Server Offline with either llama.cpp or HF-Waitress

[Rotoslider]

Ubuntu 22.04, Installed LARS today using instruction from your github readme. Using Conda environment.
llama-server --version version: 3912 (edc26566)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu
added to path in barshrc
model is in model directory.
if I run llama-server -m /home/asus/builds/LARS/lars_storage/models/dolphin-2.9.3-mistral-7B-32k-Q8_0.gguf -c 2048 from any where the server starts and I can chat with it.
main: server is listening on 127.0.0.1:8080 - starting the main loop
srv update_slots: all slots are idle

config.json
lars_server_log.log
llama_cpp_server_output_log.txt
The hf_waitress output log is blank and no json was created for it

What else do i need to give you to help troubleshoot?

[Rotoslider]

Removed the LARS folder and the Conda env. Created a venv per your readme and reinstalled everything. Same issue as before.
I can run llama-server stand alone. I can run Ollama or LMStudio or Open webUI. So sure sure where the disconnect here is.
Your app works other than it will not start either server.
Tried in various browsers no change.
Tried from another computer on the network. no change

[Stwerp]

What error were you getting? I was having trouble with llama-server not being detected and an "idle_slots" error. This turned out to be a key error on line 2144 of app.py.

LARS/web_app/app.py

Line 2144 in 710489f

print(f"llama.cpp Server ready: {data['slots_idle']} idle slots, {data['slots_processing']} processing slots.")

            print(f"llama.cpp Server ready: {data['slots_idle']} idle slots, {data['slots_processing']} processing slots.")

The data['slots_idle'] was triggering an error the try/except block was detecting. Just commenting out that line (or just printing "llama.cpp ready" instead) seemed to work.