Longer (5 vs 0.2s) guard_cycle for Django-Q
To reduce average CPU load. Given paperless is mostly idling and when not, just spawns long-running operations (e.g., OCR), a 0.2s health-check of the Django-Q cluster is unnecessary. See also #1084
This commit is contained in:
		
							parent
							
								
									9dc1989507
								
							
						
					
					
						commit
						20c597b1d7
					
				@ -458,6 +458,7 @@ PAPERLESS_WORKER_RETRY: Final[int] = __get_int(
 | 
			
		||||
 | 
			
		||||
Q_CLUSTER = {
 | 
			
		||||
    "name": "paperless",
 | 
			
		||||
    "guard_cycle": 5,
 | 
			
		||||
    "catch_up": False,
 | 
			
		||||
    "recycle": 1,
 | 
			
		||||
    "retry": PAPERLESS_WORKER_RETRY,
 | 
			
		||||
 | 
			
		||||
		Loading…
	
	
			
			x
			
			
		
	
		Reference in New Issue
	
	Block a user