CUDA Docker image used in devcontainer is deprecated

[robrussell]

After rebuilding with the devcontainer this morning it emits this log message at startup:

Container started

==========
== CUDA ==
==========

CUDA Version 10.2.89

Container image Copyright (c) 2016-2022, NVIDIA CORPORATION & AFFILIATES. All rights reserved.

This container image and its contents are governed by the NVIDIA Deep Learning Container License.
By pulling and using the container, you accept the terms and conditions of this license:
https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license

A copy of this license is made available in this container at /NGC-DL-CONTAINER-LICENSE for your convenience.

*************************
** DEPRECATION NOTICE! **
*************************
THIS IMAGE IS DEPRECATED and is scheduled for DELETION.
    https://gitlab.com/nvidia/container-images/cuda/blob/master/doc/support-policy.md

It seems like this is coming from the nvidia/cuda:10.2-devel-ubuntu18.04 base image but I'm not certain.

I haven't tried changing the base image but would how hard would it be to jump to an Ubuntu 22.04 image with CUDA 12?

[blackhawk-17]

Facing the same issue with nvidia/cuda:12.0.0-cudnn8-devel-ubuntu22.04 . It goes to crashloopbackoff once deployed. Can you please mention how you solved this?

[robrussell]

Unfortunately I never found a solution with a devcontainer. Instead I now rely on the prebuilt binary releases for my application.

[zeke-john]

Any update? I'm also using the 12.0.0-cudnn8-devel-ubuntu20.04 image and am getting the same error

[MagneticGlitter]

Hi all, I'm new to the instant-ngp, and I'm having this error after i build the image and run cmake . -B build -DCMAKE_BUILD_TYPE=RelWithDebInfo . So I'm wondering if I could get some help from someone.