Update Dockerfile

Dockerfile

@@ -4,28 +4,43 @@ FROM ghcr.io/ggml-org/llama.cpp:server AS source
 # Stage 2: Setup your environment using the full image
 FROM ghcr.io/ggml-org/llama.cpp:full
 
-# 1. FIX: Copy from /app/llama-server instead of /llama-server
+# 1. Copy llama-server binary
 COPY --from=source /app/llama-server /usr/local/bin/llama-server
 
 # 2. Setup Environment
 WORKDIR /app
 ENV HF_HOME=/app/.cache
 ENV NUMBA_CACHE_DIR=/app/.cache
+ENV LLAMA_CACHE=/app/.cache/llama.cpp
 
-# 3. Install Python & Fix Permissions Tools
+# 3. Install Python & Tools
 RUN apt-get update && \
-    apt-get install -y python3 python3-pip dos2unix && \
-    rm -rf /var/lib/apt/lists/*
+    apt-get install -y --no-install-recommends \
+    python3 \
+    python3-pip \
+    dos2unix \
+    curl \
+    && rm -rf /var/lib/apt/lists/* \
+    && apt-get clean
 
 # 4. Install Python Libraries
-RUN pip3 install --no-cache-dir huggingface_hub
+RUN pip3 install --no-cache-dir --break-system-packages huggingface_hub
 
 # 5. Copy and Fix the Start Script
 COPY start.sh /app/start.sh
 RUN dos2unix /app/start.sh && chmod +x /app/start.sh
 
-# 6. Permissions
-RUN mkdir -p /app/.cache && chmod -R 777 /app
+# 6. Create directories and set permissions
+RUN mkdir -p /app/.cache/llama.cpp \
+    && mkdir -p /app/models \
+    && chmod -R 777 /app
 
-# 7. Start
-ENTRYPOINT [ "/app/start.sh" ]
+# 7. Expose port
+EXPOSE 7860
+
+# 8. Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
+    CMD curl -f http://localhost:7860/health || exit 1
+
+# 9. Start
+ENTRYPOINT ["/app/start.sh"]