In production currently, there are 2 flavors of LiveKit: Cloud and Hosted service.
This still represents a weak failure point if the LiveKit cloud stops working. A disaster recovery process or a fallback mechanism should be implemented to mitigate this. For the time being, this can be accomplished by changing the branch to comms-v2 and starting the node on P2P comms but this requires maintaining 2 branches of the catalyst-owner repo.
In production currently, there are 2 flavors of LiveKit: Cloud and Hosted service.
This still represents a weak failure point if the LiveKit cloud stops working. A disaster recovery process or a fallback mechanism should be implemented to mitigate this. For the time being, this can be accomplished by changing the branch to comms-v2 and starting the node on P2P comms but this requires maintaining 2 branches of the catalyst-owner repo.