New answers tagged mysql-replication
The error saying "mysql.gtid_slave_pos does not exist" is blocking the replication process on the slave. You need to fix this error before replication can go on. You can try to ignore this error and see what happens. Connect with mysql CLI on the slave and type the following: mysql> STOP SLAVE; mysql> SET GLOBAL SQL_SLAVE_SKIP_COUNTER = 1; mysql> ...
You seem a bit confused about MySQL replication modes. They are all Master/Master or Master/Slave. The first choice to make is whether you want to use asynchronous, semi-synchronous or synchronous replication. The former is available on all platforms out-of-the-box, but does have technical limitations. The latter needs additional components. How you then ...
The answer in large part will be "It depends". Are you replicating to take backups from Slaves? Are you replicating to have Slaves provide read servicing for your application? Do you need read-after-write consistency? Are you replicating for high availability? However you may want to investigate MySQL 5.7's new capability of multi-source replication since ...
I used a Master Slave replication for many years exp on distibuited network, but I suggest you a solution based on GALERA http://galeracluster.com/products/ it used also inside openstack , it's absolute the best one comparing to Master to master for me.
Your subnets are perfect the way they are. No change needed there. What you need is an internal load balancer sitting in from your database servers. If you are using Azure Resource Manager, here are the steps to set this up? If you are using the older/classic networks and virtual machines, then here are the steps to configure in that environment.
Top 50 recent answers are included