I am using MySQL 5.7 on RHEL 7.1. For cluster management I am using Pacemaker and corosync. For shared storage, I am using GFS2 file system with iSCSI server

I am trying to implement MySQL active-active cluster (both nodes running at the same time). I have two nodes with MySQL server running. I am not able to give same data directory (shared storage) for both the nodes.

I tried to change the data directory in my.cnf file. After that the service didn't start. Currently, I am using the default storage engine i.e. InnoDB storage engine.

Is this a problem of the underlying storage engine or something else we need to change in the configuration?

  • Define "cluster". NDB Cluster? Galera Cluster? Something else? – Rick James Feb 15 '17 at 0:32

Two instances of mysqld must not use the same data files. The code is not designed to handle unexpected changes to the file, which will happen by the other instance.

(I assume you are talking about Master-Master replication when you say "cluster"?)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.