Skip to content
This repository has been archived by the owner on Jan 22, 2024. It is now read-only.

https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release' does not have a Release file #1009

Closed
dittothat opened this issue Jul 10, 2019 · 20 comments
Closed

https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release' does not have a Release file #1009

dittothat opened this issue Jul 10, 2019 · 20 comments

Comments

@dittothat
Copy link

dittothat commented Jul 10, 2019 •
edited

1. Issue or feature description

In nvidia/cuda:10.1-devel-ubuntu18.04 when I run apt-get update, I get the error:

Ign:2 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 InRelease
Err:3 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release
Could not handshake: The TLS connection was non-properly terminated. [IP: 10.41.13.6 3128]
Err:4 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release
Could not handshake: The TLS connection was non-properly terminated. [IP: 10.41.13.6 3128]

E: The repository 'https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release' does not have a Release file.
E: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release' does not have a Release file.

2. Steps to reproduce the issue

Build this dockerfile:
FROM nvidia/cuda:10.1-devel-ubuntu18.04
RUN apt-get update

I note that apt-get is functioning fine in a container I build and run using:
docker build .
FROM nvidia/cuda:9.1-devel-ubuntu16.04
RUN apt-get update
@demizer
Copy link

demizer commented Jul 10, 2019 •
edited

I have been building ubuntu images all morning without issue. Are you behind a proxy by chance? We've had users behind proxies that have had issues clear up a few hours later. The CDN can be sensitive at times.

@dittothat
Copy link
Author

Yes, I am behind a proxy. I will try again tomorrow. Thanks.

@dittothat
Copy link
Author

Yes, this was a proxy issue. All is well now. Thank you.

@zhanghm1995
Copy link

@dittothat I encoutered this problems as well, I wonder how to solve this if I behind the proxy?

@uonxhou
Copy link

uonxhou commented Jul 30, 2021

Have you resolved this problem?

@drjasonharrison
Copy link

drjasonharrison commented Mar 30, 2022 •
edited

How would we know if we're behind a proxy and it is causing the problem. Is there a curl or wget command that can be used to test the network connection to the PPA?

So this works (or fails) depending on the connection:

curl https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64

giving you:

<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
         "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
        <head>
                <title>404 - Not Found</title>
        </head>
        <body>
                <h1>404 - Not Found</h1>
                <script type="text/javascript" src="//wpc.75674.betacdn.net/0075674/www/ec_tpm_bcon.js"></script>
        </body>
</html>

the URL is from both the error message and from /etc/apt/sources.list.d/nvidia-ml.list within the base image.

@taoisu
Copy link

taoisu commented Mar 30, 2022

This issue happens again for me today. It seems something is wrong on the nvidia side, it was working yesterday.

The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64  Release' does not have a Release file

The repo returns 404

@jimsparkman
Copy link

See #1624

@taoisu
Copy link

taoisu commented Mar 30, 2022

Thanks for the link!

@gaitanignacio
Copy link

I'm getting the same issue right now on a docker build from an old Dockerfile that I saw working properly before!

Basically:

FROM 763104351884.dkr.ecr.us-east-1.amazonaws.com/pytorch-training:1.6.0-gpu-py36-cu110-ubuntu18.04

RUN apt-get update

E: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64  Release' does not have a Release file.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.

@pensapav
Copy link

Yup, same for us:

E: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64  Release' does not have a Release file.
The command '/bin/sh -c apt-get update && apt-get install -y         libceres1         libcgal13         libfreeimage3         libglew2.0         libopengl0' returned a non-zero code: 100
make: *** [Makefile:7: docker-build] Error 100

The whole https://developer.download.nvidia.com/compute/machine-learning/repos/ repository is returning a 404.

It seems that what is failing is an apt-get update call, so if we ignore that failure, then move onto our apt-get install flow things work just fine.

@seonho
Copy link

seonho commented Mar 31, 2022

Same for me.

See https://forums.developer.nvidia.com/t/failed-to-fetch-https-developer-download-nvidia-com-compute-machine-learning-repos-ubuntu1804-x86-64-packages-gz/156287/3

This was referenced Apr 4, 2022
Closed
Closed
Closed
@Goddard
Copy link

Goddard commented Jul 12, 2022

2022 still getting this issue

@0xleowang
Copy link

Starting having this issue since yesterday.

@AlexandreBrown
Copy link

AlexandreBrown commented Jul 12, 2022 •
edited

We started hitting this issue today.

I reported the issue on NVIDIA forum https://forums.developer.nvidia.com/t/error-during-apt-update-on-ubuntu-18-04-file-has-unexpected-size/220436

@Hariraj
Copy link

Hariraj commented Jul 12, 2022

We too started getting this issue today.

@AlexandreBrown
Copy link

AlexandreBrown commented Jul 12, 2022 •
edited

I found a fix that works (at least for me under ubuntu 18.04), see https://forums.developer.nvidia.com/t/error-during-apt-update-on-ubuntu-18-04-file-has-unexpected-size/220436/2

The fix for me was to add rm /etc/apt/sources.list.d/cuda.list before any apt command.

@muzuiget
Copy link

2022 and a half still getting this issue

E: Failed to fetch https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/Packages.gz  File has unexpected size (580845 != 580040). Mirror sync in progress?

@figkim
Copy link

figkim commented Jul 12, 2022 •
edited

In my case, add RUN rm /etc/apt/sources.list.d/cuda.list before apt-get update solves the issue temporarily while using dockerfile. (ubuntu2004)

@jesusa-nv
Copy link

Hi folks,

Sorry about these constant CDN troubles. We are working to resolve these issues, but due to the nature of the CDN and our workflows, it will take some time.

In the future, please report issues with the CUDA repo to https://github.com/NVIDIA/cuda-repo-management.

Thanks!

@NVIDIA NVIDIA locked as resolved and limited conversation to collaborators Jul 12, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests