https://livekit.io logo
Join Slack
Powered by
> {"level":30,"time":1766007044459,"pid":138,"h...
# ask-ai
r
Copy code
{
  "level": 30,
  "time": 1766007044459,
  "pid": 138,
  "hostname": "deployment-p-551rhlf5aqv-ca-ubuzhpvxljvl-5645d59794-kr72n",
  "version": "0.1.0",
  "jobId": "AJ_BfWUWeLLhUBF",
  "resuming": false,
  "agentName": "voice-agent",
  "msg": "received job request"
}
{"level":40,"time":1766007044486,"pid":138,"hostname":"deployment-p-551rhlf5aqv-ca-ubuzhpvxljvl-5645d59794-kr72n","err":{"type":"Error","message":"Channel closed","stack":"Error [ERR_IPC_CHANNEL_CLOSED]: Channel closed\n at target.send (nodeinternal/child process753:16)\n at JobProcExecutor.launchJob (file:///app/node_modules/.pnpm/@livekit+agents@1.0.25_@livekit+rtc-node@0.13.22_zod@3.25.76/node_modules/@livekit/agents/dist/ipc/job_proc_executor.js7815)\n at ProcPool.launchJob (file:///app/node_modules/.pnpm/@livekit+agents@1.0.25_@livekit+rtc-node@0.13.22_zod@3.25.76/node_modules/@livekit/agents/dist/ipc/proc_pool.js6116)\n at process.processTicksAndRejections (nodeinternal/process/task queues105:5)\n at async JobRequest.onAccept (file:///app/node_modules/.pnpm/@livekit+agents@1.0.25_@livekit+rtc-node@0.13.22_zod@3.25.76/node_modules/@livekit/agents/dist/worker.js5779)","code":"ERR_IPC_CHANNEL_CLOSED"},"msg":"job process exited unexpectedly; this likely means the error above caused a crash"} -> what is this caused by
t
Thanks for reaching out, Bobvi. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.
You're seeing Node's 
Error [ERR_IPC_CHANNEL_CLOSED]: Channel closed
because the parent agent process attempted to send a message over the IPC channel after the child job process had already terminated. In your log the parent tried to 
launchJob
and the job process exited unexpectedly, which closed the IPC channel and produced that error. Likely root causes to investigate: • An uncaught exception or crash in the agent's entrypoint or startup code (check the child's stdout/stderr or crash logs). • A prewarm function or required resource failing during initialization. • The child process being killed by the OS (OOM or an external kill). • An issue during 
@livekit/agents
process initialization. Suggested next steps: 1. Check the child job process logs (stdout/stderr) around the same timestamp for an exception or crash reason. 2. Verify any prewarm code or model/resource loads run during startup — wrap them with try/catch and log errors. 3. Monitor system resource usage (memory/CPU) for OOM or kills and check the host/kernel message log for SIGKILL/OOM entries. 4. If you can reproduce, add early logging in the agent entrypoint to show progress through initialization steps so you can see where it exits. If you want, share the child process logs or the agent startup code snippet and I’ll help pinpoint the likely failing step. Sources: Agent server | LiveKit docs | Job lifecycle | LiveKit docs | Agent server options | LiveKit docs
2 Views
Open in Slack
PreviousNext
Powered by