SystemError during setup vlad automatic

[errchh]

Sorry for a newbie question.

I set up vlad automatic, there's some error messages. I can enter the webui, I have a prompt but it gives an error and quits. Below are some info:

-- fedora 37
-- python 3.10.9
-- I guess CUDA is ok since automatic1111 is running fine.
-- ran bash webui.sh

Available models: /home/eric/automatic/models/Stable-diffusion 0
Download the default model? (y/N) y
Downloading: "https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors" to /home/eric/automatic/models/Stable-diffusion/v1-5-pruned-emaonly.safetensors

100.0%
ControlNet v1.1.116
ControlNet v1.1.116
Module load: 
/home/eric/automatic/extensions-builtin/stable-diffusion-webui-rembg/scripts/postprocessing_rembg.py: 
SystemError
╭────────────────────────────────── Traceback (most recent call last) ──────────────────────────────────╮
│ /home/eric/automatic/modules/script_loading.py:10 in load_module                                      │
│                                                                                                       │
│    9 │   try:                                                                                         │
│ ❱ 10 │   │   module_spec.loader.exec_module(module)                                                   │
│   11 │   except Exception as e:                                                                       │
│ in exec_module:883                                                                                    │
│                                                                                                       │
│                                        ... 8 frames hidden ...                                        │
│                                                                                                       │
│ /home/eric/automatic/venv/lib/python3.10/site-packages/numba/np/ufunc/__init__.py:3 in <module>       │
│                                                                                                       │
│    2                                                                                                  │
│ ❱  3 from numba.np.ufunc.decorators import Vectorize, GUVectorize, vectorize, guvectorize             │
│    4 from numba.np.ufunc._internal import PyUFunc_None, PyUFunc_Zero, PyUFunc_One                     │
│                                                                                                       │
│ /home/eric/automatic/venv/lib/python3.10/site-packages/numba/np/ufunc/decorators.py:3 in <module>     │
│                                                                                                       │
│     2                                                                                                 │
│ ❱   3 from numba.np.ufunc import _internal                                                            │
│     4 from numba.np.ufunc.parallel import ParallelUFuncBuilder, ParallelGUFuncBuilder                 │
╰───────────────────────────────────────────────────────────────────────────────────────────────────────╯
SystemError: initialization of _internal failed without raising an exception
Loading theme: black-orange
Image Browser: Creating database
Image Browser: Database created
Running on local URL:  http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
Initializing middleware
Checkpoint model.ckpt not found; loading fallback v1-5-pruned-emaonly.safetensors
Loading weights: /home/eric/automatic/models/Stable-diffusion/v1-5-pruned-emaonly.safetensors ━━━ 0… -:-…
                                                                                                  GB     
Creating model from config: /home/eric/automatic/configs/v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Calculating sha256: /home/eric/automatic/models/Stable-diffusion/v1-5-pruned-emaonly.safetensors6ce0161689b3853acaa03779ec93eafe75a02f4ced659bee03f50797806fa2fa
Applying scaled dot product cross attention optimization
Embeddings loaded:  (0)
Model loaded in 4.8s (load=0.5s create=0.4s hash=2.2s apply=0.8s vae=0.5s move=0.5s)
Startup time: 151.2s (torch=2.9s gradio=1.0s libraries=0.7s models=138.8s codeformer=0.1s scripts=1.2s 
ui=1.5s start=0.1s scripts app_started_callback=0.1s checkpoint=4.9s)
^CExiting

[vladmandic]

Torch may be fine, but numba is not, no idea why, but I've heard similar stories on fedora.

It's used by rembg extension and not critical for normal operations. Just to verify, go to settings, disable that extension and see how does startup looks.