Development environment

[anfiel]

[+] Building 31.7s (2/2) FINISHED docker:default
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 4.35kB 0.0s
=> ERROR [internal] load metadata for docker.io/nvidia/cudagl:11.4.2-de 31.7s

[internal] load metadata for docker.io/nvidia/cudagl:11.4.2-devel-ubuntu18.04:

---

Dockerfile:12

10 | # https://hub.docker.com/r/nvidia/cudagl/
11 | # https://github.com/NVIDIA/nvidia-docker/wiki/CUDA
12 | >>> FROM nvidia/cudagl:11.4.2-devel-ubuntu18.04
13 |
14 | # setup timezone

ERROR: failed to solve: nvidia/cudagl:11.4.2-devel-ubuntu18.04: failed to resolve source metadata for docker.io/nvidia/cudagl:11.4.2-devel-ubuntu18.04: failed to copy: httpReadSeeker: failed open: failed to do request: Get "https://production.cloudflare.docker.com/registry-v2/docker/registry/v2/blobs/sha256/bc/bc4eea507469a745615c69a35a7f1c14978366be1066714aadfbb834be098bbf/data?verify=1718679139-aE%2Bztz17t1cRMJKtbuHvhJrzhZo%3D": dial tcp [2a03:2880:f10e:83:face:b00c:0:25de]:443: i/o timeout

[peci1]

Please, provide more context and describe what you tried to do. The report is not useful in its current state and there's nothing we can do for you unless you provide more information.

[anfiel]

My host computer is ubuntu20.04 and cuda is 11.8. When I use this command, the above error is displayed. When I change the cuda version in dockerfile, the error is still displayed. Why is this?

[peci1]

Thanks for the update. Could you check if #17 resolves the issue for you?

[anfiel]

It seems like the same, maybe Docker.io/nvidia/cudagl:10.1-devel-ubuntu18.04. Is there a problem here?

[peci1]

No, there's not in general. I succeeded building it from scratch yesterday. Can you try over a VPN or a different ISP's network?

Or, as I see you access it via IPv6, try disabling IPv6 in your network connection settings.

[anfiel]

Every time ./build_common_image.bash is run, an error is reported when it reaches the 1178th line, as shown in the following figure

[peci1]

Your networking seems wrong. Look at the Bad Gateway [IP 127.0.0.1]. Not sure what does this, but it is almost 100% wrong. Check your /etc/hosts.

[anfiel]

Ok, thanks for your reply, I'll check it carefully.