Error: stat(/usr/local/cuda/lib64/libcudart_static.a): Bad message

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
1

If CUDA is already installed and I attempted to install CUDA 11.8 but misconfigured it, resulting in the following error, and now commands like sudo apt-get update no longer work. It seems like I need to modify the filesystem, but would it be faster to just reinstall JetPack? If there are other solutions, please let me know.

ls: cannot access ‘/usr/local/cuda’: Bed Message
ls: cannot access ‘/usr/local/cuda-11’: Bed Message
total 80
drwxr-xr-x 15 root root 4096 Nov 16 09:34 .
drwxr-xr-x 11 root root 4096 Mar 4 2021 …
drwxr-xr-x 2 root root 4096 Nov 22 14:53 bin
l??? ? ? ? ? ? cuda
l??? ? ? ? ? ? cuda-11
drwxr-xr-x 11 root root 4096 Mar 20 2023 cuda-11.4
drwxr-xr-x 13 root root 4096 Nov 15 17:05 cuda-11.8
drwxr-xr-x 3 root root 4096 May 19 2023 doc
drwxr-xr-x 2 root root 4096 Aug 1 2020 etc
drwxr-xr-x 2 root root 4096 Aug 1 2020 games
drwxr-xr-x 8 root root 4096 Dec 21 11:52 include
drwxr-xr-x 7 root root 20480 Dec 21 11:52 lib
drwxr-xr-x 4 root root 4096 May 19 2023 man
drwxrwxr-x 26 mingi mingi 4096 Nov 16 11:57 onnxruntime
drwxr-xr-x 2 root root 4096 Aug 1 2020 sbin
drwxr-xr-x 18 root root 4096 Nov 19 13:52 share
drwxr-xr-x 2 root root 4096 Aug 1 2020 src

E: Failed to fetch store:/var/lib/apt/lists/partial/ports.ubuntu.com_ubuntu-ports_dists_focal-backports_universe_dep11_Components-arm64.yml.xz Renaming failed. Bed Message (/var/lib/apt/lists/partial/ports.ubuntu.com_ubuntu-ports_dists_focal-backports_universe_dep11_Components-arm64.yml.gz → /var/lib/apt/lists/ports.ubuntu.com_ubuntu-ports_dists_focal-backports_universe_dep11_Components-arm64.yml.gz).
E: Some index files failed to download. They have been ignored, or old ones used instead.

3

Hi,

Do you want to use CUDA 11.8 on the JetPack 5 environment which contains an 11.4 GPU driver?
If so, you will need our CUDA Compatibility feature.
Please find below the command to set up the global variable. (please update the version accordingly).

But for Orin Nano, there are some releases containing newer CUDA versions.
For example, it’s more recommended to set up your environment with JetPack 6.x which uses CUDA 12.x.

Thanks.

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.