以下：
[root@localhost ~]# cat /opt/my3307.cnf
[client]
port = 3307
[mysql]
auto-rehash
prompt="\u@\h [\d]>"
#pager="less -i -n -S"
#tee=/opt/mysql/query.log
[mysqld]
####: for global
user =mysql
basedir =/usr/local/mysql/
datadir =/home/mysql3307/data
server_id =2
port =3307
character_set_server =utf8
explicit_defaults_for_timestamp =off
log_timestamps =system
socket =/tmp/mysql3307.sock
read_only =0
skip_name_resolve =1
auto_increment_increment =1
auto_increment_offset =1
lower_case_table_names =1
secure_file_priv =
open_files_limit =65536
max_connections =1000
thread_cache_size =64
table_open_cache =81920
table_definition_cache =4096
table_open_cache_instances =64
max_prepared_stmt_count =1048576
####: for binlog
binlog_format =row
log_bin =/home/mysql3307/binlog/mysql-bin
binlog_rows_query_log_events =on
#log_slave_updates =on
expire_logs_days =7
binlog_cache_size =65536
#binlog_checksum =none
sync_binlog =1
slave-preserve-commit-order =ON
####: for error-log
log_error =/home/mysql3307/logs/error.log
general_log =off
general_log_file =/home/mysql3307/logs/general.log
####: for slow query log
slow_query_log =on
slow_query_log_file =/home/mysql3307/logs/slow.log
#log_queries_not_using_indexes =on
long_query_time =1.000000
####: for gtid
#gtid_executed_compression_period =1000
gtid_mode =on
enforce_gtid_consistency =on
####: for replication
skip_slave_start =1
#master_info_repository =table
#relay_log_info_repository =table
slave_parallel_type =logical_clock
slave_parallel_workers =4
#rpl_semi_sync_master_enabled =1
#rpl_semi_sync_slave_enabled =1
#rpl_semi_sync_master_timeout =1000
#plugin_load_add =semisync_master.so
#plugin_load_add =semisync_slave.so
binlog_group_commit_sync_delay =100
binlog_group_commit_sync_no_delay_count = 10
####for group_replication
master_info_repository=TABLE
relay_log_info_repository=TABLE
binlog_checksum=NONE
log_slave_updates=ON
transaction_write_set_extraction=XXHASH64
loose-group_replication_group_name="3db33b36-0e51-409f-a61d-c99756e90155"
loose-group_replication_start_on_boot=off
loose-group_replication_local_address= "192.168.1.233:33071"
loose-group_replication_group_seeds= "192.168.1.233:33061,192.168.1.233:33071,192.168.1.233:33081"
loose-group_replication_ip_whitelist="192.168.1.233/24"
##loose-group_replication_single_primary_mode=off ##在配置MGR的mutl-master模式時開啓
##loose-group_replication_enforce_update_everywhere_checks=on ##在配置MGR的mutl-master模式時開啓
####: for innodb
default_storage_engine =innodb
default_tmp_storage_engine =innodb
innodb_data_file_path =ibdata1:1024M:autoextend
innodb_temp_data_file_path =ibtmp1:12M:autoextend
innodb_buffer_pool_filename =ib_buffer_pool
innodb_log_group_home_dir =/home/mysql3307/data
innodb_log_files_in_group =3
innodb_log_file_size =1024M
innodb_file_per_table =on
innodb_online_alter_log_max_size =128M
innodb_open_files =65535
innodb_page_size =16k
innodb_thread_concurrency =0
innodb_read_io_threads =4
innodb_write_io_threads =4
innodb_purge_threads =4
innodb_page_cleaners =4
# 4(刷新lru髒頁)
innodb_print_all_deadlocks =on
innodb_deadlock_detect =on
innodb_lock_wait_timeout =20
innodb_spin_wait_delay =128
innodb_autoinc_lock_mode =2
innodb_io_capacity =200
innodb_io_capacity_max =2000
#--------Persistent Optimizer Statistics
innodb_stats_auto_recalc =on
innodb_stats_persistent =on
innodb_stats_persistent_sample_pages =20
innodb_adaptive_hash_index =on
innodb_change_buffering =all
innodb_change_buffer_max_size =25
innodb_flush_neighbors =1
#innodb_flush_method =
innodb_doublewrite =on
innodb_log_buffer_size =128M
innodb_flush_log_at_timeout =1
innodb_flush_log_at_trx_commit =1
innodb_buffer_pool_size =4096M
innodb_buffer_pool_instances =4
autocommit =1
#--------innodb scan resistant
innodb_old_blocks_pct =37
innodb_old_blocks_time =1000
#--------innodb read ahead
innodb_read_ahead_threshold =56
innodb_random_read_ahead =OFF
#--------innodb buffer pool state
innodb_buffer_pool_dump_pct =25
innodb_buffer_pool_dump_at_shutdown =ON
innodb_buffer_pool_load_at_startup =ON
到此處mysql5.7.24的Grid+MGR模式演示完畢
相關文章
- 1. mysql5.7.24 gtid雙主複製+atlas+keepalived
- 2. centos7下配置mysql5.7.24主從複製
- 3. mysql複製 開啓事務
- 4. Centos7.5安裝mysql5.7.24二進制包方式部署(主從複製)
- 5. centOS6.5 MySQL5.7.24安裝
- 6. linux安裝mysql5.7.24
- 7. Centos7安裝mysql5.7.24
- 8. Mysql組複製
- 9. 數組複製
- 10. mysql組複製
- 更多相關文章...
- • ionic 複選框 - ionic 教程
- • 啓動MySQL服務 - MySQL教程
- • PHP開發工具
- • 漫談MySQL的鎖機制
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
