0

I'm trying to install tensorflow-gpu on Ubuntu 18.04 LTS. I have an ASUS GL553VD with NVIDIA GTX-1050. I'm following the installation instructions for CUDA 9.0 as seen here: https://developer.nvidia.com/cuda-90-download-archive?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1604&target_type=deblocal

Everything works fine until I run sudo apt-get install cuda (i.e. the 4th step) and then this happens:Terminal Image

I have been following this tutorial to install tensorflow-gpu: https://github.com/markjay4k/Install-Tensorflow-on-Ubuntu-17.10-/blob/master/Tensorflow%20Install%20instructions.ipynb

I've tried changing mirrors a lot, selecting the best Ubuntu finds as well as others. I've run sudo apt-get update multiple times successfully. but installation for CUDA always gets stuck at Waiting for headers:99% and then aborts installation.

Please tell me how can I get this to work or please tell me any other configurations of GPU drivers, CUDA and cuDNN that Tensorflow supports. I have trying to install CUDA 9.0 since its mentioned on the official website: https://www.tensorflow.org/install/gpu

  • 1
    The graphics-drivers ppa is (trying) to bring in the 390.87 driver -- maybe get rid of that ppa , clean out any (partially) installed nvidia drivers, and just use the standard repo to get the 390.77 driver. – ubfan1 Feb 28 at 5:49
0

I hope that you have resolved your issue, i faced a similar issue when running sudo apt-get install cuda with similar error message of 99% Waiting for header

E: Failed to fetch http://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/./cuda-cupti-10-1_10.1.168-1_amd64.deb  Connection failed [IP: 192.229.232.112 80]
E: Aborting install.

All i did was repeatedly running sudo apt-get install cuda --fix-missing until the installation continued to fix the issue.

Hope this helps whoever who would potentially meet the same issue :)

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.