deployment 16

Dockerfile

@@ -11,6 +11,8 @@ RUN apt-get update && apt-get install -y \
     build-essential \
     curl \
     ca-certificates \
+    dnsutils \
+    iputils-ping \
     libopenblas-dev \
     && rm -rf /var/lib/apt/lists/*
 
@@ -36,6 +38,10 @@ RUN pip install --no-cache-dir --force-reinstall 'numpy<2.0.0'
 # Create storage directory
 RUN mkdir -p /app/storage && chmod 777 /app/storage
 
+# Copy and make startup script executable
+COPY start-backend.sh /app/start-backend.sh
+RUN chmod +x /app/start-backend.sh
+
 # Expose HF Spaces port
 EXPOSE 7860
 
@@ -52,5 +58,5 @@ ENV USE_FAKE_REDIS=true
 HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
     CMD curl -f http://localhost:7860/health || exit 1
 
-# Start FastAPI server
-CMD ["python", "-m", "uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
+# Start with diagnostic script
+CMD ["/app/start-backend.sh"]

backend/pipeline.py

@@ -139,6 +139,7 @@ class TranscriptionPipeline:
         """Download audio from YouTube URL using yt-dlp."""
         output_path = self.temp_dir / "audio.wav"
 
+        # Try with different extractors and network options
         cmd = [
             "yt-dlp",
             "-x",  # Extract audio
@@ -147,17 +148,22 @@ class TranscriptionPipeline:
             "--output", str(output_path.with_suffix('')),  # yt-dlp adds .wav
             "--force-ipv4",  # Force IPv4 to avoid DNS issues
             "--socket-timeout", "30",
-            "--source-address", "0.0.0.0",  # Bind to all interfaces
-            "--legacy-server-connect",  # Use legacy connection method
-            # Workarounds for YouTube restrictions
-            "--extractor-args", "youtube:player_client=android,web",
+            "--retries", "10",
+            "--fragment-retries", "10",
+            # Try alternative extractors
+            "--extractor-args", "youtube:player_client=android,ios,web",
             "--no-check-certificates",
+            # Add verbose output for debugging
+            "--verbose",
             self.youtube_url
         ]
 
         result = subprocess.run(cmd, capture_output=True, text=True)
 
         if result.returncode != 0:
+            # Log the full error for debugging
+            print(f"yt-dlp stderr: {result.stderr}")
+            print(f"yt-dlp stdout: {result.stdout}")
             raise RuntimeError(f"yt-dlp failed: {result.stderr}")
 
         if not output_path.exists():

start-backend.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -e
+
+echo "🔍 Checking DNS configuration..."
+
+# Try to resolve youtube.com
+if ! nslookup youtube.com > /dev/null 2>&1; then
+    echo "⚠️ DNS resolution failed, attempting fixes..."
+    
+    # Check if we can ping Google's DNS
+    if ping -c 1 8.8.8.8 > /dev/null 2>&1; then
+        echo "✓ Network connectivity exists"
+        echo "⚠️ DNS resolution issue - this may be a HF Spaces network policy"
+    else
+        echo "✗ No network connectivity"
+    fi
+    
+    # Show current DNS config
+    echo "Current DNS configuration:"
+    cat /etc/resolv.conf || echo "Cannot read /etc/resolv.conf"
+fi
+
+echo ""
+echo "🚀 Starting backend server..."
+cd /app/backend && python -u main.py