Chapter 45. Oh, No: No Logs
Laura Santamaria
Logging is the least glamorous part of any server, much less any cloud-based system. However, it’s one of the most critical processes on a server for operations teams, whether they’re responding to an incident or monitoring a system for anomalies. Great logs are even more crucial for anyone working with a distributed, cloud-based system that has not been properly maintained—the dreaded legacy system. However, what do you do when you don’t have logs?
If you don’t have logs, don’t panic. There are many ways to debug a system without logs. It just requires ingenuity and patience, along with creative thinking. If you panic before you can get too far, you will narrow your mental pathways and take up precious thinking space worrying about your boss breathing down your neck. Take a deep breath and sit down to think.
First, try hitting any available endpoints or services. Whether it’s an API call through curl or a ping to a server, you will get something you can work with, even if it’s no response at all. For example, a 403 response from an API with known good credentials indicates an error with authentication, perhaps from a compromised authentication system or someone updating a database without following a procedure your company has established. No response from a server or an API tells you to take a closer look ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access