Add missing Docker configuration files for Ollama container

- Added Dockerfile with Ollama + Python 3.11 setup
- Added .dockerignore for build optimization
- Updated cerebrium.toml and main.py for Docker deployment
- Enables GitHub Actions testing of container build

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

Author: YedPool

cerebrium/.dockerignore

@@ -0,0 +1,37 @@
+# Ignore unnecessary files for Docker build
+__pycache__/
+*.pyc
+*.pyo
+*.pyd
+.Python
+env/
+pip-log.txt
+pip-delete-this-directory.txt
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.log
+.git/
+.mypy_cache/
+.pytest_cache/
+.hypothesis/
+
+# OS files
+.DS_Store
+.DS_Store?
+._*
+.Spotlight-V100
+.Trashes
+ehthumbs.db
+Thumbs.db
+
+# IDE files
+.vscode/
+.idea/
+*.swp
+*.swo
+*~

cerebrium/Dockerfile

@@ -0,0 +1,74 @@
+# Custom Ollama + Voice Agent Docker Image
+FROM ollama/ollama:latest
+
+# Install Python and system dependencies
+RUN apt-get update && apt-get install -y \
+    python3.11 \
+    python3-pip \
+    python3.11-venv \
+    curl \
+    wget \
+    && rm -rf /var/lib/apt/lists/*
+
+# Create symbolic links for python commands
+RUN ln -s /usr/bin/python3.11 /usr/bin/python
+RUN ln -s /usr/bin/pip3 /usr/bin/pip
+
+# Set working directory
+WORKDIR /app
+
+# Copy Python requirements and install dependencies
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application code
+COPY main.py .
+COPY *.py .
+
+# Create startup script that runs both Ollama and our app
+RUN cat > /app/start.sh << 'EOF'
+#!/bin/bash
+set -e
+
+echo "🚀 Starting Ollama + Voice Agent Container"
+
+# Start Ollama in background
+echo "📦 Starting Ollama server..."
+ollama serve &
+OLLAMA_PID=$!
+
+# Wait for Ollama to be ready
+echo "⏳ Waiting for Ollama to be ready..."
+for i in {1..30}; do
+    if curl -s http://localhost:11434/api/tags > /dev/null 2>&1; then
+        echo "✅ Ollama is ready!"
+        break
+    fi
+    echo "   Attempt $i/30 - waiting..."
+    sleep 2
+done
+
+# Download the model we need
+echo "📥 Downloading llama3.2:1b model..."
+ollama pull llama3.2:1b
+
+# Start our voice agent
+echo "🎤 Starting Voice Agent..."
+python main.py
+
+# Clean shutdown
+trap 'kill $OLLAMA_PID' EXIT
+wait
+EOF
+
+RUN chmod +x /app/start.sh
+
+# Expose ports
+EXPOSE 8600 11434
+
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
+    CMD curl -f http://localhost:11434/api/tags || exit 1
+
+# Start both services
+CMD ["/app/start.sh"]

cerebrium/cerebrium.toml

@@ -15,15 +15,8 @@ max_replicas = 3
 cooldown = 120
 
 [cerebrium.build]
-dockerfile = false
-python_packages = "./requirements.txt"
-include = [".", "*.py"]
-# Install system dependencies then download models
-shell_commands = [
-    "apt-get update && apt-get install -y curl",
-    "curl -fsSL https://ollama.com/install.sh | sh",
-    "python main.py download-files"
-]
+dockerfile = "./Dockerfile"
+include = [".", "*.py", "requirements.txt", "Dockerfile"]
 
 # Custom Runtime Configuration - LiveKit Agent Worker
 [cerebrium.runtime.custom]

cerebrium/main.py

@@ -30,29 +30,9 @@
 # Set cache directory for models
 os.environ["HF_HOME"] = "/cortex/.cache/"
 
-def setup_ollama():
-    """Start Ollama service (binary installed during build)."""
-    try:
-        logger.info("Starting Ollama service...")
-        
-        # Start Ollama service in background (binary pre-installed)
-        ollama_process = subprocess.Popen(
-            ["ollama", "serve"],
-            stdout=subprocess.PIPE,
-            stderr=subprocess.PIPE,
-            text=True
-        )
-        
-        logger.info("✓ Ollama service started")
-        return ollama_process
-        
-    except Exception as e:
-        logger.error(f"Failed to start Ollama: {e}")
-        return None
-
-def wait_for_ollama():
-    """Wait for Ollama to be ready."""
-    max_attempts = 30
+def check_ollama_ready():
+    """Check if Ollama is ready (should be started by Docker container)."""
+    max_attempts = 10
     for attempt in range(max_attempts):
         try:
             result = subprocess.run(
@@ -70,7 +50,7 @@ def wait_for_ollama():
         logger.info(f"Waiting for Ollama... ({attempt + 1}/{max_attempts})")
         time.sleep(2)
 
-    logger.error("✗ Ollama failed to start within timeout")
+    logger.error("✗ Ollama not available")
     return False
 
 class GmailAssistant(Agent):
@@ -205,18 +185,16 @@ async def log_usage():
         logger.info("✓ Build phase complete")
 
     else:
-        # Start Ollama service first
+        # Check if Ollama is ready (should be started by Docker container)
         logger.info("🚀 Starting Gmail Voice Assistant Agent...")
-        logger.info("Setting up Ollama...")
-        
-        ollama_process = setup_ollama()
+        logger.info("Checking Ollama availability...")
 
-        if ollama_process and wait_for_ollama():
-            logger.info("✓ Ollama is ready, starting LiveKit agent worker")
+        if check_ollama_ready():
+            logger.info("✓ Ollama is available, starting LiveKit agent worker")
         else:
-            logger.error("❌ Ollama setup failed - agent cannot function without LLM")
-            logger.error("Please check Ollama installation and model availability")
-            # Don't start agent if Ollama failed - it will definitely fail
+            logger.error("❌ Ollama not available - agent cannot function without LLM")
+            logger.error("Make sure Ollama is running in the Docker container")
+            # Don't start agent if Ollama not available - it will definitely fail
             sys.exit(1)
 
         # Start the LiveKit agent worker
@@ -229,10 +207,6 @@ async def log_usage():
             ))
         except KeyboardInterrupt:
             logger.info("Shutting down...")
-            if ollama_process:
-                ollama_process.terminate()
         except Exception as e:
             logger.error(f"Agent failed: {e}")
-            if ollama_process:
-                ollama_process.terminate()
             raise