Database migrations need to happen frequently to apply changes in database applications. Those migrations can be small, like adding new objects or changing some procedures or functions. Migrations can also be more extensive, like migrating large tables. In both cases, the database migrations need to be robust and repeatable if something goes wrong. In this PL/SQL Office Hours, we will explore practices to ensure high quality in your PL/SQL database migrations. We will have a look which elements make your database migrations repeatable and give an introduction on how to get started with automated database migrations if you have never done any before.

Highlights

00:00 Introduction from Steven
02:10 Overview of database migrations
06:24 challenges with deployments
10:18 Team performance and delivery
12:37 Characteristics of good migrations
17:49 Maximizing automation of deployments
21:50 Q&A deployment vs migration, release every day?, ebr
25:03 Code for repeatable and non-repeatable database objects
27:26 Migrating tables
36:05 Requirements for migration scripts
38:50 Use an existing logging framework!
40:13 Q&A: Recompile invalid objects after migration; 24x7 shop should use EBR; rely on flashback for undo; CI/D tools