Error: Failed to fetch completions: Error processing prompt (see logs with DEBUG>=2): Wait timeout: 10000 ms! (the signal is not set to 13884, but 13842)

[svm87601]

I have deployed the exo platform on two Linux servers, both equipped with RTX 3090 GPUs. I successfully ran the web service and achieved distributed inference, but I encountered an error after the third inference attempt. My system is Ubuntu 22.04, with CUDA version 12.5 and driver version 555.42.02. I would greatly appreciate any solutions you could provide.