Thanos Query Instant Latency High
critical

Description Thanos Query {{$labels.job}} has a 99th percentile latency of {{$value}} seconds for instant queries.
Query for 10m
>>>
	
				
					(
				
			
				
					
					
						histogram_quantile
					
				
			
				
					(0.99, sum by (job, le) (
				
			
				
					
					
						rate
					
				
			
				
					(
				
			
				
					
				
			
				
					{job=~".*thanos-query.*", handler="query"}[5m]))) > 40 and sum by (job) (
				
			
				
					
					
						rate
					
				
			
				
					(
				
			
				
					
				
			
				
					{job=~".*thanos-query.*", handler="query"}[5m])) > 0)
				
			
    
Query Explanation

The rule computes the 99th‑percentile latency of Thanos Query's instant‑query HTTP requests over the past 5 minutes using histogram_quantile on the rate of the request‑duration buckets, and it triggers when that latency exceeds 40 seconds while there is any query traffic (i.e., the total request rate is greater than zero).

Get Alert
Download
Copy to Clipboard