Add debug Dockerfile with better error handling and logging

Dockerfile

@@ -7,6 +7,7 @@ RUN apt-get update && apt-get install -y \
     python3.11-venv \
     git \
     build-essential \
+    curl \
     && rm -rf /var/lib/apt/lists/*
 
 # Create a non-root user (required by HF Spaces)
@@ -27,42 +28,65 @@ COPY --chown=user hono-proxy $HOME/app/hono-proxy
 RUN python3.11 -m venv $HOME/venv
 ENV PATH="$HOME/venv/bin:$PATH"
 
-# Install Python dependencies for embedding API
-RUN pip install --upgrade pip
-RUN pip install -r requirements_embedding.txt
+# Install Python dependencies with error handling
+RUN pip install --upgrade pip && \
+    pip install -r requirements_embedding.txt || \
+    (echo "Failed to install Python dependencies" && exit 1)
 
-# Install pnpm and Node dependencies for Hono proxy
+# Install pnpm and Node dependencies
 RUN npm install -g pnpm
 WORKDIR $HOME/app/hono-proxy
-RUN pnpm install
+RUN pnpm install || (echo "Failed to install Node dependencies" && exit 1)
 
-# Copy Vespa certificates (these need to be included in the repo)
+# Copy Vespa certificates
 RUN mkdir -p $HOME/.vespa/il-infra.colpali-server.default
-COPY --chown=user vespa-certs/* $HOME/.vespa/il-infra.colpali-server.default/ || true
+COPY --chown=user vespa-certs/* $HOME/.vespa/il-infra.colpali-server.default/ || echo "Warning: Vespa certs not found"
 
-# Create startup script for backend services only
+# Create a more robust startup script
 WORKDIR $HOME/app
-RUN printf '#!/bin/bash\n\
-\n\
-# Start embedding API on port 8001\n\
-echo "Starting ColPali embedding API on port 8001..."\n\
-python embedding_api.py &\n\
-EMBED_PID=$!\n\
-\n\
-# Wait for embedding API to be ready\n\
-sleep 10\n\
-\n\
-# Start Hono proxy on HF Spaces port 7860\n\
-echo "Starting Hono proxy on port 7860..."\n\
-cd hono-proxy && PORT=7860 CORS_ORIGIN="*" EMBEDDING_API_URL="http://localhost:8001" VESPA_ENDPOINT="https://f5acf536.ed2ceb09.z.vespa-app.cloud" npx tsx src/index.ts\n\
-\n\
-# If Hono exits, kill embedding service\n\
-kill $EMBED_PID\n' > start-backend.sh
+RUN echo '#!/bin/bash' > start-backend.sh && \
+    echo 'set -e' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo 'echo "Starting backend services..."' >> start-backend.sh && \
+    echo 'echo "Environment variables:"' >> start-backend.sh && \
+    echo 'env | grep -E "VESPA|HF_|HUGGING" || true' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo '# Start embedding API with logging' >> start-backend.sh && \
+    echo 'echo "Starting ColPali embedding API on port 8001..."' >> start-backend.sh && \
+    echo 'python embedding_api.py 2>&1 | tee embedding.log &' >> start-backend.sh && \
+    echo 'EMBED_PID=$!' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo '# Wait and check if embedding API started' >> start-backend.sh && \
+    echo 'sleep 10' >> start-backend.sh && \
+    echo 'if ! kill -0 $EMBED_PID 2>/dev/null; then' >> start-backend.sh && \
+    echo '    echo "ERROR: Embedding API failed to start"' >> start-backend.sh && \
+    echo '    cat embedding.log' >> start-backend.sh && \
+    echo '    exit 1' >> start-backend.sh && \
+    echo 'fi' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo '# Test embedding API' >> start-backend.sh && \
+    echo 'curl -f http://localhost:8001/health || echo "Warning: Embedding API health check failed"' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo '# Start Hono proxy' >> start-backend.sh && \
+    echo 'echo "Starting Hono proxy on port 7860..."' >> start-backend.sh && \
+    echo 'cd hono-proxy' >> start-backend.sh && \
+    echo 'export PORT=7860' >> start-backend.sh && \
+    echo 'export CORS_ORIGIN="*"' >> start-backend.sh && \
+    echo 'export EMBEDDING_API_URL="http://localhost:8001"' >> start-backend.sh && \
+    echo 'export VESPA_ENDPOINT="${VESPA_ENDPOINT:-https://f5acf536.ed2ceb09.z.vespa-app.cloud}"' >> start-backend.sh && \
+    echo 'npx tsx src/index.ts 2>&1 | tee ../hono.log' >> start-backend.sh && \
+    echo '' >> start-backend.sh && \
+    echo '# If we get here, Hono exited' >> start-backend.sh && \
+    echo 'echo "Hono proxy exited, killing embedding service..."' >> start-backend.sh && \
+    echo 'kill $EMBED_PID 2>/dev/null || true' >> start-backend.sh && \
+    chmod +x start-backend.sh
 
-RUN chmod +x start-backend.sh
+# Add health check
+HEALTHCHECK --interval=30s --timeout=30s --start-period=5s --retries=3 \
+  CMD curl -f http://localhost:7860/api/health || exit 1
 
-# Expose HF Spaces port (Hono proxy will run on this)
+# Expose HF Spaces port
 EXPOSE 7860
 
-# Run the startup script
-CMD ["./start-backend.sh"]
+# Run with better error handling
+CMD ["bash", "-c", "./start-backend.sh || (echo 'Startup failed, showing logs:' && cat *.log 2>/dev/null && exit 1)"]

embedding_api.py

@@ -23,7 +23,10 @@ if hf_token:
     os.environ["HF_TOKEN"] = hf_token
 
 # Setup logging
-logging.basicConfig(level=logging.INFO)
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)
 logger = logging.getLogger(__name__)
 
 # Initialize FastAPI