43 lines
1,021 B
Docker
43 lines
1,021 B
Docker
# Rerank API Server
|
|
# Multi-stage build for smaller final image
|
|
|
|
FROM python:3.11.12-slim AS builder
|
|
|
|
# Install uv
|
|
COPY --from=ghcr.io/astral-sh/uv:0.10 /uv /usr/local/bin/uv
|
|
|
|
WORKDIR /app
|
|
|
|
# Copy project files
|
|
COPY pyproject.toml README.md ./
|
|
COPY src/ ./src/
|
|
|
|
# Install dependencies (with optional extras for backends)
|
|
RUN uv sync --frozen --no-dev --all-extras || uv sync --no-dev --all-extras
|
|
|
|
# Production image
|
|
FROM python:3.11.12-slim
|
|
|
|
WORKDIR /app
|
|
|
|
# Copy virtual environment from builder
|
|
COPY --from=builder /app/.venv /app/.venv
|
|
|
|
# Copy source code
|
|
COPY src/ ./src/
|
|
|
|
# Set environment variables
|
|
ENV PATH="/app/.venv/bin:$PATH"
|
|
ENV PYTHONUNBUFFERED=1
|
|
ENV RERANK_HOST=0.0.0.0
|
|
ENV RERANK_PORT=14200
|
|
|
|
# Health check (using Python since curl not available in slim image)
|
|
HEALTHCHECK --interval=30s --timeout=10s --retries=3 \
|
|
CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:14200/health')" || exit 1
|
|
|
|
# Expose port
|
|
EXPOSE 14200
|
|
|
|
# Run the server
|
|
CMD ["python", "-m", "rerank.cli"]
|