diff --git a/web_server.py b/web_server.py
index bcc4640f..49091a96 100644
--- a/web_server.py
+++ b/web_server.py
@@ -312,7 +312,14 @@ class WebUIDownloadMonitor:
                 return False
             else:
                 queue_time = current_time - task['queued_start_time']
-                if queue_time > 120:  # Increased timeout to reduce retry spam
+                
+                # Use context-aware timeouts like GUI:
+                # - 15 seconds for artist album downloads (streaming context)
+                # - 90 seconds for background playlist downloads  
+                is_streaming_context = task.get('track_info', {}).get('is_album_download', False)
+                timeout_threshold = 15.0 if is_streaming_context else 90.0
+                
+                if queue_time > timeout_threshold:
                     # Track retry attempts to prevent rapid loops
                     retry_count = task.get('stuck_retry_count', 0)
                     last_retry = task.get('last_retry_time', 0)
@@ -343,7 +350,14 @@ class WebUIDownloadMonitor:
                 return False
             else:
                 download_time = current_time - task['downloading_start_time']
-                if download_time > 120:  # Increased timeout to reduce retry spam
+                
+                # Use context-aware timeouts like GUI:
+                # - 15 seconds for artist album downloads (streaming context)  
+                # - 90 seconds for background playlist downloads
+                is_streaming_context = task.get('track_info', {}).get('is_album_download', False)
+                timeout_threshold = 15.0 if is_streaming_context else 90.0
+                
+                if download_time > timeout_threshold:
                     retry_count = task.get('stuck_retry_count', 0)
                     last_retry = task.get('last_retry_time', 0)
                     
@@ -408,21 +422,36 @@ class WebUIDownloadMonitor:
             
             # Only attempt cancellation if we have what looks like a proper download ID
             # (not a filename fallback which would be much longer)
-            if download_id and username and isinstance(download_id, str) and len(download_id) < 100:
-                print(f"🚫 Attempting to cancel stuck download: {os.path.basename(str(download_id))} from {username} (task: {task_id[:8]}...)")
-                try:
-                    success = asyncio.run(soulseek_client.cancel_download(download_id, username, remove=False))
-                    if success:
-                        print(f"✅ Successfully cancelled download {str(download_id)[:8]}... from {username}")
-                        # Clear any stored download info to prevent status conflicts
-                        task.pop('soulseek_download_id', None)
-                        task.pop('soulseek_username', None)
+            # Handle cancellation attempts based on download ID type
+            if download_id and username and isinstance(download_id, str):
+                # Check if this is a synthetic download ID (generated by web server)
+                if download_id.startswith('web_dl_'):
+                    print(f"⚠️ Skipping cancellation for synthetic download ID {download_id} - cannot cancel via slskd API")
+                    # Use original download ID for potential cancellation if available
+                    original_id = task.get('original_download_id')
+                    if original_id and len(str(original_id)) > 100:
+                        # This is still a filename, can't cancel
+                        print(f"⚠️ Original download ID is filename, cannot cancel: {str(original_id)[:50]}...")
                     else:
-                        print(f"⚠️ Cancel request failed for {str(download_id)[:8]}... - all API endpoints returned errors, proceeding with retry anyway")
-                except Exception as e:
-                    print(f"⚠️ Cancel exception for {str(download_id)[:8]}...: {str(e)[:150]}, proceeding with retry anyway")
+                        print(f"⚠️ No usable original download ID for cancellation")
+                elif len(download_id) < 100:
+                    # This looks like a proper slskd download ID, try to cancel it
+                    print(f"🚫 Attempting to cancel stuck download: {download_id[:8]}... from {username} (task: {task_id[:8]}...)")
+                    try:
+                        success = asyncio.run(soulseek_client.cancel_download(download_id, username, remove=False))
+                        if success:
+                            print(f"✅ Successfully cancelled download {download_id[:8]}... from {username}")
+                            # Clear any stored download info to prevent status conflicts
+                            task.pop('soulseek_download_id', None)
+                            task.pop('soulseek_username', None)
+                        else:
+                            print(f"⚠️ Cancel request failed for {download_id[:8]}... - API error, proceeding with retry anyway")
+                    except Exception as e:
+                        print(f"⚠️ Cancel exception for {download_id[:8]}...: {str(e)[:100]}, proceeding with retry anyway")
+                else:
+                    print(f"⚠️ Download ID too long ({len(download_id)} chars) - likely filename, cannot cancel")
             else:
-                print(f"⚠️ Skipping cancellation - invalid data (ID len={len(str(download_id)) if download_id else 0}, username='{username}'), proceeding with retry")
+                print(f"⚠️ Missing download ID or username for cancellation, proceeding with retry")
                     
         except Exception as e:
             print(f"⚠️ Error in cancellation logic, proceeding with retry: {e}")
@@ -5437,7 +5466,18 @@ def _attempt_download_with_candidates(task_id, candidates, track, batch_id=None)
                                 _on_download_completed(batch_id, task_id, success=False)
                             return False
                         
-                        download_tasks[task_id]['download_id'] = download_id
+                        # Store download information with proper ID handling
+                        # If download_id is actually a filename (long path), generate a proper ID
+                        if download_id and len(str(download_id)) > 100:
+                            # This is likely a filename fallback, generate a proper download ID
+                            import uuid
+                            actual_download_id = f"web_dl_{uuid.uuid4().hex[:8]}"
+                            print(f"⚠️ Generated synthetic download ID {actual_download_id} for filename fallback: {str(download_id)[:50]}...")
+                            download_tasks[task_id]['download_id'] = actual_download_id
+                            download_tasks[task_id]['original_download_id'] = download_id  # Keep original for reference
+                        else:
+                            download_tasks[task_id]['download_id'] = download_id
+                        
                         download_tasks[task_id]['username'] = username
                         download_tasks[task_id]['filename'] = filename