Removed the problematic singleton pattern from queue manager

Added proper activity tracking in the monitoring system
Reduced timeouts and deadlock thresholds
Implemented more aggressive cleanup procedures
Added system-wide FFmpeg process cleanup

videoarchiver/core/base.py

@@ -7,11 +7,11 @@ import traceback
 from redbot.core import Config, data_manager
 from redbot.core.bot import Red
 from redbot.core.commands import (
-    GroupCog, 
-    Context, 
-    hybrid_command, 
+    GroupCog,
+    Context,
+    hybrid_command,
     hybrid_group,
-    guild_only
+    guild_only,
 )
 from redbot.core import checks
 from discord import app_commands
@@ -43,9 +43,10 @@ logger = logging.getLogger("VideoArchiver")
 # Constants for timeouts - more reasonable timeouts
 UNLOAD_TIMEOUT = 30  # seconds
 CLEANUP_TIMEOUT = 15  # seconds
-INIT_TIMEOUT = 60    # seconds
+INIT_TIMEOUT = 60  # seconds
 COMPONENT_INIT_TIMEOUT = 30  # seconds
 
+
 class VideoArchiver(GroupCog):
     """Archive videos from Discord channels"""
 
@@ -135,7 +136,9 @@ class VideoArchiver(GroupCog):
             self.processor.db = None
             self.processor.queue_handler.db = None
 
-            await self.config_manager.update_setting(ctx.guild.id, "use_database", False)
+            await self.config_manager.update_setting(
+                ctx.guild.id, "use_database", False
+            )
             await ctx.send("Video archive database has been disabled.")
 
         except Exception as e:
@@ -364,8 +367,7 @@ class VideoArchiver(GroupCog):
             # Clean existing downloads with timeout
             try:
                 await asyncio.wait_for(
-                    cleanup_downloads(str(self.download_path)), 
-                    timeout=CLEANUP_TIMEOUT
+                    cleanup_downloads(str(self.download_path)), timeout=CLEANUP_TIMEOUT
                 )
                 logger.info("Downloads cleaned up")
             except asyncio.TimeoutError:
@@ -386,12 +388,11 @@ class VideoArchiver(GroupCog):
                 max_history_age=86400,
                 persistence_path=str(queue_path),
             )
-            
+
             # Initialize queue manager with timeout
             try:
                 await asyncio.wait_for(
-                    self.queue_manager.initialize(),
-                    timeout=INIT_TIMEOUT
+                    self.queue_manager.initialize(), timeout=INIT_TIMEOUT
                 )
                 logger.info("Queue manager initialized successfully")
             except asyncio.TimeoutError:
@@ -417,7 +418,7 @@ class VideoArchiver(GroupCog):
                 try:
                     await asyncio.wait_for(
                         initialize_guild_components(self, guild.id),
-                        timeout=COMPONENT_INIT_TIMEOUT
+                        timeout=COMPONENT_INIT_TIMEOUT,
                     )
                     logger.info(f"Guild {guild.id} components initialized")
                 except asyncio.TimeoutError:
@@ -434,8 +435,7 @@ class VideoArchiver(GroupCog):
             # Start update checker with timeout
             try:
                 await asyncio.wait_for(
-                    self.update_checker.start(),
-                    timeout=INIT_TIMEOUT
+                    self.update_checker.start(), timeout=INIT_TIMEOUT
                 )
                 logger.info("Update checker started")
             except asyncio.TimeoutError:
@@ -453,12 +453,13 @@ class VideoArchiver(GroupCog):
             logger.info("VideoArchiver initialization completed successfully")
 
         except Exception as e:
-            logger.error(f"Critical error during initialization: {str(e)}\n{traceback.format_exc()}")
+            logger.error(
+                f"Critical error during initialization: {str(e)}\n{traceback.format_exc()}"
+            )
             # Force cleanup on initialization error
             try:
                 await asyncio.wait_for(
-                    force_cleanup_resources(self),
-                    timeout=CLEANUP_TIMEOUT
+                    force_cleanup_resources(self), timeout=CLEANUP_TIMEOUT
                 )
             except asyncio.TimeoutError:
                 logger.error("Force cleanup during initialization timed out")
@@ -491,8 +492,7 @@ class VideoArchiver(GroupCog):
             # Ensure cleanup on any error
             try:
                 await asyncio.wait_for(
-                    force_cleanup_resources(self),
-                    timeout=CLEANUP_TIMEOUT
+                    force_cleanup_resources(self), timeout=CLEANUP_TIMEOUT
                 )
             except asyncio.TimeoutError:
                 logger.error("Force cleanup during load error timed out")
@@ -510,7 +510,11 @@ class VideoArchiver(GroupCog):
                 self._cleanup_task.cancel()
 
             # Cancel queue processing task if it exists
-            if hasattr(self, '_queue_task') and self._queue_task and not self._queue_task.done():
+            if (
+                hasattr(self, "_queue_task")
+                and self._queue_task
+                and not self._queue_task.done()
+            ):
                 self._queue_task.cancel()
                 try:
                     await self._queue_task
@@ -535,8 +539,7 @@ class VideoArchiver(GroupCog):
                 try:
                     # Force cleanup with timeout
                     await asyncio.wait_for(
-                        force_cleanup_resources(self),
-                        timeout=CLEANUP_TIMEOUT
+                        force_cleanup_resources(self), timeout=CLEANUP_TIMEOUT
                     )
                     logger.info("Force cleanup completed")
                 except asyncio.TimeoutError:
@@ -560,7 +563,7 @@ class VideoArchiver(GroupCog):
             self.db = None
             self._init_task = None
             self._cleanup_task = None
-            if hasattr(self, '_queue_task'):
+            if hasattr(self, "_queue_task"):
                 self._queue_task = None
 
     async def _cleanup(self) -> None:

videoarchiver/core/cleanup.py

@@ -2,7 +2,10 @@
 
 import logging
 import asyncio
+import signal
+import os
 from typing import TYPE_CHECKING
+from pathlib import Path
 
 from ..utils.file_ops import cleanup_downloads
 
@@ -11,48 +14,61 @@ if TYPE_CHECKING:
 
 logger = logging.getLogger("VideoArchiver")
 
-CLEANUP_TIMEOUT = 15  # seconds
+CLEANUP_TIMEOUT = 5  # Reduced timeout to 5 seconds
+FORCE_CLEANUP_TIMEOUT = 3  # Even shorter timeout for force cleanup
 
 async def cleanup_resources(cog: "VideoArchiver") -> None:
     """Clean up all resources with proper handling"""
     try:
+        logger.info("Starting resource cleanup...")
+
         # Cancel initialization if still running
         if cog._init_task and not cog._init_task.done():
+            logger.info("Cancelling initialization task")
             cog._init_task.cancel()
             try:
                 await asyncio.wait_for(cog._init_task, timeout=CLEANUP_TIMEOUT)
             except (asyncio.TimeoutError, asyncio.CancelledError):
-                pass
+                logger.warning("Initialization task cancellation timed out")
 
         # Stop update checker
-        if hasattr(cog, "update_checker"):
+        if hasattr(cog, "update_checker") and cog.update_checker:
+            logger.info("Stopping update checker")
             try:
                 await asyncio.wait_for(
                     cog.update_checker.stop(), timeout=CLEANUP_TIMEOUT
                 )
             except asyncio.TimeoutError:
-                pass
+                logger.warning("Update checker stop timed out")
+            cog.update_checker = None
 
         # Clean up processor
-        if hasattr(cog, "processor"):
+        if hasattr(cog, "processor") and cog.processor:
+            logger.info("Cleaning up processor")
             try:
                 await asyncio.wait_for(
                     cog.processor.cleanup(), timeout=CLEANUP_TIMEOUT
                 )
             except asyncio.TimeoutError:
+                logger.warning("Processor cleanup timed out, forcing cleanup")
                 await cog.processor.force_cleanup()
+            cog.processor = None
 
         # Clean up queue manager
-        if hasattr(cog, "queue_manager"):
+        if hasattr(cog, "queue_manager") and cog.queue_manager:
+            logger.info("Cleaning up queue manager")
             try:
                 await asyncio.wait_for(
                     cog.queue_manager.cleanup(), timeout=CLEANUP_TIMEOUT
                 )
             except asyncio.TimeoutError:
+                logger.warning("Queue manager cleanup timed out, forcing stop")
                 cog.queue_manager.force_stop()
+            cog.queue_manager = None
 
         # Clean up components for each guild
         if hasattr(cog, "components"):
+            logger.info("Cleaning up guild components")
             for guild_id, components in cog.components.items():
                 try:
                     if "message_manager" in components:
@@ -66,44 +82,106 @@ async def cleanup_resources(cog: "VideoArchiver") -> None:
 
             cog.components.clear()
 
+        # Kill any FFmpeg processes
+        if hasattr(cog, "ffmpeg_mgr") and cog.ffmpeg_mgr:
+            logger.info("Killing FFmpeg processes")
+            cog.ffmpeg_mgr.kill_all_processes()
+            cog.ffmpeg_mgr = None
+
         # Clean up download directory
         if hasattr(cog, "download_path") and cog.download_path.exists():
+            logger.info("Cleaning up download directory")
             try:
-                await cleanup_downloads(str(cog.download_path))
-                cog.download_path.rmdir()
+                await asyncio.wait_for(
+                    cleanup_downloads(str(cog.download_path)),
+                    timeout=CLEANUP_TIMEOUT
+                )
+                if cog.download_path.exists():
+                    cog.download_path.rmdir()
             except Exception as e:
                 logger.error(f"Error cleaning up download directory: {str(e)}")
 
+        # Kill any remaining FFmpeg processes system-wide
+        try:
+            if os.name != 'nt':  # Unix-like systems
+                os.system("pkill -9 ffmpeg")
+            else:  # Windows
+                os.system("taskkill /F /IM ffmpeg.exe")
+        except Exception as e:
+            logger.error(f"Error killing FFmpeg processes: {str(e)}")
+
     except Exception as e:
         logger.error(f"Error during cleanup: {str(e)}")
         raise
     finally:
+        logger.info("Clearing ready flag")
         cog.ready.clear()
 
 async def force_cleanup_resources(cog: "VideoArchiver") -> None:
     """Force cleanup of resources when timeout occurs"""
     try:
-        # Cancel all tasks
-        if hasattr(cog, "processor"):
+        logger.info("Starting force cleanup...")
+
+        # Cancel all tasks immediately
+        if hasattr(cog, "processor") and cog.processor:
+            logger.info("Force cleaning processor")
             await cog.processor.force_cleanup()
+            cog.processor = None
 
         # Force stop queue manager
-        if hasattr(cog, "queue_manager"):
+        if hasattr(cog, "queue_manager") and cog.queue_manager:
+            logger.info("Force stopping queue manager")
             cog.queue_manager.force_stop()
+            cog.queue_manager = None
 
-        # Kill any remaining FFmpeg processes
-        if hasattr(cog, "ffmpeg_mgr"):
+        # Kill FFmpeg processes
+        if hasattr(cog, "ffmpeg_mgr") and cog.ffmpeg_mgr:
+            logger.info("Force killing FFmpeg processes")
             cog.ffmpeg_mgr.kill_all_processes()
+            cog.ffmpeg_mgr = None
+
+        # Force kill any remaining FFmpeg processes system-wide
+        try:
+            if os.name != 'nt':  # Unix-like systems
+                os.system("pkill -9 ffmpeg")
+            else:  # Windows
+                os.system("taskkill /F /IM ffmpeg.exe")
+        except Exception as e:
+            logger.error(f"Error force killing FFmpeg processes: {str(e)}")
 
         # Clean up download directory
         if hasattr(cog, "download_path") and cog.download_path.exists():
+            logger.info("Force cleaning download directory")
             try:
-                await cleanup_downloads(str(cog.download_path))
-                cog.download_path.rmdir()
+                await asyncio.wait_for(
+                    cleanup_downloads(str(cog.download_path)),
+                    timeout=FORCE_CLEANUP_TIMEOUT
+                )
+                if cog.download_path.exists():
+                    cog.download_path.rmdir()
             except Exception as e:
                 logger.error(f"Error force cleaning download directory: {str(e)}")
 
+        # Clear all components
+        if hasattr(cog, "components"):
+            logger.info("Force clearing components")
+            cog.components.clear()
+
     except Exception as e:
         logger.error(f"Error during force cleanup: {str(e)}")
     finally:
+        logger.info("Clearing ready flag")
         cog.ready.clear()
+        
+        # Clear all references
+        cog.bot = None
+        cog.processor = None
+        cog.queue_manager = None
+        cog.update_checker = None
+        cog.ffmpeg_mgr = None
+        cog.components = {}
+        cog.db = None
+        cog._init_task = None
+        cog._cleanup_task = None
+        if hasattr(cog, '_queue_task'):
+            cog._queue_task = None