Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Unreliable determination of nv_jetson_model in run.sh - fixed #694

Open
TangmereCottage opened this issue Oct 22, 2024 · 0 comments
Open

Comments

@TangmereCottage
Copy link
Contributor

TangmereCottage commented Oct 22, 2024

In run.sh,

cat /proc/device-tree/model > /tmp/nv_jetson_model
cat is used to generate a file containing the model descriptor in /tmp/nv_jetson_model. The problem is that if that file is not found, the run.sh command creates a new folder, which then hoses the AI, such as nano_llm, due to "failed to find/open file /proc/device-tree/model" (which makes sense because it's not a file but a folder).

The bug is sort of tricky because once the folder has been created by a broken cat command, subsequent invocations of run.sh will not remove it and your stack will remain broken.

Solution:
Make sure that /tmp does not erroneously contain a folder called "nv_jetson_model". If there is one, manually delete it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant