Skip to content

A new image for fah-client Bastet (v8.x)#37

Open
sanori wants to merge 13 commits intoFoldingAtHome:masterfrom
sanori:feat/bastet
Open

A new image for fah-client Bastet (v8.x)#37
sanori wants to merge 13 commits intoFoldingAtHome:masterfrom
sanori:feat/bastet

Conversation

@sanori
Copy link
Copy Markdown
Contributor

@sanori sanori commented Nov 23, 2025

I propose and have made a new docker image for fah-client v8 (Bastet)
support both nVidia CUDA and AMD ROCm and configurable version number.
How about this? Explanations are in fah-gpu-bastet/README.md

This PR may resolve issues #33 #36 .

I have tested it on my new machine, which is consist of Ryzen 9600 and RTX 3070.

Please feel free to comment this approach.

@mharsch
Copy link
Copy Markdown
Contributor

mharsch commented Feb 27, 2026

FWIW, I've been testing this for several weeks with current gen AMD and NVIDIA cards with no issues. I've mainly been running a custom build using the latest ROCm and CUDA releases to match the latest drivers.

The biggest issue I hit was lack of a documented headless initialization procedure i.e. seeding account-token and machine-name in config.xml within a fah volume before the first run of fah-gpu-bastet container.

@sanori
Copy link
Copy Markdown
Contributor Author

sanori commented Apr 1, 2026

@mharsch Thank you for your comment

The biggest issue I hit was lack of a documented headless initialization procedure i.e. seeding account-token and machine-name in config.xml within a fah volume before the first run of fah-gpu-bastet container.

I assumed that most people in the Folding@home community, except me, already knew that fah-client Bastet (version 8.x) could be configured via the GUI (https://v8-4.foldingathome.org/machines). So, I only added the GUI configuration link. 😅

I will add a few explanations for users running fah-client Bastet for the first time.

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

Successfully merging this pull request may close these issues.

2 participants