MySQL Cluster is a different and separate product from the clustering
approaches, such as DRBD with Pacemaker, implemented to protect a
regular MySQL server
MySQL Cluster uses the NDB storage engine to automatically shard,
distribute and replicate your database across a cluster of nodes over a
network. You can learn more about MySQL Cluster here:
MySQL with the InnoDB storage engine is often deployed with DRBD and
Pacemaker / Corosync to provide lossless failover. It is a stack that is
supported by Oracle as part of the MySQL Enterprise Edition offering.
The whitepaper posted as follows provides details on how to configure
(note, registration required):
I can put you in touch with one of our consultants if you wish to
discuss DRBD further
On 07/03/2013 11:56, Ravi Kiran PENTYALA wrote:
> Hi All,
> Article on Active/Passive MYSQLSQL Cluster, setup using Pacemaker with
> Corosync and DRBD
> We use two nodes, one active (that answers requests from apps) and the
> other on passive mode. If the active server is down, the passive one will
> automatically take its position (being the active node).
> One issue that must be remembered is that we use DRBD to replicate the
> data between the nodes, then we need to have a disk or partition exclusive
> to DRBD, which should be mounted to mysql directory /data01 , from our
> cluster setup 'Rancho' servers db701a is in active mode &701b is in
> passive mode(standby)
> root@db701a:~# mount
> /dev/drbd0 on /data01 type ext4 (rw)
> root@db701b:~# mount
> /dev/sda1 on /boot type ext4 (rw)
> If there is an failover in the active mode(i.e db701a) ,will automatically
> switches to db701b .then passive server will be active , sync will be from
> b to a.
> Important note :We need to leave one partition exclusive to DRBD, on
> active node of the server .
> Overview of MySQL with Clustering Application:
> DRDB (Distributed Replication Block Device) is one of the leading
> solution for Mysql when combined with Pacemaker , Corosync. DRBD
> synchronizes data at the block device (typically a spinning or solid state
> disk) – transparent to the application, database and even the file system.
> Any changes made on the active host are synchronously replicated to the
> standby host by DRBD, they are accessible for reads and writes on exactly
> the same on both the nodes , the two hosts (Active/Standby) are required in
> order to provide physical redundancy; for suppose if we are using virtual
> environments those 2 hosts should be on different physical machines, one
> important feature no shared storage is required at any point of time , the
> services will be active on one host and in standby mode on the other.
> Pacemaker and Corosync combine to provide the clustering layer that sits
> between the services & Hosts
> Pacemaker is responsible for starting and stopping services, ensuring that
> they are running on exactly one host at time.
> Corosync provides the messaging infrastructure between the nodes that
> enables Pacemaker to do its job,
> it also handles the nodes , within the cluster and informs Pacemaker for
> any changes required in case of failovers.
> · Building of HighAvailability clusters from commodity hardware,
> without the requirement for shared-storage.
> · Automatic failover and recovery for service continuity.
> · An end-to-end, integrated stack .
> · synchronous replication, to ensure failover between nodes
> without the risk of losing committed transactions
> Three important things to be noted
> · drdb should be mounted to active node
> · Making sure mysql data directory , & my.cnf files are pointed
> to specified location.
> root@db701a:~# ls -l /etc/mysql/my.cnf
> lrwxrwxrwx 1 root root 20 Mar 6 10:54 /etc/mysql/my.cnf ->
> root@db701a:~# ls -l /var/lib/mysql
> lrwxrwxrwx 1 mysql root 13 Mar 3 16:47 /var/lib/mysql -> /data01/mysql
> · Virtual IP should be pointed to active node first .
> Please share your some more points .
MySQL Product Marketing & Management
+44 (0)7765 898-605