Fix: Install llama-cpp-python at startup to /tmp to avoid build timeout

Dockerfile

@@ -2,28 +2,23 @@ FROM python:3.12-slim
 
 WORKDIR /app
 
-# Install system dependencies for building llama-cpp-python
+# Install system dependencies for building llama-cpp-python at runtime
 RUN apt-get update && apt-get install -y \
     build-essential \
     cmake \
     git \
     && rm -rf /var/lib/apt/lists/*
 
-# Copy requirements and install
+# Copy requirements and install (excluding llama-cpp-python)
 COPY requirements.txt .
-
-# Set environment variables for CPU-only build (no BLAS, no CUDA)
-ENV CMAKE_ARGS="-DLLAMA_BLAS=OFF -DLLAMA_CUBLAS=OFF -DLLAMA_METAL=OFF"
-
-# Install llama-cpp-python (will build from source with CPU-only)
-RUN pip install --no-cache-dir llama-cpp-python==0.3.14 --verbose
-
-# Install remaining requirements
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy application code
+# Copy application code and startup scripts
 COPY . .
 
+# Make startup script executable
+RUN chmod +x start.sh
+
 # Create data directory for persistence
 RUN mkdir -p /data
 
@@ -33,13 +28,14 @@ ENV MODEL_REPO=daskalos-apps/phi4-cybersec-Q4_K_M
 ENV MODEL_FILENAME=phi4-mini-instruct-Q4_K_M.gguf
 ENV USE_RAG=true
 ENV CACHE_ENABLED=true
+ENV CMAKE_ARGS="-DLLAMA_BLAS=OFF -DLLAMA_CUBLAS=OFF -DLLAMA_METAL=OFF"
 
 # Expose port
 EXPOSE 8000
 
-# Health check
-HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+# Health check (with longer startup period for llama-cpp build)
+HEALTHCHECK --interval=30s --timeout=10s --start-period=600s --retries=3 \
   CMD python -c "import requests; requests.get('http://localhost:8000/health')"
 
-# Run the application
-CMD ["python", "main.py"]
+# Run the startup script which installs llama-cpp-python then starts the app
+CMD ["bash", "start.sh"]

install_llama.py

@@ -0,0 +1,59 @@
+#!/usr/bin/env python3
+"""
+Install llama-cpp-python at startup to avoid Docker build timeout.
+This script runs before the main application starts.
+"""
+import subprocess
+import sys
+import os
+import time
+
+def install_llama_cpp():
+    """Install llama-cpp-python to /tmp directory"""
+    print("=" * 60)
+    print("🔧 Installing llama-cpp-python at startup...")
+    print("=" * 60)
+
+    # Set installation directory to /tmp (always writable)
+    install_dir = "/tmp/llama_cpp_install"
+    os.makedirs(install_dir, exist_ok=True)
+
+    # Set environment variables
+    env = os.environ.copy()
+    env['PYTHONUSERBASE'] = install_dir
+    env['CMAKE_ARGS'] = '-DLLAMA_BLAS=OFF -DLLAMA_CUBLAS=OFF -DLLAMA_METAL=OFF'
+
+    start_time = time.time()
+
+    try:
+        # Install to custom location
+        result = subprocess.run(
+            [sys.executable, "-m", "pip", "install",
+             "--target", install_dir,
+             "--no-cache-dir",
+             "llama-cpp-python==0.3.14"],
+            env=env,
+            capture_output=False,
+            text=True,
+            check=True
+        )
+
+        elapsed = time.time() - start_time
+        print(f"\n✅ llama-cpp-python installed successfully in {elapsed:.1f}s")
+
+        # Add to Python path
+        if install_dir not in sys.path:
+            sys.path.insert(0, install_dir)
+
+        print(f"📦 Installation directory: {install_dir}")
+        return True
+
+    except subprocess.CalledProcessError as e:
+        elapsed = time.time() - start_time
+        print(f"\n❌ Installation failed after {elapsed:.1f}s")
+        print(f"Error: {e}")
+        return False
+
+if __name__ == "__main__":
+    success = install_llama_cpp()
+    sys.exit(0 if success else 1)

start.sh

@@ -0,0 +1,14 @@
+#!/bin/bash
+set -e
+
+echo "🚀 Starting Cybersecurity Chatbot..."
+
+# Install llama-cpp-python at startup
+python3 install_llama.py
+
+# Add installation directory to PYTHONPATH
+export PYTHONPATH="/tmp/llama_cpp_install:$PYTHONPATH"
+
+# Start the application
+echo "▶️  Launching FastAPI application..."
+python3 main.py