460 driver installation on Tesla 2070

Dear Nvidia team, we have :

  1. Installed Nvidia Tesla driver for Centos 8 according to: NVIDIA Driver Installation Quickstart Guide :: NVIDIA Tesla Documentation
  2. Installed cuda 11.2 according to redhat8 cuda installation : https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#redhat8-installation

nvidia-smi returns the following message:

NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.

dkms status returns
nvidia, 460.32.03, 4.18.0-240.10.1.el8_3.x86_64, x86_64: installed

lspci -v | grep VGA. returns

04:00.0 VGA compatible controller: NVIDIA Corporation GF100GL [Tesla C2050 / C2070] (rev a3) (prog-if 00 [ VGA controller])

05:01.0 VGA compatible controller: Matrox Electronics Systems Ltd. MGA G200eW WPCM450 (rev 0a) (prog-if 00 [ VGA controller])

The system is without display

The Nvidia bug report gives in some places strange message like: The NVIDIA Tesla C2070 GPU installed in this system is supported through the NVIDIA 390.xx Legacy drivers. … The 455.32.00 NVIDIA driver will ignore …
(see full file attached)
nvidia-bug-report.log.gz (92.6 KB)

Please help.

Thank you very much in advance,
Best regards,
Evgeniya

This is an old Fermi gen Tesla, only supported up to driver 390 and cuda 8.