DevConf.US 2022

One of the top priorities in every company nowadays is to ensure High Availability(HA) and Disaster Recovery in their services. Despite this emphasis, you can hear news almost every single day about businesses losing part of their revenue due to inoperative applications or outages! While moving applications to the cloud helps improve availability, additional modifications are needed to take full advantage of the cloud. If you are interested in learning more about how to support High Availability in your applications, join us in this talk as we walk through the process the Open Data Hub team went through to implement HA for the JupyterHub. We will also explain how to implement load balancing with Traefik for networking, and how to ensure only one HA instance is “actively” running via leader election. Finally, we will discuss some common traps as well as lessons learned.