As of its latest release, Instana is introducing support for Traefik, an open source cloud-native reverse proxy and ingress controller originally developed by Traefik Labs (formerly Containous) in 2016. This closes an important gap in the observability puzzle.
What is a reverse proxy and what’s reversed about it?
Before we talk about Traefik, we should probably talk about what a reverse proxy is in the first place. A reverse proxy is a server that sits in front of web servers and forwards service to those servers; they’re kind of like traffic directors for computers. Reverse proxies can be used for a number of different reasons, but mostly they serve to increase reliability and performance or security. A reverse proxy is different from a traditional forward proxy because it sits in front of web servers and intercepts and directs ingress traffic coming from clients where a forward proxy sits in front of clients to direct information coming from servers. For further clarification, check out the diagram below:
Okay, so why Traefik, specifically?
What makes Traefik stand out to us from other reverse proxies such as nginx is that it can handle a ton of different upstreams with easy and straight-forward configuration. When Traefik inspects a user’s infrastructure, it automatically finds relevant information and figures out which server processes serve requests based on container labels, Kubernetes meta information and explicit configuration. Traefik is also natively compliant with modern infrastructure, whether it be Kubernetes, Docker, AWS, Mesos, or just about anything else you could think of.
Traefik’s main goal is to make Traefik simple and easy to use (see the synergy with Instana here?). There is no need to keep or maintain a separate configuration file for your proxy because everything happens automatically in real time. When we rolled out Traefik support, we expected to find some Traefik processes in our customer landscapes; and after analyzing the data, we found more than 500 unique deployments of Traefik in different scenarios.
Traefik is at the core of customer infrastructure
Traefik represents an invaluable piece of the observability picture for Instana and its customers. By providing visibility into the routing of traffic between workloads and pods in a Kubernetes setting, Traefik is closing an important gap in terms of visibility that enables Instana’s users to identify bottlenecks down to the process level. It’s a Traefik ingress controller in a Kubernetes cluster. With Instana’s automatic monitoring of Traefik processes combined with tracing, every single call to an application will natively be forwarded by reverse proxy capabilities – and Instana’s users can identify the complete physical context of the calls; including the pod and the node.
Other use cases for Traefik involve overhead detection. For example, maybe you have some middleware in your stack that adds too much latency on top of a single transaction. With Traefik and Instana, you can identify the exact overhead in terms of timing and tweak your configuration accordingly.
Either way, Instana has got you covered. Find out more about Traefik in Instana by checking our documentation at https://www.instana.com/docs/ecosystem/traefik/.