mysql 8.0.18 mgr 搭建及其切换功能

yum -y install make gcc-c++ cmake bison-devel ncurses-devel readline-devel libaio-devel perl libaio wget lrzsz vim libnuma* bzip2 xz

sed -i 's/SELINUX=enforcing/SELINUX=disabled/' /etc/selinux/config
setenforce 0

/etc/init.d/iptables stop

echo "/etc/init.d/iptables stop">>/etc/rc.local

cat >> /etc/security/limits.conf  EOF
#
###custom
#
* soft nofile 20480
* hard nofile 65535
* soft nproc 20480
* hard nproc 65535
EOF

cat >> /etc/hosts "EOF"

10.10.146.28 bj-db-m1 
10.10.1.139 bj-db-m2 
10.10.173.84 bj-db-m3 

EOF

cat >>/etc/sysctl.conf "EOF"
vm.swappiness=0
#增加tcp支持的队列数
net.ipv4.tcp_max_syn_backlog = 65535
#减少断开连接时 ，资源回收
net.ipv4.tcp_max_tw_buckets = 8000
net.ipv4.tcp_tw_reuse = 1
net.ipv4.tcp_tw_recycle = 1
net.ipv4.tcp_fin_timeout = 10
#改变本地的端口范围
net.ipv4.ip_local_port_range = 1024 65535
#允许更多的连接进入队列
net.ipv4.tcp_max_syn_backlog = 4096 
#对于只在本地使用的数据库服务器
net.ipv4.tcp_fin_timeout = 30
#端口监听队列
net.core.somaxconn=65535
#接受数据的速率
net.core.netdev_max_backlog=65535
net.core.wmem_default=87380
net.core.wmem_max=16777216
net.core.rmem_default=87380
net.core.rmem_max=16777216
EOF

sysctl -p

wget https://dev.mysql.com/get/Downloads/MySQL-8.0/mysql-8.0.18-linux-glibc2.12-x86_64.tar.xz
# 解压安装包
tar -xJf mysql-8.0.18-linux-glibc2.12-x86_64.tar.xz
# 进入目录，做软连接，方便以后升级
cd /usr/local/
ln -s /opt/mysql-8.0.18-linux-glibc2.12-x86_64 mysql
# 创建用户
groupadd mysql
useradd -g mysql mysql -d /home/mysql -s /sbin/nologin

# 创建相应的目录

mkdir -p /data/mysql/mysql_3306/{logs,tmp,undolog}

# 第一台
if [ -f /etc/my.cnf ]; then
 mv /etc/my.cnf /etc/my.cnf.`date +%Y%m%d%H%m`.bak
fi 
# node1
cat >/data/mysql/mysql_3306/my_3306.cnf "EOF"
[client]
port = 3306
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock

[mysql]
prompt="\u@\h \R:\m:\s [\d]> "
no-auto-rehash

[mysqld]
user = mysql
port = 3306
admin_address = 127.0.0.1
basedir = /usr/local/mysql
datadir = /data/mysql/mysql_3306/data
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock
pid-file = mysql_3306.pid
character-set-server = utf8mb4
skip_name_resolve = 1

#replicate-wild-ignore-table=mysql.%
#replicate-wild-ignore-table=test.%
#replicate-wild-ignore-table=information_schema.%

# Two-Master configure
#server-1 
#auto-increment-offset = 1
#auto-increment-increment = 2 

#server-2    
#auto-increment-offset = 2
#auto-increment-increment = 2


# semi sync replication settings #
#plugin_dir = /usr/local/mysql/lib/mysql/plugin
#plugin_load = "validate_password.so;rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"
plugin_dir = /usr/local/mysql/lib/plugin #官方版本的路径
plugin_load = "rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so" #官方版本的路径

slave_parallel_workers = 4
slave_parallel_type = LOGICAL_CLOCK
slave_preserve_commit_order = 1

open_files_limit = 65535
back_log = 1024
max_connections = 1024
max_connect_errors = 1000000
table_open_cache = 1024
table_definition_cache = 1024
table_open_cache_instances = 64
thread_stack = 512K
external-locking = FALSE
max_allowed_packet = 32M
sort_buffer_size = 4M
join_buffer_size = 4M
thread_cache_size = 1536
interactive_timeout = 600
wait_timeout = 600
tmp_table_size = 32M
max_heap_table_size = 32M
slow_query_log = 1
log_timestamps = SYSTEM
slow_query_log_file = /data/mysql/mysql_3306/logs/slow.log
log-error = /data/mysql/mysql_3306/logs/error.log
long_query_time = 0.1
log_queries_not_using_indexes =1
log_throttle_queries_not_using_indexes = 60
min_examined_row_limit = 100
log_slow_admin_statements = 1
log_slow_slave_statements = 1
server-id = 1423306
log-bin = /data/mysql/mysql_3306/logs/mysql-bin
sync_binlog = 1
binlog_cache_size = 4M
max_binlog_cache_size = 2G
max_binlog_size = 1G
binlog_expire_logs_seconds=2592000 
master_info_repository = TABLE
relay_log_info_repository = TABLE
gtid_mode = on
enforce_gtid_consistency = 1
binlog_checksum=NONE
log_slave_updates
slave-rows-search-algorithms = 'INDEX_SCAN,HASH_SCAN'
binlog_format = row
binlog_row_image=FULL

relay_log_recovery = 1
relay-log-purge = 1
key_buffer_size = 32M
read_buffer_size = 8M
read_rnd_buffer_size = 4M
bulk_insert_buffer_size = 64M
myisam_sort_buffer_size = 128M
myisam_max_sort_file_size = 10G
myisam_repair_threads = 1
lock_wait_timeout = 3600
explicit_defaults_for_timestamp = 1
innodb_thread_concurrency = 0
innodb_sync_spin_loops = 100
innodb_spin_wait_delay = 30

#transaction_isolation = REPEATABLE-READ
transaction_isolation = READ-COMMITTED
#innodb_additional_mem_pool_size = 16M
innodb_buffer_pool_size = 2867M
innodb_buffer_pool_instances = 4
innodb_buffer_pool_load_at_startup = 1
innodb_buffer_pool_dump_at_shutdown = 1
innodb_data_file_path = ibdata1:1G:autoextend
innodb_flush_log_at_trx_commit = 1
innodb_log_buffer_size = 32M
innodb_log_file_size = 2G
innodb_log_files_in_group = 3
innodb_max_undo_log_size = 4G
innodb_undo_directory = /data/mysql/mysql_3306/undolog

# 根据您的服务器IOPS能力适当调整
# 一般配普通SSD盘的话，可以调整到 10000 - 20000
# 配置高端PCIe SSD卡的话，则可以调整的更高，比如 50000 - 80000
innodb_io_capacity = 4000
innodb_io_capacity_max = 8000
innodb_flush_sync = 0
innodb_flush_neighbors = 0
innodb_write_io_threads = 8
innodb_read_io_threads = 8
innodb_purge_threads = 4
innodb_page_cleaners = 4
innodb_open_files = 65535
innodb_max_dirty_pages_pct = 50
innodb_flush_method = O_DIRECT
innodb_lru_scan_depth = 4000
innodb_checksum_algorithm = crc32
innodb_lock_wait_timeout = 10
innodb_rollback_on_timeout = 1
innodb_print_all_deadlocks = 1
innodb_file_per_table = 1
innodb_online_alter_log_max_size = 4G
innodb_stats_on_metadata = 0

# some var for MySQL 8
log_error_verbosity = 3
innodb_print_ddl_logs = 1
binlog_expire_logs_seconds = 2592000
#innodb_dedicated_server = 0

innodb_status_file = 1
# 注意: 开启 innodb_status_output  innodb_status_output_locks 后, 可能会导致log-error文件增长较快
innodb_status_output = 0
innodb_status_output_locks = 0

#performance_schema
performance_schema = 1
performance_schema_instrument = '%memory%=on'
performance_schema_instrument = '%lock%=on'

#innodb monitor
innodb_monitor_enable="module_innodb"
innodb_monitor_enable="module_server"
innodb_monitor_enable="module_dml"
innodb_monitor_enable="module_ddl"
innodb_monitor_enable="module_trx"
innodb_monitor_enable="module_os"
innodb_monitor_enable="module_purge"
innodb_monitor_enable="module_log"
innodb_monitor_enable="module_lock"
innodb_monitor_enable="module_buffer"
innodb_monitor_enable="module_index"
innodb_monitor_enable="module_ibuf_system"
innodb_monitor_enable="module_buffer_page"
innodb_monitor_enable="module_adaptive_hash"

#MGR
#GR配置项 基中loose前缀表示若group Replication plugin未加载 mysql server仍明治维新启动
transaction_write_set_extraction = XXHASH64 #对每个事务获取write set，并且用XXHASH64算法获取hash值
loose-group_replication_group_name = "58f6e65e-9309-11e9-9d88-525400184a0a" #组名，此处可拿select uuid()生成
loose-group_replication_start_on_boot = off #在mysqld启动时不自动启动组复制
loose-group_replication_local_address = "10.10.146.28:33006" #本节点的IP地址和端口，注意该端口是组内成员之间通信的端口，而不是MySQL对外提供服务的端口
loose-group_replication_group_seeds = "10.10.146.28:33006,10.10.1.139:33006,10.10.173.84:33006" #种子节点的IP和端口号，新成员加入到集群的时候需要联系种子节点，启动集群的节点不使用该选项
loose-group_replication_bootstrap_group = off #关闭，如果打开会造成脑裂 #是否启动集群，注意，该选项任何时候只能用于一个节点，通常情况下启动集群的时候使用，启动之后需要关闭该选项

loose-group_replication_member_weight = 50 #权重选择


[mysqldump]
quick
max_allowed_packet = 32M

[mysqld_safe]
#malloc-lib=/usr/local/mysql/lib/jmalloc.so 
nice=-19
open-files-limit=65535

EOF

# 第二台
if [ -f /etc/my.cnf ]; then
 mv /etc/my.cnf /etc/my.cnf.`date +%Y%m%d%H%m`.bak
fi 
# node1
cat >/data/mysql/mysql_3306/my_3306.cnf "EOF"
[client]
port = 3306
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock

[mysql]
prompt="\u@\h \R:\m:\s [\d]> "
no-auto-rehash

[mysqld]
user = mysql
port = 3306
admin_address = 127.0.0.1
basedir = /usr/local/mysql
datadir = /data/mysql/mysql_3306/data
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock
pid-file = mysql_3306.pid
character-set-server = utf8mb4
skip_name_resolve = 1

#replicate-wild-ignore-table=mysql.%
#replicate-wild-ignore-table=test.%
#replicate-wild-ignore-table=information_schema.%

# Two-Master configure
#server-1 
#auto-increment-offset = 1
#auto-increment-increment = 2 

#server-2    
#auto-increment-offset = 2
#auto-increment-increment = 2


# semi sync replication settings #
#plugin_dir = /usr/local/mysql/lib/mysql/plugin
#plugin_load = "validate_password.so;rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"
plugin_dir = /usr/local/mysql/lib/plugin #官方版本的路径
plugin_load = "rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so" #官方版本的路径

slave_parallel_workers = 4
slave_parallel_type = LOGICAL_CLOCK
slave_preserve_commit_order = 1

open_files_limit = 65535
back_log = 1024
max_connections = 1024
max_connect_errors = 1000000
table_open_cache = 1024
table_definition_cache = 1024
table_open_cache_instances = 64
thread_stack = 512K
external-locking = FALSE
max_allowed_packet = 32M
sort_buffer_size = 4M
join_buffer_size = 4M
thread_cache_size = 1536
interactive_timeout = 600
wait_timeout = 600
tmp_table_size = 32M
max_heap_table_size = 32M
slow_query_log = 1
log_timestamps = SYSTEM
slow_query_log_file = /data/mysql/mysql_3306/logs/slow.log
log-error = /data/mysql/mysql_3306/logs/error.log
long_query_time = 0.1
log_queries_not_using_indexes =1
log_throttle_queries_not_using_indexes = 60
min_examined_row_limit = 100
log_slow_admin_statements = 1
log_slow_slave_statements = 1
server-id = 1433306
log-bin = /data/mysql/mysql_3306/logs/mysql-bin
sync_binlog = 1
binlog_cache_size = 4M
max_binlog_cache_size = 2G
max_binlog_size = 1G
binlog_expire_logs_seconds=2592000 
master_info_repository = TABLE
relay_log_info_repository = TABLE
gtid_mode = on
enforce_gtid_consistency = 1
binlog_checksum=NONE
log_slave_updates
slave-rows-search-algorithms = 'INDEX_SCAN,HASH_SCAN'
binlog_format = row
binlog_row_image=FULL

relay_log_recovery = 1
relay-log-purge = 1
key_buffer_size = 32M
read_buffer_size = 8M
read_rnd_buffer_size = 4M
bulk_insert_buffer_size = 64M
myisam_sort_buffer_size = 128M
myisam_max_sort_file_size = 10G
myisam_repair_threads = 1
lock_wait_timeout = 3600
explicit_defaults_for_timestamp = 1
innodb_thread_concurrency = 0
innodb_sync_spin_loops = 100
innodb_spin_wait_delay = 30

#transaction_isolation = REPEATABLE-READ
transaction_isolation = READ-COMMITTED
#innodb_additional_mem_pool_size = 16M
innodb_buffer_pool_size = 2867M
innodb_buffer_pool_instances = 4
innodb_buffer_pool_load_at_startup = 1
innodb_buffer_pool_dump_at_shutdown = 1
innodb_data_file_path = ibdata1:1G:autoextend
innodb_flush_log_at_trx_commit = 1
innodb_log_buffer_size = 32M
innodb_log_file_size = 2G
innodb_log_files_in_group = 3
innodb_max_undo_log_size = 4G
innodb_undo_directory = /data/mysql/mysql_3306/undolog

# 根据您的服务器IOPS能力适当调整
# 一般配普通SSD盘的话，可以调整到 10000 - 20000
# 配置高端PCIe SSD卡的话，则可以调整的更高，比如 50000 - 80000
innodb_io_capacity = 4000
innodb_io_capacity_max = 8000
innodb_flush_sync = 0
innodb_flush_neighbors = 0
innodb_write_io_threads = 8
innodb_read_io_threads = 8
innodb_purge_threads = 4
innodb_page_cleaners = 4
innodb_open_files = 65535
innodb_max_dirty_pages_pct = 50
innodb_flush_method = O_DIRECT
innodb_lru_scan_depth = 4000
innodb_checksum_algorithm = crc32
innodb_lock_wait_timeout = 10
innodb_rollback_on_timeout = 1
innodb_print_all_deadlocks = 1
innodb_file_per_table = 1
innodb_online_alter_log_max_size = 4G
innodb_stats_on_metadata = 0

# some var for MySQL 8
log_error_verbosity = 3
innodb_print_ddl_logs = 1
binlog_expire_logs_seconds = 2592000
#innodb_dedicated_server = 0

innodb_status_file = 1
# 注意: 开启 innodb_status_output  innodb_status_output_locks 后, 可能会导致log-error文件增长较快
innodb_status_output = 0
innodb_status_output_locks = 0

#performance_schema
performance_schema = 1
performance_schema_instrument = '%memory%=on'
performance_schema_instrument = '%lock%=on'

#innodb monitor
innodb_monitor_enable="module_innodb"
innodb_monitor_enable="module_server"
innodb_monitor_enable="module_dml"
innodb_monitor_enable="module_ddl"
innodb_monitor_enable="module_trx"
innodb_monitor_enable="module_os"
innodb_monitor_enable="module_purge"
innodb_monitor_enable="module_log"
innodb_monitor_enable="module_lock"
innodb_monitor_enable="module_buffer"
innodb_monitor_enable="module_index"
innodb_monitor_enable="module_ibuf_system"
innodb_monitor_enable="module_buffer_page"
innodb_monitor_enable="module_adaptive_hash"

#MGR
#GR配置项 基中loose前缀表示若group Replication plugin未加载 mysql server仍明治维新启动
transaction_write_set_extraction = XXHASH64 #对每个事务获取write set，并且用XXHASH64算法获取hash值
loose-group_replication_group_name = "58f6e65e-9309-11e9-9d88-525400184a0a" #组名，此处可拿select uuid()生成
loose-group_replication_start_on_boot = off #在mysqld启动时不自动启动组复制
loose-group_replication_local_address = "10.10.1.139:33006" #本节点的IP地址和端口，注意该端口是组内成员之间通信的端口，而不是MySQL对外提供服务的端口
loose-group_replication_group_seeds = "10.10.146.28:33006,10.10.1.139:33006,10.10.173.84:33006" #种子节点的IP和端口号，新成员加入到集群的时候需要联系种子节点，启动集群的节点不使用该选项
loose-group_replication_bootstrap_group = off #关闭，如果打开会造成脑裂 #是否启动集群，注意，该选项任何时候只能用于一个节点，通常情况下启动集群的时候使用，启动之后需要关闭该选项

loose-group_replication_member_weight = 50 #权重选择


[mysqldump]
quick
max_allowed_packet = 32M

[mysqld_safe]
#malloc-lib=/usr/local/mysql/lib/jmalloc.so 
nice=-19
open-files-limit=65535

EOF

# 第三台
if [ -f /etc/my.cnf ]; then
 mv /etc/my.cnf /etc/my.cnf.`date +%Y%m%d%H%m`.bak
fi 
# node1
cat >/data/mysql/mysql_3306/my_3306.cnf "EOF"
[client]
port = 3306
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock

[mysql]
prompt="\u@\h \R:\m:\s [\d]> "
no-auto-rehash

[mysqld]
user = mysql
port = 3306
admin_address = 127.0.0.1
basedir = /usr/local/mysql
datadir = /data/mysql/mysql_3306/data
socket = /data/mysql/mysql_3306/tmp/mysql_3306.sock
pid-file = mysql_3306.pid
character-set-server = utf8mb4
skip_name_resolve = 1

#replicate-wild-ignore-table=mysql.%
#replicate-wild-ignore-table=test.%
#replicate-wild-ignore-table=information_schema.%

# Two-Master configure
#server-1 
#auto-increment-offset = 1
#auto-increment-increment = 2 

#server-2    
#auto-increment-offset = 2
#auto-increment-increment = 2


# semi sync replication settings #
#plugin_dir = /usr/local/mysql/lib/mysql/plugin
#plugin_load = "validate_password.so;rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so"
plugin_dir = /usr/local/mysql/lib/plugin #官方版本的路径
plugin_load = "rpl_semi_sync_master=semisync_master.so;rpl_semi_sync_slave=semisync_slave.so" #官方版本的路径

slave_parallel_workers = 4
slave_parallel_type = LOGICAL_CLOCK
slave_preserve_commit_order = 1

open_files_limit = 65535
back_log = 1024
max_connections = 1024
max_connect_errors = 1000000
table_open_cache = 1024
table_definition_cache = 1024
table_open_cache_instances = 64
thread_stack = 512K
external-locking = FALSE
max_allowed_packet = 32M
sort_buffer_size = 4M
join_buffer_size = 4M
thread_cache_size = 1536
interactive_timeout = 600
wait_timeout = 600
tmp_table_size = 32M
max_heap_table_size = 32M
slow_query_log = 1
log_timestamps = SYSTEM
slow_query_log_file = /data/mysql/mysql_3306/logs/slow.log
log-error = /data/mysql/mysql_3306/logs/error.log
long_query_time = 0.1
log_queries_not_using_indexes =1
log_throttle_queries_not_using_indexes = 60
min_examined_row_limit = 100
log_slow_admin_statements = 1
log_slow_slave_statements = 1
server-id = 1443306
log-bin = /data/mysql/mysql_3306/logs/mysql-bin
sync_binlog = 1
binlog_cache_size = 4M
max_binlog_cache_size = 2G
max_binlog_size = 1G
binlog_expire_logs_seconds=2592000 
master_info_repository = TABLE
relay_log_info_repository = TABLE
gtid_mode = on
enforce_gtid_consistency = 1
binlog_checksum=NONE
log_slave_updates
slave-rows-search-algorithms = 'INDEX_SCAN,HASH_SCAN'
binlog_format = row
binlog_row_image=FULL

relay_log_recovery = 1
relay-log-purge = 1
key_buffer_size = 32M
read_buffer_size = 8M
read_rnd_buffer_size = 4M
bulk_insert_buffer_size = 64M
myisam_sort_buffer_size = 128M
myisam_max_sort_file_size = 10G
myisam_repair_threads = 1
lock_wait_timeout = 3600
explicit_defaults_for_timestamp = 1
innodb_thread_concurrency = 0
innodb_sync_spin_loops = 100
innodb_spin_wait_delay = 30

#transaction_isolation = REPEATABLE-READ
transaction_isolation = READ-COMMITTED
#innodb_additional_mem_pool_size = 16M
innodb_buffer_pool_size = 2867M
innodb_buffer_pool_instances = 4
innodb_buffer_pool_load_at_startup = 1
innodb_buffer_pool_dump_at_shutdown = 1
innodb_data_file_path = ibdata1:1G:autoextend
innodb_flush_log_at_trx_commit = 1
innodb_log_buffer_size = 32M
innodb_log_file_size = 2G
innodb_log_files_in_group = 3
innodb_max_undo_log_size = 4G
innodb_undo_directory = /data/mysql/mysql_3306/undolog

# 根据您的服务器IOPS能力适当调整
# 一般配普通SSD盘的话，可以调整到 10000 - 20000
# 配置高端PCIe SSD卡的话，则可以调整的更高，比如 50000 - 80000
innodb_io_capacity = 4000
innodb_io_capacity_max = 8000
innodb_flush_sync = 0
innodb_flush_neighbors = 0
innodb_write_io_threads = 8
innodb_read_io_threads = 8
innodb_purge_threads = 4
innodb_page_cleaners = 4
innodb_open_files = 65535
innodb_max_dirty_pages_pct = 50
innodb_flush_method = O_DIRECT
innodb_lru_scan_depth = 4000
innodb_checksum_algorithm = crc32
innodb_lock_wait_timeout = 10
innodb_rollback_on_timeout = 1
innodb_print_all_deadlocks = 1
innodb_file_per_table = 1
innodb_online_alter_log_max_size = 4G
innodb_stats_on_metadata = 0

# some var for MySQL 8
log_error_verbosity = 3
innodb_print_ddl_logs = 1
binlog_expire_logs_seconds = 2592000
#innodb_dedicated_server = 0

innodb_status_file = 1
# 注意: 开启 innodb_status_output  innodb_status_output_locks 后, 可能会导致log-error文件增长较快
innodb_status_output = 0
innodb_status_output_locks = 0

#performance_schema
performance_schema = 1
performance_schema_instrument = '%memory%=on'
performance_schema_instrument = '%lock%=on'

#innodb monitor
innodb_monitor_enable="module_innodb"
innodb_monitor_enable="module_server"
innodb_monitor_enable="module_dml"
innodb_monitor_enable="module_ddl"
innodb_monitor_enable="module_trx"
innodb_monitor_enable="module_os"
innodb_monitor_enable="module_purge"
innodb_monitor_enable="module_log"
innodb_monitor_enable="module_lock"
innodb_monitor_enable="module_buffer"
innodb_monitor_enable="module_index"
innodb_monitor_enable="module_ibuf_system"
innodb_monitor_enable="module_buffer_page"
innodb_monitor_enable="module_adaptive_hash"

#MGR
#GR配置项 基中loose前缀表示若group Replication plugin未加载 mysql server仍明治维新启动
transaction_write_set_extraction = XXHASH64 #对每个事务获取write set，并且用XXHASH64算法获取hash值
loose-group_replication_group_name = "58f6e65e-9309-11e9-9d88-525400184a0a" #组名，此处可拿select uuid()生成
loose-group_replication_start_on_boot = off #在mysqld启动时不自动启动组复制
loose-group_replication_local_address = "10.10.173.84:33006" #本节点的IP地址和端口，注意该端口是组内成员之间通信的端口，而不是MySQL对外提供服务的端口
loose-group_replication_group_seeds = "10.10.146.28:33006,10.10.1.139:33006,10.10.173.84:33006" #种子节点的IP和端口号，新成员加入到集群的时候需要联系种子节点，启动集群的节点不使用该选项
loose-group_replication_bootstrap_group = off #关闭，如果打开会造成脑裂 #是否启动集群，注意，该选项任何时候只能用于一个节点，通常情况下启动集群的时候使用，启动之后需要关闭该选项

loose-group_replication_member_weight = 50 #权重选择


[mysqldump]
quick
max_allowed_packet = 32M

[mysqld_safe]
#malloc-lib=/usr/local/mysql/lib/jmalloc.so 
nice=-19
open-files-limit=65535

EOF

chown -R mysql.mysql /data/mysql/mysql_3306
chown -R mysql.mysql /usr/local/mysql/

#初始化
# /usr/local/mysql/bin/mysqld --user=mysql --basedir=/usr/local/mysql --datadir=/data/mysql --initialize-insecure
# 官方推荐使用--initialize，会在错误日志中生成难以输入的临时密码，我这里使用的免密码的方式。
/usr/local/mysql/bin/mysqld --defaults-file=/data/mysql/mysql_3306/my_3306.cnf --initialize-insecure --user=mysql 

#启动数据库
/usr/local/mysql/bin/mysqld_safe --defaults-file=/data/mysql/mysql_3306/my_3306.cnf 

#9、查看日志
# tail -f /data/mysql/mysql_3306/logs/error.log

#10、初次登陆
/usr/local/mysql/bin/mysql -S /data/mysql/mysql_3306/tmp/mysql_3306.sock

# 修改密码方法

set sql_log_bin = 0;

ALTER USER 'root'@'localhost' IDENTIFIED WITH mysql_native_password BY 'GJjumB6g4FcwdF3R6AZc' PASSWORD EXPIRE NEVER ;
create user 'root'@'127.0.0.1' identified WITH mysql_native_password by 'GJjumB6g4FcwdF3R6AZc' PASSWORD EXPIRE NEVER ;
grant all privileges on *.* to 'root'@'127.0.0.1' with grant option;

create user 'admin_m'@'127.0.0.1' identified WITH mysql_native_password by 'fcfmTbRw1tz2x5L5GvjJ' PASSWORD EXPIRE NEVER ; 
grant all privileges on *.* to 'admin_m'@'127.0.0.1' with grant option;

create user 'admin_m'@'%' identified WITH mysql_native_password by 'fcfmTbRw1tz2x5L5GvjJ' PASSWORD EXPIRE NEVER ; 
grant all privileges on *.* to 'admin_m'@'%' with grant option;

create user 'test_w'@'%' identified with mysql_native_password by 'EeCrfUDO6wRzn72BBQ52' PASSWORD EXPIRE NEVER ;
grant insert,delete,update,select on db144.* to 'test_w'@'%' ;

create user 'test_r'@'%' identified with mysql_native_password by 'EeCrfUDO6wRzn72BBQ52' PASSWORD EXPIRE NEVER ;
grant insert,delete,update,select on db144.* to 'test_r'@'%' ;

create user 'repl'@'%' IDENTIFIED with mysql_native_password by 'replpfhOTnWffQdQL3F3' ;
GRANT REPLICATION SLAVE ON *.* TO 'repl'@'%' ;
set sql_log_bin = 1;

ln -s /usr/local/mysql/lib/libmysqlclient.so /usr/lib/
ln -s /usr/local/mysql/lib/libmysqlclient.so.21 /usr/lib/libmysqlclient.so.21
ln -s /usr/local/mysql/lib/libmysqlclient.so /usr/lib64/
ln -s /usr/local/mysql/lib/libmysqlclient.so.21 /usr/lib64/libmysqlclient.so.21


ln -s /data/mysql/mysql_3306/tmp/mysql_3306.sock /tmp/mysql.sock
ln -s /usr/local/mysql/bin/* /usr/bin/

cat >>~/.bashrc "EOF"
##########


alias mysql.3306.start="/usr/local/mysql/bin/mysqld_safe --defaults-file=/data/mysql/mysql_3306/my_3306.cnf "
alias mysql.3306.stop="/usr/local/mysql/bin/mysqladmin -h127.0.0.1 -P 3306 -uroot -p'GJjumB6g4FcwdF3R6AZc' shutdown "
alias mysql.3306.login="/usr/local/mysql/bin/mysql -h127.0.0.1 -P 3306 -uroot -p'GJjumB6g4FcwdF3R6AZc'"


##########
EOF


source /root/.bash_profile

cat >>/etc/ld.so.conf "EOF"
/usr/local/mysql/lib
EOF

ldconfig 
mysql.3306.login

# MGR 第一台配置：
# 第一步：创建用于复制的用户

set sql_log_bin=0;
create user 'repuser'@'%' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'%';

create user 'repuser'@'127.0.0.1' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'127.0.0.1';

create user 'repuser'@'localhost' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'localhost';

set sql_log_bin=1;

# 第二步：配置复制所使用的用户

change master to master_user='repuser',master_password='JhXpMK44ju8Vp5bxvO2N' for channel 'group_replication_recovery';

# 第三步：安装mysql group replication这个插件


# 备注：如果在my.cnf里写写入 plugin_load="group_replication=group_replication.so" 这步就可以不用操作

install plugin group_replication soname 'group_replication.so';

# 通过show plugins;查看是否安装成功
show plugins;

# 第四步：建个群(官方点的说法就是初始化一个复制组

set global group_replication_bootstrap_group=on;
start group_replication;
set global group_replication_bootstrap_group=off;

select * from performance_schema.replication_group_members;

##########################################################################
#MGR 配置其他从节点
#在所有从主机上的mysql中执行
# 第一步：创建用于复制的用户

set sql_log_bin=0;
create user 'repuser'@'%' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'%';

create user 'repuser'@'127.0.0.1' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'127.0.0.1';

create user 'repuser'@'localhost' identified by 'JhXpMK44ju8Vp5bxvO2N';

grant replication slave,replication client on *.* to 'repuser'@'localhost';

set sql_log_bin=1;

# 第二步：配置复制所使用的用户

change master to master_user='repuser',master_password='JhXpMK44ju8Vp5bxvO2N' for channel 'group_replication_recovery';

# 第三步：安装mysql group replication这个插件


# 备注：如果在my.cnf里写写入 plugin_load="group_replication=group_replication.so" 这步就可以不用操作

install plugin group_replication soname 'group_replication.so';

# 通过show plugins;查看是否安装成功
show plugins;

# 第四步：加入前面创建好的复制组

start group_replication;
select * from performance_schema.replication_group_members;
#########################################################################################

# 检查状态
mysql> select * from performance_schema.replication_group_members;
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| CHANNEL_NAME  | MEMBER_ID    | MEMBER_HOST | MEMBER_PORT | MEMBER_STATE | MEMBER_ROLE | MEMBER_VERSION |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
| group_replication_applier | d955da6d-0048-11ea-b7b4-525400f4342d | bj-db-m1 | 3306 | ONLINE | PRIMARY | 8.0.18  |
| group_replication_applier | e050c34f-0048-11ea-917d-52540021fab9 | bj-db-m3 | 3306 | ONLINE | SECONDARY | 8.0.18  |
| group_replication_applier | e6c56347-0048-11ea-9e8b-5254007c241f | bj-db-m2 | 3306 | ONLINE | SECONDARY | 8.0.18  |
+---------------------------+--------------------------------------+-------------+-------------+--------------+-------------+----------------+
3 rows in set (0.00 sec)

######################################################################
# 单主切换到多主
# MGR切换模式需要重新启动组复制，因此需要在所有节点上先关闭组复制，
#设置 group_replication_single_primary_mode=OFF 等参数，再启动组复制。
1) 停止组复制(在所有MGR节点上执行)：
stop group_replication; 
set global group_replication_single_primary_mode=OFF;
set global group_replication_enforce_update_everywhere_checks=ON;

2) 随便选择某个MGR节点执行 (比如这里选择在MGR-node1节点):
set global group_replication_recovery_get_public_key=1;
SET GLOBAL group_replication_bootstrap_group=ON;
START GROUP_REPLICATION;
SET GLOBAL group_replication_bootstrap_group=OFF;

3) 然后在其他的MGR节点执行 (这里指MGR-node2和MGR-node3节点上执行):
set global group_replication_recovery_get_public_key=1;
START GROUP_REPLICATION;

4) 查看MGR组信息 (在任意一个MGR节点上都可以查看)
SELECT * FROM performance_schema.replication_group_members;


# 可以看到所有MGR节点状态都是online，角色都是PRIMARY，MGR多主模式搭建成功。

##########################################################################
# 多主切回单主模式
1) 停止组复制(在所有MGR节点上执行)：
stop group_replication;
set global group_replication_enforce_update_everywhere_checks=OFF;
set global group_replication_single_primary_mode=ON;

2) 选择一个节点作为主节点, 在主节点上执行 (这里选择MGR-node1节点作为主节点)
SET GLOBAL group_replication_bootstrap_group=ON;
START GROUP_REPLICATION;
SET GLOBAL group_replication_bootstrap_group=OFF;
 
3) 在其他剩余的节点, 也就是从库节点上执行 (这里从库节点指的就是MGR-node2和MGR-node3):
START GROUP_REPLICATION;

4) 查看MGR组信息 (在任意一个MGR节点上都可以查看)
SELECT * FROM performance_schema.replication_group_members;
##########################################################################

# 故障注意点：
# 单主模式，恢复MGR-node1节点, 恢复后, 需要手动激活下该节点的组复制功能
# 如果节点发生故障, 在恢复后需要重新加入到MGR集群里, 正确的做法是:
STOP GROUP_REPLICATION;
START GROUP_REPLICATION;

# 如果某个节点挂了, 则其他的节点继续进行同步.
# 当故障节点恢复后, 只需要手动激活下该节点的组复制功能("START GROUP_REPLICATION;"),
# 即可正常加入到MGR组复制集群内并自动同步其他节点数据.


# 如果是i/o复制出现异常
# 确定数据无误后
# 查找主库的gtid情况
mysql> show global variables like '%gtid%' ;
+----------------------------------------------+-------------------------------------------------------+
| Variable_name    | Value       |
+----------------------------------------------+-------------------------------------------------------+
| binlog_gtid_simple_recovery   | ON       |
| enforce_gtid_consistency   | ON       |
| group_replication_gtid_assignment_block_size | 1000000      |
| gtid_executed    | 58f6e65e-9309-11e9-9d88-525400184a0a:1-946050:1000003 |
| gtid_executed_compression_period  | 1000       |
| gtid_mode     | ON       |
| gtid_owned     |       |
| gtid_purged     |       |
| session_track_gtids    | OFF       |
+----------------------------------------------+-------------------------------------------------------+
rows in set (0.00 sec)

# 在有故障的从库中操作
stop GROUP_REPLICATION;
reset master;
set global gtid_purged='58f6e65e-9309-11e9-9d88-525400184a0a:1-946055:1000003';
START GROUP_REPLICATION;

# 添加白名单网段
stop group_replication;
set global group_replication_ip_whitelist="127.0.0.1/32,172.16.60.0/24,172.16.50.0/24,172.16.51.0/24";
start group_replication;
show variables like "group_replication_ip_whitelist";

# 一定要注意: 配置白名单前面一定要先关闭 Group Replication, 及先要执行"stop group_replication;"