NVRM: RmInitAdapter failed! missing one gpu

I have 8 gpus but the nvidia-smi commands only shows 7. NVRM: RmInitAdapter failed, message appears in the log report. Tried to reinstall the driver and reboot. Is there anything else I can do to fix this issue?nvidia-bug-report.log (8.1 MB)

That’s a hardware issue, resources are properly assigned so the gpu is either broken or improperly seated or dust on the slot. Try reseating it in its pcie slot. If still doesn’t work, have it replaced.

BTW, the logs are flooded with error messages from the intel ethernet adapter:

ixgbe 0000:01:00.1: Warning firmware error detected FWSM: 0x0118801F

you should look into that.

For anyone else who comes here because of the ixgbe error, the fix is to update the firmware on the Ethernet adapter. At least in my case the exact error message came from an X550 adapter, which is updated using the “Non-Volatile Memory (NVM) Update Utility for Intel® Ethernet Network Adapter X550 Series” software. Good luck!