Spaces:
Sleeping
Sleeping
File size: 3,796 Bytes
5fe93dd | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 | import os
import asyncio
import httpx
from fastapi import FastAPI
from fastapi.middleware.cors import CORSMiddleware
import uvicorn
import subprocess
from config import settings
from utils.logger import logger
from utils.hardware import check_hardware
from api.routes import openenv, websocket, config_routes, scenario_routes, model_routes
app = FastAPI(title="NEXUS Backend API")
# Setup CORS
app.add_middleware(
CORSMiddleware,
allow_origins=["*"],
allow_credentials=True,
allow_methods=["*"],
allow_headers=["*"],
)
app.include_router(openenv.router)
app.include_router(websocket.router)
app.include_router(config_routes.router)
app.include_router(scenario_routes.router)
app.include_router(model_routes.router)
# ── Health check (required by HF Space automated ping) ─────────────────────
@app.get("/health")
async def health():
return {"status": "ok", "env": "nexus-incident-investigation"}
@app.get("/api")
async def root():
return {"name": "NEXUS", "version": "1.0.0", "status": "running"}
# Serve frontend statically if available
frontend_dist = os.path.join(os.path.dirname(os.path.dirname(__file__)), "frontend", "dist")
if os.path.exists(frontend_dist):
from fastapi.staticfiles import StaticFiles
from fastapi.responses import FileResponse
app.mount("/assets", StaticFiles(directory=os.path.join(frontend_dist, "assets")), name="assets")
@app.get("/")
@app.get("/{catchall:path}")
async def serve_frontend(catchall: str = ""):
return FileResponse(os.path.join(frontend_dist, "index.html"))
async def check_ollama():
try:
async with httpx.AsyncClient() as client:
resp = await client.get(settings.OLLAMA_BASE_URL.replace("/v1", ""), timeout=2.0)
if resp.status_code == 200:
logger.info(f"Ollama running at {settings.OLLAMA_BASE_URL}")
return True
except Exception:
logger.error(f"Ollama is NOT reachable at {settings.OLLAMA_BASE_URL}.")
logger.error("Please install from https://ollama.com and run it.")
return False
@app.on_event("startup")
async def startup_event():
# Print endpoints list
logger.info("Starting NEXUS Backend...")
for route in app.routes:
logger.info(f"Endpoint: {route.path} ({getattr(route, 'methods', 'WS')})")
hw = check_hardware()
logger.info(f"Hardware setup: VRAM available = {hw['vram_available_gb']} GB. GPU Mode enabled = {hw['use_gpu']}")
ollama_ok = await check_ollama()
if ollama_ok:
# Just check models via httpx to Ollama's local tags
async with httpx.AsyncClient() as client:
try:
resp = await client.get(settings.OLLAMA_BASE_URL.replace("/v1", "/api/tags"))
models = [m["name"] for m in resp.json().get("models", [])]
if settings.AGENT_A_MODEL not in models:
logger.warning(f"Default Agent A model {settings.AGENT_A_MODEL} not found. Run: ollama pull {settings.AGENT_A_MODEL}")
if settings.AGENT_B_MODEL not in models:
logger.warning(f"Default Agent B model {settings.AGENT_B_MODEL} not found. Run: ollama pull {settings.AGENT_B_MODEL}")
except Exception:
pass
# Run MCP server in bg
def run_mcp():
base_dir = os.path.dirname(__file__)
subprocess.Popen([os.sys.executable, "-m", "tools.tool_server"], cwd=base_dir)
run_mcp()
if __name__ == "__main__":
try:
uvicorn.run(app, host=settings.HOST, port=settings.PORT)
except Exception as e:
import traceback
print(f"FATAL ERROR AT STARTUP: {str(e)}")
traceback.print_exc()
os._exit(1)
|