English

I’m Herbert, lead developer of LibreQoS. I joined not long after Robert and Dave got the ball rolling, added IPv6 support, and created most of the Rust parts, C parts, etc.

One thing we’ve heard from people is that Insight looks great, but they ask, “How do I use this thing?” and, more importantly, “How does this help me?” We use Insight daily; it’s become our one-stop shop for finding network issues and understanding what’s going on. However, we’re all developers and have decades of experience as internet service providers. So we set out to demystify the experience.

A huge thank you to NLnet foundation for sponsoring this development.

We just launched version 0 (Alpha) of the Insight API. LibreQoS shapers have an API, too — and that’ll be the topic of future posts. This one is all about the Insight API.

One of the first things many people notice when they launch LibreQoS Insight is that our bandwidth graphs look different. For example:

Spot congestion, latency spikes, and retransmissions before tickets arrive.

Prior to version 1.5-RC1, LibreQoS read ShapedDevices.csv and network.json (in the LibreQoS.py script) to determine network hierarchy, device configuration and design the traffic control (TC) setup. Once the files were validated, the script would create a big set of tc commands, clear the current TC state, and apply the new configuration.