No heartbeat for components

[DHrobotman]

[2023-09-18 21:35:46,519][196071] No heartbeat for components :InferenceWorker_p0-w0 (337 seconds)
[2023-09-18 21:35:46,519][196071] Stopping training due to lack of heartbeats from <class 'sample_factory.algo.sampling.inference_worker.InferenceWorker‘>

When I start train.py, an error will occur. So how can I solve the problem and why does it happen? I hope you can help me answer this question

[alex-petrenko]

Hi @DHrobotman

This error typically means something happened earlier, i.e. one of the system components crashed. Possible reasons: bug in the code, OOM, etc.
Try to look at the entire log file and search for Exceptions and errors.

It would also help if you could attach your log file and configuration here.