Re: [webrtc-stats] What does a "transport" consider as "payload"? (#507)

> This becomes complex due to the way RTCP is structured.

Aye! Spec looks mostly good on that and libWebRTC does not count RTCP ✅ 
The transport *might* count them - looks like it already does but how would one tell them apart from SCTP packets?
Use-case would be monitoring the 5% rule - you know how big those 😱 feedback packets get

> (Really happy to see transport + candidate-pair add up exactly in this case, though!)

But that is *not* what you wanted [a while back](https://github.com/w3c/webrtc-stats/issues/507#issuecomment-570494862)?
Looking [here](https://source.chromium.org/chromium/chromium/src/+/main:third_party/webrtc/p2p/base/connection.cc;l=1697;drc=533610f3b0cddaf48453f570591a39acc7b8042a;bpv=1;bpt=1) I would say [this](https://source.chromium.org/chromium/chromium/src/+/main:third_party/webrtc/p2p/base/connection.cc;l=454;drc=533610f3b0cddaf48453f570591a39acc7b8042a;bpv=0;bpt=1)  does not count STUN packets.

The more interesting case is what happens if you have two connections (pairs) and switch between them. ✅ for libWebRTC, partial happyness?

-- 
GitHub Notification of comment by fippo
Please view or discuss this issue at https://github.com/w3c/webrtc-stats/issues/507#issuecomment-2727012758 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Saturday, 15 March 2025 21:43:27 UTC