The error signal_event taking too much time: Answer(SessionDescription ...) indicates that your LiveKit worker is taking longer than 10 seconds to handle the signaling event for processing the participant's answer in the WebRTC negotiation. This usually points to a performance bottleneck or contention within the worker process, or potentially a synchronization/blocking issue in your environment. Recommended troubleshooting steps: • Review system resource usage (CPU, memory) on the worker machine during these events. • Check the logs for additional errors and warnings immediately before and after the timeout; look for repeated patterns, or "failed to handle signal" errors. • Isolate whether this is client-specific, server-side, or load-related by noting if it happens under certain conditions (e.g., number of participants, room size, etc.). • Verify all LiveKit components are using compatible versions. • Monitor the timing and flow of the offer/answer exchange and check for network or signaling delays. This behavior commonly results from system resource contention, negotiation queue state issues, or thread pool/resource exhaustion. If you are able to share more about recent changes in your environment (e.g. updates, configuration changes, workload change), or if you have observed a pattern tied to particular clients or timings, this will help further pinpoint the cause. Sources: livekit/src/room/mod.rs | Community & Support | livekit/livekit_composite | DeepWiki