Update Dockerfile

Dockerfile

@@ -1,20 +1,13 @@
 # Dockerfile
-FROM python:3.11-slim
+FROM python:3.11-slim AS builder
 
-# Set environment variables to reduce Python buffering and improve logging
+# Set environment variables for Python optimization
 ENV PYTHONUNBUFFERED=1 \
     PYTHONDONTWRITEBYTECODE=1 \
     OLLAMA_MODELS=/home/ollama/.ollama \
     OLLAMA_HOST=0.0.0.0
 
-# Create a non-root user
-RUN useradd -m -u 1000 ollama && \
-    mkdir -p /home/ollama/.ollama && \
-    chown -R ollama:ollama /home/ollama
-
-WORKDIR /app
-
-# Install system dependencies efficiently
+# Install build dependencies
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
     curl \
@@ -25,6 +18,19 @@ RUN apt-get update && \
 # Install Ollama
 RUN curl -fsSL https://ollama.ai/install.sh | sh
 
+# Final stage
+FROM python:3.11-slim
+
+# Create a non-root user
+RUN useradd -m -u 1000 ollama && \
+    mkdir -p /home/ollama/.ollama && \
+    chown -R ollama:ollama /home/ollama
+
+WORKDIR /app
+
+# Copy Ollama binaries from builder stage
+COPY --from=builder /usr/local/bin/ollama /usr/local/bin/ollama
+
 # Copy requirements and install Python dependencies
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
@@ -39,40 +45,41 @@ RUN chown -R ollama:ollama /app && \
 # Switch to ollama user
 USER ollama
 
-# Create a startup script with configurable model pulling
+# Create a startup script with configurable model pulling and enhanced logging
 RUN echo '#!/bin/bash\n\
 set -e\n\
-echo "Starting Ollama server..."\n\
-ollama serve &\n\
+LOG_FILE=/home/ollama/startup.log\n\
+echo "Starting Ollama server at $(date)" >> $LOG_FILE\n\
+ollama serve >> $LOG_FILE 2>&1 &\n\
 sleep 15\n\
 MODELS_TO_PULL="${MODELS_TO_PULL:-hf.co/gguf-org/gemma-3-270m-gguf:Q5_K_S}"\n\
-echo "Pulling models: $MODELS_TO_PULL"\n\
+echo "Pulling models: $MODELS_TO_PULL" | tee -a $LOG_FILE\n\
 IFS=',' read -ra MODEL_ARRAY <<< "$MODELS_TO_PULL"\n\
 for model in "${MODEL_ARRAY[@]}"; do\n\
-    echo "Pulling model $model..."\n\
+    echo "Pulling model $model..." | tee -a $LOG_FILE\n\
     for attempt in {1..3}; do\n\
-        if ollama pull "$model"; then\n\
-            echo "Model $model pulled successfully"\n\
+        if ollama pull "$model" >> $LOG_FILE 2>&1; then\n\
+            echo "Model $model pulled successfully" | tee -a $LOG_FILE\n\
             break\n\
         else\n\
-            echo "Attempt $attempt: Failed to pull model $model, retrying in 10 seconds..."\n\
+            echo "Attempt $attempt: Failed to pull model $model, retrying in 10 seconds..." | tee -a $LOG_FILE\n\
             sleep 10\n\
         fi\n\
         if [ $attempt -eq 3 ]; then\n\
-            echo "Error: Failed to pull model $model after 3 attempts"\n\
+            echo "Error: Failed to pull model $model after 3 attempts" | tee -a $LOG_FILE\n\
             exit 1\n\
         fi\n\
     done\n\
 done\n\
-echo "Starting Gunicorn server..."\n\
-exec python3 -m gunicorn --bind 0.0.0.0:7860 --workers 1 --timeout 120 --log-level info app:app' > /app/start.sh && \
+echo "Starting Gunicorn server at $(date)" | tee -a $LOG_FILE\n\
+exec python3 -m gunicorn --bind 0.0.0.0:7860 --workers 1 --timeout 120 --log-level info app:app >> $LOG_FILE 2>&1' > /app/start.sh && \
     chmod +x /app/start.sh
 
 # Expose port
 EXPOSE 7860
 
-# Health check with enhanced timeout
-HEALTHCHECK --interval=30s --timeout=15s --start-period=30s --retries=3 \
+# Health check with optimized parameters
+HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
     CMD curl -f http://localhost:7860/health || exit 1
 
 # Run the startup script