# WSL2 - TensorFlow Install Issue Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered

**URL:** https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508
**Category:** cuDNN
**Tags:** tensorflow, cudnn, cuda
**Created:** [February 12, 2024, 12:25am UTC](https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508 "2024-02-12T00:25:53Z")
**Posts on this page:** 4
**Page:** 1

<div class="post-metadata">

### Author: ![vatsal.raicha](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@vatsal.raicha](https://forums.developer.nvidia.com/u/vatsal.raicha)
#### Post date: [February 12, 2024, 12:25am UTC](https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508/1 "2024-02-12T00:25:53Z")

</div>

## Description

Facing this error message when trying to use tensorflow on WSL2 Ubuntu - Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered.  
TF-TRT Warning: Could not find TensorRT  
could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa\_node  
Your kernel may have been built without NUMA support.

## Environment

**TensorRT Version** : 8.6.1.6  
**GPU Type** : 4080 Laptop GPU  
**Nvidia Driver Version** : NVIDIA-SMI 546.17 Driver Version: 546.17  
**CUDA Version** : 12.3, CUDA Toolkit Version - 11.8  
**CUDNN Version** : v8.6  
**Operating System + Version** : Windows 11  
**Python Version (if applicable)**: Python 3.10.13  
**TensorFlow Version (if applicable)**: 2.15.0  
**PyTorch Version (if applicable)**: N/A  
**Baremetal or Container (if container which image + tag)**: N/A  
pyenv Version - 2.3.35-11-g9908daf8  
WSL version: 2.0.14.0  
Kernel version: 5.15.133.1-1  
[tensorFlowErrors.txt](https://forums.developer.nvidia.com/uploads/short-url/5xY4TQ6zXBI3FFVhi4IWIHGG4CX.txt) (2.3 KB)

## Relevant Files

Please attach or include links to any models, data, files, or scripts necessary to reproduce your issue. (Github repo, Google Drive, Dropbox, etc.)

## Steps To Reproduce

Followed the steps in here to install tensorflow and verification

> **[Windows WSL2 - Install TensorFlow with pip](https://www.tensorflow.org/install/pip#windows-wsl2)**
>
> The following GPU-enabled devices are supported:

Please include:

- Exact steps/commands to build your repro
- Exact steps/commands to run your repro
- Full traceback of errors encountered

---

<div class="post-metadata">

### Author: ![lsm.liborio](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@lsm.liborio](https://forums.developer.nvidia.com/u/lsm.liborio)
#### Post date: [February 19, 2024, 12:44pm UTC](https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508/2 "2024-02-19T12:44:22Z")

</div>

I have the same error, but just installing Nvidia Container Toolkit in linux mint 21.3 in [https://github.com/NVIDIA/nvidia-container-toolkit](https://github.com/NVIDIA/nvidia-container-toolkit).

I get the following error message after importing tensorflow:

 ![image](https://global.discourse-cdn.com/nvidia/original/4X/7/2/8/7288d26ebaf746ff1b682c1b475f86b8b3f25f82.jpeg)

GPU: RTX 3060 12GB  
Linux mint 21.3 fresh new install  
tensorflow 2.14.0

---

<div class="post-metadata">

### Author: ![AakankshaS](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/aakankshas/32/14047_2.png) [@AakankshaS](https://forums.developer.nvidia.com/u/AakankshaS)
#### Post date: [February 29, 2024, 1:59pm UTC](https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508/3 "2024-02-29T13:59:14Z")

</div>

Hi @vatsal.raicha ,  
Can you please raise this issue on [Issues · tensorflow/tensorflow · GitHub](https://github.com/tensorflow/tensorflow/issues)

Thanks

---

<div class="post-metadata">

### Author: ![vatsal.raicha](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@vatsal.raicha](https://forums.developer.nvidia.com/u/vatsal.raicha)
#### Post date: [March 4, 2024, 1:28am UTC](https://forums.developer.nvidia.com/t/wsl2-tensorflow-install-issue-unable-to-register-cudnn-factory-attempting-to-register-factory-for-plugin-cudnn-when-one-has-already-been-registered/282508/4 "2024-03-04T01:28:14Z")

</div>

Thanks @AakankshaS,

I have raised this on the TensorFlow GitHub - [WSL2 - TensorFlow Install Issue Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered · Issue #63109 · tensorflow/tensorflow · GitHub](https://github.com/tensorflow/tensorflow/issues/63109)