Member-only story

Simplified Vertica Copy Cluster Failover

George Seah
5 min readJul 9, 2020

--

After working with Vertica database for about a year, I came to realize that there is a backup method called “copycluster”.
Don’t shame me please!

Photo by Abigail Keenan on Unsplash

Reference: https://www.vertica.com/docs/9.2.x/HTML/Content/Authoring/AdministratorsGuide/BackupRestore/SampleConfigFiles/copycluster.htm

Basically, what it does is to copy the entire Vertica contents into another cold backup Vertica server. The setup requirement was 1 active Vertica server and 2 backup Vertica servers.

The scripts provided by Vertica was useful, but to execute a failover manually, there had to be a lot of steps to be done. Thus, I have created a simple shell script to automate the failover process (Technically, just a script calling other scripts.)

Photo by Markus Winkler on Unsplash

The goal is simple — to get the user to perform the failover with as little effort as possible. This is to minimize any human errors and also reduce the dependency required from my side.

However, I will not be able to install external packages to simplify this as we are using CentOS minimal image. Everything has to be done within a shell script.

--

--

No responses yet