How to setup a single master with one or more readonly slaves

First configure the master by editing the mysql configuration file, this could be /etc/my.cnf or on Centos 7 it is /etc/my.cnf.d/server.cnf

Find the [server] section and add the following lines:

bind-address = 12.34.56.78 # replace 12.34.56.78 with the IP of your server
log_bin
server_id = 1
log_basename = master1
datadir = /var/lib/mysql
binlog-ignore-db = mysql

The binlog-ignore-db line tells the master not to replicate the ‘mysql’ database, you can add additional entries to not replicate other databases if you wish, e.g.

binlog-ignore-db = information_schema
binlog-ignore-db = performance_schema

Adding the following additional entries allows you to tweak the performance of your server:

max_allowed_packet=64M
max_heap_table_size = 64M
tmp_table_size = 128M
join_buffer_size = 128M
innodb_buffer_pool_size = 256M
innodb_doublewrite = OFF
innodb_additional_mem_pool_size = 128M
innodb_flush_log_at_timeout = 4
innodb_read_io_threads = 48
innodb_write_io_threads = 32
max_connections = 128

Following any changes to the config files you will need to restart MySQL (or MariaDB if you use that instead), so you would do one of the following:

systemctl restart mysqld
systemctl restart mariadb

Now login to your local MySQL server, e.g.

mysql -u root -p

Once logged into the MySQL CLI add a user that the slaves will use to replicate:

GRANT REPLICATION SLAVE ON *.* TO 'replication'@'1.2.3.4' IDENTIFIED BY 'password';
FLUSH PRIVILEGES;

Replace 1.2.3.4 with the IP of your slave server. You can run the above command multiple times, once for each slave server. Remember to change ‘password’ to an actual password, a good choice would be 12 (or more) characters consisting of a random mix of upper/lowercase letters, numbers and punctuation characters.

Now we’re ready to take a backup of the master, for this you will need to open a second session on your master server. In the first session, still logged into the MySQL CLI issue the following commands:

FLUSH TABLES WITH READ LOCK;
SHOW MASTER STATUS;

You should see something along these lines:

+--------------------+----------+--------------+------------------+
| File               | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+--------------------+----------+--------------+------------------+
| master1-bin.000001 |   456    |              | mysql            |
+--------------------+----------+--------------+------------------+
1 row in set (0.00 sec)

Leave this session in place whilst you move to the second session to backup the databases you want to replicate by using your preferred method, e.g.

mysqldump -u root -p --all-databases > dump.sql

Now you can go back to the first window and release the lock:

UNLOCK TABLES;
QUIT;

Now copy this backup to your slave server(s). I use scp for this, e.g.

scp dump.sql username@12.34.56.78:.

Now login to the slave server and import the dumped databases:

mysql -u root -p < dump.sql

Now configure MySQL (or mariadb) to be a slave by editing it’s configuration file, e.g. /etc/my.cnf.d/server.cnf and add the following to the [server] section:

server-id = 2
datadir = /var/lib/mysql
relay-log = /var/lib/mysql/mysql-relay-bin.log
log_bin = /var/lib/mysql/mysql-bin.log

As per the master server, following any changes to the config files you will need to restart MySQL (or MariaDB if you use that instead), so you would do one of the following:

systemctl restart mysqld
systemctl restart mariadb

The next step is to login to the MySQL CLI and tell it where to find the master along with the login details and the starting position in the log:

CHANGE MASTER TO MASTER_HOST='12.34.56.78',MASTER_USER='replication', MASTER_PASSWORD='password', MASTER_LOG_FILE='master1-bin.000001', MASTER_LOG_POS=456;

The last thing we need to do is start the slave process, and check it is running. Do this by issuing the following commands on the slave MySQL CLI:

START SLAVE;
SHOW SLAVE STATUS\G

You can repeat the slave part on each server you want to setup replication, just remember to give each slave a different server_id

Remove any existing packages:

yum remove maria*

Update:

yum update

Add the official repo for MariaDB by creating the file /etc/yum.repos.d/MariaDB.repo

[mariadb]
name = MariaDB
baseurl = http://yum.mariadb.org/10.1/centos7-amd64
gpgkey=https://yum.mariadb.org/RPM-GPG-KEY-MariaDB
gpgcheck=1

Now install MariaDB:

yum install -y MariaDB-server MariaDB-client MariaDB-compat galera socat jemalloc

Setup MariaDB:

systemctl start mariadb
mysql_secure_installation
systemctl stop mariadb

To generate the CA certificate:

openssl genrsa 2048 > ca-key.pem
openssl req -new -x509 -nodes -days 3600 -key ca-key.pem -out ca.pem

To generate the server certificate, remove passphrase, and sign it:

openssl req -newkey rsa:2048 -days 3600 -nodes -keyout server-key.pem -out server-req.pem
openssl rsa -in server-key.pem -out server-key.pem
openssl x509 -req -in server-req.pem -days 3600 -CA ca.pem -CAkey ca-key.pem -set_serial -1 -out server-cert.pem

(Optional) To generate the client certificate, remove passphrase, and sign it:

openssl req -newkey rsa:2048 -days 3600 -nodes -keyout client-key.pem -out client-req.pem
openssl rsa -in client-key.pem -out client-key.pem
openssl x509 -req -in client-req.pem -days 3600 -CA ca.pem -CAkey ca-key.pem -set_serial 01 -out client-cert.pem

Edit the file: /etc/my.cnf.d/server.cnf

[sst]
encrypt=4
ssl-ca=/etc/pki/ca.pem
ssl-cert=/etc/pki/server-cert.pem
ssl-key=/etc/pki/server-key.pem

[galera]
wsrep_on=ON
wsrep_provider=/usr/lib64/galera/libgalera_smm.so
wsrep_cluster_address='gcomm://a.a.a.a,b.b.b.b,c.c.c.c'
wsrep_cluster_name='cluster.name'
wsrep_node_address='10.0.0.11'
wsrep_node_name='node1'
wsrep_sst_method=rsync
wsrep_sst_receive_address='x.x.x.x'
wsrep_provider_options='socket.ssl_key=/etc/pki/server-key.pem;socket.ssl_cert=/etc/pki/server-cert.pem;socket.ssl_ca=/etc/pki/ca.pem;evs.inactive_timeout=PT45S;evs.install_timeout=PT45S;evs.keepalive_period=PT3S;evs.max_install_timeouts=8;evs.send_window=512;evs.suspect_timeout=PT30S;evs.user_send_window=256;'
binlog_format=row
default_storage_engine=InnoDB
innodb_autoinc_lock_mode=2

In the file above the line “wsrep_sst_receive_address=’x.x.x.x'” is required if any of the nodes are behind a NAT router on private IP addresses, where x.x.x.x is the public IP address of the router. Without this SST donors will try to send snapshot data to the nodes private IP address which will invariably fail.

The “wsrep_provider_options” are tailored to for nodes that talk to each other over a WAN (i.e. the internet). If your nodes are all on the same LAN then you can leave this option out completely – it adjusts some timeout default values to better cope with varying connectivity quality across a WAN.

Start the primary node:

galera_new_cluster

Start the other nodes:

systemctl start mariadb

Login to any of the nodes and check status:

show status like 'wsrep%';