Book Image

Learn PostgreSQL

By : Luca Ferrari, Enrico Pirozzi
Book Image

Learn PostgreSQL

By: Luca Ferrari, Enrico Pirozzi

Overview of this book

PostgreSQL is one of the fastest-growing open source object-relational database management systems (DBMS) in the world. As well as being easy to use, it’s scalable and highly efficient. In this book, you’ll explore PostgreSQL 12 and 13 and learn how to build database solutions using it. Complete with hands-on tutorials, this guide will teach you how to achieve the right database design required for a reliable environment. You'll learn how to install and configure a PostgreSQL server and even manage users and connections. The book then progresses to key concepts of relational databases, before taking you through the Data Definition Language (DDL) and commonly used DDL commands. To build on your skills, you’ll understand how to interact with the live cluster, create database objects, and use tools to connect to the live cluster. You’ll then get to grips with creating tables, building indexes, and designing your database schema. Later, you'll explore the Data Manipulation Language (DML) and server-side programming capabilities of PostgreSQL using PL/pgSQL, before learning how to monitor, test, and troubleshoot your database application to ensure high-performance and reliability. By the end of this book, you'll be well-versed with the Postgres database and be able to set up your own PostgreSQL instance and use it to build robust solutions.
Table of Contents (27 chapters)
Section 1: Getting Started
Section 2: Interacting with the Database
Section 3: Administering the Cluster
Section 4: Replication
Section 5: The PostegreSQL Ecosystem

Implementing auditing

Auditing is the capability of performing introspection over an application or user session, in other words, to be able to reproduce, step by step, what the user or the application asked the cluster to do.

Auditing is slightly different from logging, as logging provides a simple way of saving whatever action of the user, but without providing an easy way to reconstruct the user or application interactions with the cluster. In fact, in a highly concurrent cluster, many actions made by different users will coexist in the logs in a mixed bunch of lines. Moreover, logging does not provide any particular logic on what it is storing, and therefore it becomes hard to find out what a user has done. This becomes even more true when the user or the application executes complex statements, in particular, statements where parameters and values are not explicitly provided.

As an example, consider the following simple section:

forumdb=> PREPARE my_query( text ) AS SELECT * FROM...