Skip to content

Implement ICE server probing to prevent webrtc service downtime #35

Open
@ivelin

Description

@ivelin

Currently aiortc does not implement ICETrickle nor is it able to fall back from a failing primary STUN or TURN server to a fall back option provided in the API standard RTCConfiguration list.

As a workaround, we can implement a background thread that periodically (e.g. every minute) polls the servers in the RTCConfig list and removes the dead ones. Respectively as needed, signal the peerjs code to reconnect via healthy servers.

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or requesthelp wantedExtra attention is needed

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions