2.7

Table Of Contents
vPostgres Database Replication
You can replicate vPostgres databases to improve reliability, fault tolerance, and accessibility if a database
instance fails, or otherwise becomes unavailable.
The vPostgres database within Data Director uses Write-Ahead Logging (WAL) to replicate data between
master and slave instances. WAL is a standard approach to transaction logging in which changes to data files
are written to only after those changes have been logged. WAL allows for a significant reduction in the number
of disk writes because only the log file needs to be flushed to disk at the time of transaction committal, rather
than every data file changed by the transaction. Using WAL to replicate data is the fastest type of replication
available, allowing you to quickly create slave instances of vPostgres databases.
You can replicate a vPostgres database without any performance penalty to clients accessing the master
instance. However, when you failover to a slave instance, all connections to the database are terminated.
Prerequisites
n
Verify that you have provisioned a vPostgres database within Data Director to replicate.
n
Create slave instances on different datastores assigned to different database groups.
To ensure fault tolerance, replicate slave instances on different datastores than that of the master instance.
Create slave instances in different database groups, where each group is assigned to a different resource
pool to ensure that slave instances have their own resources with which to operate in the event of a system
failure.
Procedure
1 Replicate a vPostgres Database on page 164
You can replicate a vPostgres database to provide an added degree of fault tolerance and redundancy
to your database environment.
2 Failover to a Standby vPostgres Database on page 165
You can manually failover by promoting a slave (or standby) database to take over from its master (or
primary) database. When you promote a slave database to take over from its master database, all other
slave instances are redirected to use the newly created master instance.
Replicate a vPostgres Database
You can replicate a vPostgres database to provide an added degree of fault tolerance and redundancy to your
database environment.
vPostgres replication requires that you create a slave database from the master database. The slave database
is a full clone of its master database.
Prerequisites
n
Provision a vPostgres master (or primary) database within Data Director from which to replicate slave (or
standby) instances.
n
Verify that the vSphere deployment hosting Data Director has two or more datastores on which to create
master and slave databases. The datastores must be on different physical hosts to ensure fault tolerance
in the event of physical hardware failure.
Procedure
1 Log in to Data Director as a system administrator.
2 Click the Organization tab, and click Manage and Monitor.
VMware vFabric Data Director Administrator and User Guide
164 VMware, Inc.