Traefik provides a full-featured and efficient load balancer that allows you to expose applications to the network in a dynamic environment. It is highly flexible and can scale to handle massive amounts of network traffic. This expert guide demonstrates how it can also be configured to offer true high availability in on-premises environments.
This guide presents three scenarios that illustrate on-premises use cases for Traefik, with a proof-of-concept solution for each. The walk-through will set up multiple platforms, and each scenario will be more challenging than the last:
• Case 1: Active/Passive Nodes. A high-availability load balancer based on Traefik and keepalived in a bare metal world. This will consist of two-nodes (two VMs running on KVM) in an active/passive configuration. If one of the nodes fails, the other will take charge of handling the requests.
• Case 2: Kubernetes Ingress. Building on the first scenario, the same pair of load balancer nodes, only now deployed to manage ingress for Kubernetes. This scenario will showcase the distributed features of Traefik Enterprise, including distributed rate limiting and distributed certificate management.
• Case 3: Cloud-Scale Load Balancing. Finally, a robust load balancer capable of handling massive amounts of traffic, with multiple proxies taking advantage of networking protocols, most notably the border gateway protocol (BGP).