[Issue]: Could not load library libcudnn_cnn_infer.so.8.

[elchupacabrinski]

Issue Description

Just did a clean install parallel to auto1111, but somehow I can't generate anything and I'm always getting the following error:
Could not load library libcudnn_cnn_infer.so.8. Error: libnvrtc.so: cannot open shared object file: No such file or directory

Version Platform Description

Ubuntu 22.04

[vladmandic]

this is a torch issue, not an issue with the repo.
perhaps community can help - tranfering to discussions

[dirtyapenz]

I had the same issue on Ubuntu 22.04. I found the answer at https://www.tensorflow.org/install/pip

I did the following to get this installed using conda

conda create -n vladui python=3.10.9
conda activate vladui
conda install -c "nvidia/label/cuda-11.8.0" cuda-toolkit
python3 -m pip install nvidia-cudnn-cu11==8.7.0.84
mkdir -p $CONDA_PREFIX/etc/conda/activate.d
echo 'CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CONDA_PREFIX/lib/:$CUDNN_PATH/lib' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
source $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
cd ~
git clone https://github.com/vladmandic/automatic
cd automatic
./webui.sh

[MrUSBEN]

Thankyou! This worked for me on Ubuntu 22 for ComfyUI

[feceugur]

Thank you so much this worked for me on Ubuntu 20.04.6 LTS

[oyoanan]

fatal: 无法访问 'https://github.com/vladmandic/automatic/'：GnuTLS recv error (-110): TLS 链接非正常地终止了。

[permagreen-zz]

I'm on Arch and unfortunately the above solution didn't work for me. Whenever I hit generate I get Could not load library libcudnn_cnn_infer.so.8. Error: libnvrtc.so: cannot open shared object file: No such file or directory

This only happens with vlad, btw, not automatic or any other webui.

[dirtyapenz]

Check that your $LD_LIBRARY_PATH is something sane.
echo $LD_LIBRARY_PATH

Check cuda version installed.
nvcc --version

vlad uses tensorflow. A1111 does not.

[Genius-pig]

I found two solutions:
https://discuss.tensorflow.org/t/wsl2-installation-failing-miserably/16236
https://stackoverflow.com/questions/76016645/tensorflow-2-12-could-not-load-library-libcudnn-cnn-infer-so-8-in-wsl2

I follow first one, it solve my problems.

[sixone-Jiang]

the first one is OK

[Bortrex]

In my case, I did not have installed nvcc driver.
this fixed my problem:
sudo apt install nvidia-cuda-toolkit

[iturner72]

same :)

[ValerioSpenn]

Also for me!

[Ediblock]

Worked for me as well

[nnaaaa]

from another lib, but it still work for my problems

[hyun95roh]

You saved my life. Thank You !!!!

[rrakso]

This worked for me:

As discussed in this issue 516 it is just a matter to add this in the .bashrc:

export LD_LIBRARY_PATH=/usr/lib/wsl/lib:$LD_LIBRARY_PATH

Be sure that your library is in /usr/lib/wsl/lib, to see it you can run

ldconfig -p | grep cuda

Reference: Libcudnn_cnn_infer.so.8 library can not found

[Nik-Kras]

Just a note: vast.ai machine was Ubuntu 22.04 x84_64 machine. And I tested it for this machine only.

Here is what helped me:

1. Install NVCC driver

sudo apt install nvidia-cuda-toolkit

2. Install cuDNN (Version 8!)

It is not trivial to install version 8!
If you go to the official website to install cuDNN and enter your machine's details, you will only get the latest version — version 9. I tried it, but it doesn't work (even when you link from calls of version 8 to version 9).

Official Website

So, you would need to go to archive and submit your request for version 8.

After that, you will get a file called .deb. In my case, I got cudnn-local-repo-ubuntu2204-8.9.7.29_1.0-1_amd64.deb

Here is how you install it:

sudo dpkg -i cudnn-local-repo-ubuntu2204-8.9.7.29_1.0-1_amd64.deb
sudo cp /var/cudnn-local-repo-ubuntu2204-8.9.7/cudnn-local-*.gpg /usr/share/keyrings/

Maybe you would also need to do that:

sudo apt-get update
sudo apt-get install --reinstall libcudnn8 libcudnn8-dev libcudnn8-samples

3. Verify installation

ls /usr/lib/x86_64-linux-gnu/libcudnn* | grep libcudnn_ops_infer.so.8
ls /usr/lib/x86_64-linux-gnu/libcudnn* | grep libcudnn_cnn_infer.so.8

[rayryeng]

Thank you - this worked for me. I skipped step 1 because I am using WSL so I've already installed the CUDA toolkit using NVIDIA's recommended setup. The main problem was getting access to cuDNN v8, which I have downloaded from the archives and installed correctly.

The only other thing I had to do was make sure my LD_LIBRARY_PATH was pointing to where libcudnn was located, so I additionally did:

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/x86_64-linux-gnu/

Thanks again!