Master server

$ mysql -uroot -p

mysql> GRANT REPLICATION SLAVE ON *.* to 'repl'@'172.%' IDENTIFIED BY 'repl-password-here';

mysql> flush tables with read lock;

mysql> reset master;

# Make note of the file name and file position

mysql> show master status;

$ /etc/init.d/mysql stop

$ xfs_freeze -f /mnt/mysql

# Take AWS EC2 EBS snapshot and when it begins do next step

$ xfs_freeze -u /mnt/mysql

$ /etc/init.d/mysql start

$ mysql -uroot -p

mysql> unlock tables;

Wait for WAS to finish writing the EBS snapshot before proceeding.

Slave server

1. EBS snapshot is complete
2. create a new volume from snapshot
3. umount old volume
4. umount /data
5. Detach "/data" volume
6. Attach new "/data" volume
7. mount /data
8. lsblk
9. sudo dd if=/dev/xvdX of=/dev/null bs=1M
10. syncing to the master

$ /etc/init.d/mysql start

$ mysql -uroot -p

mysql> CHANGE MASTER TO

 MASTER_HOST='172.x.x.x',

 MASTER_USER='repl',

 MASTER_PASSWORD='repl-password-here',

 MASTER_LOG_FILE = 'mysql-bin.000001',

 MASTER_LOG_POS = 107;

mysql> start slave;

mysql> show slave status \G

Reference

• https://jrklein.com/2017/03/22/mysql-master-slave-replication-on-aws-ec2-ebs/
• https://www.tecmint.com/mariadb-master-slave-replication-on-centos-rhel-debian/