A100 not recognized on GCP instance

[davide445]

Could not recognize one A100 card of two cards by nvidia-smi

Hi NV Team, Please give me advice to resolve the issue. Issue: Suddenly our server did not recognized one A100 cards of two ones by nvidia-smi after restarting server. Note: lspic command showed the two cards like below. nvidia@a100-server-b:~$ lspci | grep NVIDIA 24:00.0 3D controller: NVIDIA Corporation Device 20f1 (rev a1) e1:00.0 3D controller: NVIDIA Corporation Device 20f1 (rev a1) nvidia@a100-server-b:~$ nvidia-smi Tue Jun 15 01:34:28 2021 +-----------------------------------…

@Kaka_m we have the same problem on a Ubuntu 18.04 GCP a2-highgpu-2g VM instance, my I ask how you solved this.