Nvidia-docker: invalid: BADSIG F60F4B3D7FA2AF80 cudatools <[email protected]>
I got the same problem as #571 and #613. When I use the image nvidia/Cuda:10.0-runtime-ubuntu18.04. I ran apt update when building an image. Here are the outputs:
Get:1 http://archive.ubuntu.com/ubuntu bionic InRelease [242 kB]
Get:2 http://security.ubuntu.com/ubuntu bionic-security InRelease [88.7 kB]
Get:5 http://security.ubuntu.com/ubuntu bionic-security/multiverse amd64 Packages [4171 B]
Get:6 http://security.ubuntu.com/ubuntu bionic-security/universe amd64 Packages [304 kB]
Ign:3 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x86_64 InRelease
Get:7 http://archive.ubuntu.com/ubuntu bionic-updates InRelease [88.7 kB]
Get:9 http://security.ubuntu.com/ubuntu bionic-security/restricted amd64 Packages [5436 B]
Get:10 http://security.ubuntu.com/ubuntu bionic-security/main amd64 Packages [424 kB]
Ign:4 https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 InRelease
Get:12 http://archive.ubuntu.com/ubuntu bionic-backports InRelease [74.6 kB]
Get:13 http://archive.ubuntu.com/ubuntu bionic/universe amd64 Packages [11.3 MB]
Get:8 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x86_64 Release [564 B]
Get:11 https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 Release [564 B]
Get:14 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x86_64 Release.gpg [819 B]
Get:15 https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 Release.gpg [801 B]
Ign:15 https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 Release.gpg
Get:16 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x86_64 Packages [71.4 kB]
Get:17 http://archive.ubuntu.com/ubuntu bionic/multiverse amd64 Packages [186 kB]
Get:18 http://archive.ubuntu.com/ubuntu bionic/main amd64 Packages [1344 kB]
Get:19 http://archive.ubuntu.com/ubuntu bionic/restricted amd64 Packages [13.5 kB]
Get:20 http://archive.ubuntu.com/ubuntu bionic-updates/restricted amd64 Packages [10.8 kB]
Get:21 http://archive.ubuntu.com/ubuntu bionic-updates/main amd64 Packages [776 kB]
Get:22 http://archive.ubuntu.com/ubuntu bionic-updates/universe amd64 Packages [1190 kB]
Get:23 http://archive.ubuntu.com/ubuntu bionic-updates/multiverse amd64 Packages [7238 B]
Get:24 http://archive.ubuntu.com/ubuntu bionic-backports/universe amd64 Packages [3652 B]
Get:25 http://archive.ubuntu.com/ubuntu bionic-backports/main amd64 Packages [942 B]
Reading package lists...
W: GPG error: https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools <[email protected]>
E: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release' is not signed.
I think it is a CDN issue because it redirected to developer.download.nvidia.cn.
Qinka
All 31 comments
Hello!
Which region are you from? I can't seem to reproduce your error :)
RenaudWasTaken
on 6 May 2019
@RenaudWasTaken China mainland.
After this issue opened, the error seemed fixed, but sometimes it's failed again.
I think this issue can be closed since it is available most times.
Qinka
on 6 May 2019
I'm also receiving this error from UK
bencevans
on 17 May 2019
I'm also receiving this error from Indonesia.
W: GPG error: http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools <[email protected]>
E: The repository 'http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64 Release' is not signed.
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.
sliawatimena
on 28 May 2019
same issue
Get:21 http://archive.ubuntu.com/ubuntu xenial-backports/main amd64 Packages [7942 B]
Get:22 http://archive.ubuntu.com/ubuntu xenial-backports/universe amd64 Packages [8532 B]
Get:23 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Release.gpg [169 B]
Get:24 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64 Release.gpg [169 B]
Ign:24 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64 Release.gpg
Get:25 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Packages [209 kB]
Err:25 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Packages
Hash Sum mismatch
Get:26 https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64 Packages [46.3 kB]
Fetched 16.1 MB in 9s (1642 kB/s)
Reading package lists...
W: GPG error: https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools <[email protected]>
W: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64 Release' is not signed.
E: Failed to fetch https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64/Packages.gz Hash Sum mismatch
E: Some index files failed to download. They have been ignored, or old ones used instead.
The command '/bin/sh -c apt-get update --fix-missing && apt-get install -y --no-install-recommends build-essential software-properties-common python3 python3-dev python3-tk python3-pip build-essential libfreetype6-dev libpng12-dev libzmq3-dev libspatialindex-dev libsm6 vim wget git zip && apt-get clean && rm -rf /var/lib/apt/lists/*' returned a non-zero code: 100
sshuair
on 31 May 2019
Having the same issue after installing the cudatools key.
W: GPG error: http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools cudatools@nvidia.com
W: The repository 'http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release' is not signed.
Performing an install of the following packages:
libcudnn7=7.6.0.64-1+cuda10.0 \
libcudnn7-dev=7.6.0.64-1+cuda10.0 \
cuda-command-line-tools-10-0 cuda-cublas-10-0
cuda-cufft-10-0 cuda-curand-10-0 \
cuda-cusolver-10-0 cuda-cusparse-10-0
gives an error saying that:
WARNING: The following packages cannot be authenticated!
libcudnn7 libcudnn7-dev
Downloading from Sweden, if that matters.
Further, this seemed to work perfectly last week (on 19 June and before) but not after the weekend (24 June and after).
johanahlkvist
on 26 Jun 2019
Looks like this was a caching issue. Closing as we haven't seen any other issues in the past few months.
RenaudWasTaken
on 18 Sep 2019
@RenaudWasTaken
Having the same issue for https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x86_64 recently.
Actually I've seen this several times.
Especially in China where the .com URL get redirected to .cn domain.
If I hook up a proxy or simply do it in US, currently it's fine.
xkszltl
on 8 Oct 2019
@RenaudWasTaken
Still happening, here from Taiwan, for https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64, no redirection as for xkszltl above though. Tried every possible solution but to no avail.
calocedrus
on 22 Nov 2019
This is not fixed; multiple users are still reporting this from all sorts of countries.
Alex-Linhares
on 24 Jan 2020
Just in case it's of help to someone else, I could avoid this error and install CUDA+the NVIDIA driver on Mint19.3. Here's what I did:
clean install of mint (let nouveau drivers be installed, do not upgrade to NVIDIA). Follow the deb-local installation method (under a VPN @ California):
https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1804&target_type=deblocal
goto 1. and repeat the process around 50 times.
Voilá! NVIDIA drivers and CUDA running on your system! Wasn't that easy?
Alex-Linhares
on 24 Jan 2020
Severely censored network in China :)
372046933
on 24 Jun 2020
@372046933 I doubt this has anything to do with censorship. More likely to be a CDN related issue.
xkszltl
on 24 Jun 2020
I am also meeting this issue on China:
W: GPG error: https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1804/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools cudatools@nvidia.com
E: The repository 'https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64 Release' is not signed.
jdhao
on 1 Jul 2020
@jdhao , Use ladder to bypass the nasty situation.
372046933
on 1 Jul 2020
Same from South Korea right now:
Ign:10 http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Packages
Hit:10 http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Packages
Err:10 http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Packages
Hash Sum mismatch
...
Reading package lists...
W: GPG error: http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudatools <[email protected]>
W: The repository 'http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1604/x86_64 Release' is not signed.
E: Failed to fetch store:/var/lib/apt/lists/partial/developer.download.nvidia.com_compute_cuda_repos_ubuntu1604_x86%5f64_Packages.gz Hash Sum mismatch
E: Some index files failed to download. They have been ignored, or old ones used instead.
dniku
on 8 Jul 2020
This is not the right place to file issues related to containers running on top of nvidia-docker.
You will need to file an issue here:
https://forums.developer.nvidia.com/c/accelerated-computing/cuda/cuda-setup-and-installation/8
klueska
on 8 Jul 2020
i got the same problem on China, the error occurred when I used the image of Nvidia/CUDa :10.2-cudnn7-devel-ubuntu18.04 and executed the apt-get update command:
Reading package lists... Done
W: GPG error: https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1804/x8 6_64 Release: The following signatures were invalid: BADSIG F60F4B3D7FA2AF80 cudat ools cudatools@nvidia.com
E: The repository 'https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1 804/x86_64 Release' is not signed.
but, i just slove this problem by executing the following command:
gpg --keyserver keyserver.ubuntu.com --recv-keys F60F4B3D7FA2AF80
Crisescode
on 17 Jul 2020
我也遇到了这个问题，解决办法如下，在dockerfile中报错的step之前添加如下命令，如果curl已经安装，可以去掉第一个RUN命令
RUN apt-get update && apt-get install -y --no-install-recommends \
curl
RUN curl -s -L https://nvidia.github.io/libnvidia-container/gpgkey | \
apt-key add -
hopesala
on 4 Aug 2020
I got the same error in China
RUN curl -fsSL https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/7fa2af80.pub | apt-key add -
RUN echo "deb https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/ ./" > /etc/apt/sources.list.d/cuda.list
RUN echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list&& \
echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list&& \
echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \
echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.list
The problem was solved.
pogevip
on 4 Oct 2020
Good to know Aliyun has CUDA mirror, unfortunately they don't have nvidia/machine-learning.
xkszltl
on 4 Oct 2020
I got the same error in China
RUN curl -fsSL https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/7fa2af80.pub | apt-key add - RUN echo "deb https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/ ./" > /etc/apt/sources.list.d/cuda.list RUN echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list&& \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list&& \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.listThe problem was solved.
Not a security expert, but to my understanding, you shouldn't use public key from untrusted source (aliyun, in your case)
qwfy
on 4 Oct 2020
For those who are using this dockerfile as a starting point for your own dockerfile, you can consider download the bad file manually and COPY it to the corresponding location.
qwfy
on 4 Oct 2020
FYI usually it's too hard to get the file, regardless of who's doing the
download (docker build or human).
Local mirror makes things better, by moving the failure to the up-front
when mirroring.
For me I put whatever nvidia related network command into a until xxx; do
echo Retrying; done, usually there will be a success within 100-10k times.
On Sun, Oct 4, 2020 at 10:53 AM Incomplete notifications@github.com wrote:
For those who are using this dockerfile as a starting point for your own
dockerfile, you can consider download the bad file manually and COPY it
to the corresponding location.—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/NVIDIA/nvidia-docker/issues/969#issuecomment-703192776,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ABHWIUJQQNPS3UJPIPCK2ZTSI7PRLANCNFSM4HKSD7XA
.
--
From LTL
xkszltl
on 4 Oct 2020
I am also facing this issue with nvidia/cuda:10.2-cudnn7-devel-ubuntu18.04 and I am from India.
praveenbadimala
on 5 Oct 2020
@praveenbadimala
Just curious, does the traffic go through any CDN in India?
You can check with curl -SILsv 'https://developer.download.nvidia.com/compute/cuda/repos'.
For me I'm seeing a lot of X-Cache and X-OCRA-Accelerator fields in header with jp.krill.zenlogic.net after being 301 to nvidia.cn.
xkszltl
on 5 Oct 2020
@xkszltl
today I dint face the issue and the output of the curl -SILsv dint show that the traffic is routed through any CDN in India.
praveenbadimala
on 6 Oct 2020
I got the same error in China
RUN curl -fsSL https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/7fa2af80.pub | apt-key add - RUN echo "deb https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64/ ./" > /etc/apt/sources.list.d/cuda.list RUN echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list&& \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty main restricted universe multiverse" >> /etc/apt/sources.list&& \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-security main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-updates main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-proposed main restricted universe multiverse" >> /etc/apt/sources.list && \ echo "deb-src http://mirrors.tuna.tsinghua.edu.cn/ubuntu/ trusty-backports main restricted universe multiverse" >> /etc/apt/sources.listThe problem was solved.
Not a security expert, but to my understanding, you shouldn't use public key from untrusted source (aliyun, in your case)
why do you think aliyun is untrusted source???
Answergeng
on 14 Oct 2020
@Answergeng
Only author and trusted CA can be treated as trusted source.
HTTPS make sure the .pub you get is from nvidia without MITM attack.
If you get it from Ali, you can only make sure it's what Ali offers, not what nvidia offers.
Technically Ali can put malware in package and sign with their own priv key matching their .pub.
I know your point, Ali is big enough to not do stupid things in this way, but it's always a bad idea to keep door open. And darkness can gradually rise when too many doors are open.
xkszltl
on 14 Oct 2020
As you're using ubuntu 18.04, run:
# ###list keys if you need###
# apt-key list
# apt-key del F60F4B3D7FA2AF80
#
# apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
# apt update
it works.
via:
yantaozhao
on 26 Nov 2020
As you're using ubuntu 18.04, run:
# ###list keys if you need### # apt-key list # apt-key del F60F4B3D7FA2AF80 # # apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub # apt updateit works.
via:
* https://www.tensorflow.org/install/gpu#install_cuda_with_apt * https://docs.nvidia.com/cuda/wsl-user-guide/index.html#running-cuda * https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html
Not so fast. Actually in my case after executing your solution, if I run
# apt update
again, the error will occur. Like a haunting ghost.
PS: NONE of the above ideas about apt-list cleaning or gpg key removing and adding again worked or lasted.
saraba2b
on 29 Nov 2020
Related issues
jonghwanhyeon
·
4Comments
SpotCrowdTech
·
3Comments
Anthony-Tatowicz
·
4Comments
adbeda
·
3Comments
terrybroad
·
4Comments
Most helpful comment
I got the same error in China
The problem was solved.