It’s great that your setup was stable before, so the issue might be due to some intermittent network hiccups, resource limits, or small configuration changes. Let’s work together to get this sorted out! Here are some steps you can try to help pinpoint and resolve the problem:
1.
Verify Network Stability: Take a moment to check if there have been any network interruptions or changes in firewall rules that could be affecting WebRTC traffic. Also, make sure TURN over TLS is enabled, as it really helps in restrictive network environments.
2.
Inspect MCP Server Logs: Have a look at the MCP server logs around the times when failures happen. Sometimes, those logs can reveal useful clues or errors.
3.
Enable Diagnostic Logs: Try turning on detailed logging in the LiveKit SDK. This will give you insights into connection states, ICE candidate gathering, and reconnection attempts, which can be super helpful to find the root cause.
4.
Review Resource Usage: It’s important to check that your MCP server and LiveKit deployment have enough resources like CPU and memory. If they’re running high, it might cause timeouts or disconnects.
5.
Check SDK and Server Compatibility: Double-check that your LiveKit Agents SDK (1.0.12) and MCP server (1.9.2) versions are compatible and up to date to avoid any unexpected issues.
If you’re still running into trouble, please feel free to share any extra logs or details about how the failures happen. I’m here to help and we can explore more troubleshooting steps together!
Sources:
SIP troubleshooting guide | LiveKit Docs |
Getting Help | livekit/livekit_composite | DeepWiki |
Connecting to LiveKit | LiveKit Docs |
src/room/RTCEngine.ts |
livekit/src/rtc_engine/rtc_session.rs