webRTC WHIP/WHEP + TURN extreme packet loss
Greetings,
Running a webRTC broadcast from host over WHIP peer connection.
Utilizing a TURN server.
All viewers, (using WHEP peer connection) of this stream are experiencing heavy / frequent packet loss to the point where the broadcast is almost unbearable to watch. Audio does not lag but video will frequently. Almost a 1 second pause 4 seconds.
Current scenario is with 125 concurrent viewers.
In monitoring the webRTC Internals system, I see frequent packet loss for the inbound rtp connection.
This issue with packet loss comes and goes between other streams. Sometimes it doesnt occur. Other times, it's constant. It appears to be happening at the point of delivery.
Can anyone else report similar issues in your instances? Is there anything that can be done to prevent or improve these issues?
2 Replies
I see this is an "ask and pray" type of help desk situation here. Is there no CloudFlare stream dev / support team that views and engages with these topics to assist developers experiencing issues with the service? If I cannot get clarity on the issue from a backend team, I will need to seek alternative service providers who support their streaming service more attentively. I've already opened a ticket and forum post. No response after weeks. 🤷♂️
Okay, and your recommendation for getting in contact with a CloudFlare support team member is what exactly?
if you want better response times get a biz or ent plan :) as floppydisk already mentioned this discord is a community forum, help arrives when it arrives (...)