nvidia-smi hangs indefinitely after ~66 days 12 hours uptime with driver 570.133.20 OpenRM on B200 and kernel 6.6.0
Summary
This GitHub issue documents a stability bug where nvidia-smi hangs after about 66 days of uptime on a system with driver 570.133.20 OpenRM on a B200 GPU and kernel 6.6.0. The post provides system details, dmesg outputs, and reproduction steps, making it a valuable primary source for hardware/driver reliability discussions relevant to long-running compute nodes.