Book Image

PostgreSQL High Availability Cookbook - Second Edition

By : Shaun Thomas
Book Image

PostgreSQL High Availability Cookbook - Second Edition

By: Shaun Thomas

Overview of this book

Databases are nothing without the data they store. In the event of a failure - catastrophic or otherwise - immediate recovery is essential. By carefully combining multiple servers, it’s even possible to hide the fact a failure occurred at all. From hardware selection to software stacks and horizontal scalability, this book will help you build a versatile PostgreSQL cluster that will survive crashes, resist data corruption, and grow smoothly with customer demand. It all begins with hardware selection for the skeleton of an efficient PostgreSQL database cluster. Then it’s on to preventing downtime as well as troubleshooting some real life problems that administrators commonly face. Next, we add database monitoring to the stack, using collectd, Nagios, and Graphite. And no stack is complete without replication using multiple internal and external tools, including the newly released pglogical extension. Pacemaker or Raft consensus tools are the final piece to grant the cluster the ability to heal itself. We even round off by tackling the complex problem of data scalability. This book exploits many new features introduced in PostgreSQL 9.6 to make the database more efficient and adaptive, and most importantly, keep it running.
Table of Contents (18 chapters)
Title Page
About the Author
About the Reviewer
Customer Feedback

Using an outage to test migration

While planned migrations are always preferred, sometimes hardware failures or server instability will introduce an aspect of surprise. If we had not used Pacemaker, a server crash would be a catastrophic event. Even if we had followed every chapter in this book this far and had Nagios and e-mail alerts galore, a DBA would need to be available to activate the alternate node.

If an outage occurred at night when everyone was sleeping, we would be faced with a worst-case scenario. Necessary personnel might not hear the alert for several minutes, and more time is lost on triage and activation steps. Such an outage could extend from a few minutes to over an hour. So much for our high availability!

Yet, at this point, we don't know if Pacemaker would negate the previous scenario. While we've tested how Pacemaker handles an expected and safe migration, what happens when a node disappears entirely? Will Pacemaker cover us in the event there is an outage when nobody...