Miloš Milunović

Backing up your AKS cluster. Because you have to!

It's tough enough trying to set up a release process for a single application, right?
What if someone gave you a task to automate integration of dozens of microservices written by multiple companies from development environment all the way to production?
... oh, and they told you that you are responsible for keeping the production up and running whatever happens ofc!
Backup of AKS cluster is a topic little referenced but much needed in a productive environment. In this session I want to go through the process of achieving a full AKS cluster backup and restore.
Full. Everything.
Kubernetes services, Managed cluster, azure resources. And ofc share some of the best practices that work for us at least!
Topics that will be covered:
Cluster backup automation with Velero
How to handle state?
AKS infrastructure backup (and how to structure your backup)
Backup to another region
Disaster recovery scenarios

Let's automate everything!!!

From dev to prod. Two years of using Kubernetes in production

It’s tough enough trying to set up a release process for a single application, right? What if someone gave you a task to automate the integration of dozens of microservices written by multiple companies from development environment all the way to production? … oh, and they told you that you are responsible for keeping the production up and running whatever happens ofc!

Imagine you have 5 Kubernetes environments used by developers from multiple companies and users alike.
How would you handle releases? How to aggregate logs? Which monitoring tools to use and how to setup your monitoring? How to give access to developers to debug their applications without endangering security? How to manage configurations of different environments without much added complexity? How to be invisible for users and developers but still satisfy both user groups of your environments? How to do security and QA?
In this talk I will share some of the knowledge and experiences we have acquired while running Kubernetes in production.
Topics that will be covered:
Release management with Helm, logging with EFK stack, monitoring setup, tracing guidelines and tools, security scans, cluster backups and restore procedures…

This talk was first time presented @Heapcon conference in Belgrade.
Link: https://www.youtube.com/watch?v=EZphST_zZUc&t=1s
Target audience intermediate Kubernetes users.