Dockerfile.orchestrator

# Lightweight Dockerfile for the AI Chat Orchestrator
# Builds a small image that runs `orchestrator_service.py` behind gunicorn

FROM python:3.11-slim

# Install runtime deps
RUN apt-get update \
     && apt-get install -y --no-install-recommends \
         ca-certificates \
         curl \
     && rm -rf /var/lib/apt/lists/*

# Set workdir
WORKDIR /app

# Copy source
COPY orchestrator_service.py /app/
COPY requirements_orchestrator.txt /app/

# Install python deps
RUN pip install --no-cache-dir -r requirements_orchestrator.txt

# Expose orchestrator port
EXPOSE 5002

# Default environment variables (can be overridden at runtime)
# On macOS Docker Desktop containers, use host.docker.internal to reach services
# running on the Docker host (e.g. Phi3/ollama). This can be overridden at
# runtime or in docker-compose when deploying to different environments.
ENV PHI3_HOST=host.docker.internal
ENV PHI3_PORT=11434
ENV PHI3_MODEL=phi3.5:latest
ENV FLASK_ENV=production

# Run with gunicorn; --preload is avoided so streaming requests can be handled by workers
CMD ["gunicorn", "-b", "0.0.0.0:5002", "orchestrator_service:app", "--workers", "1", "--threads", "4", "--timeout", "120"]