medium10 pts · first answer countspromqllatency

p99 latency across all instances

Your service exposes a Prometheus histogram named http_request_duration_seconds. Which query correctly gives the p99 latency across all instances?

A.avg(histogram_quantile(0.99, rate(http_request_duration_seconds_bucket[5m])))B.histogram_quantile(0.99, http_request_duration_seconds_sum)C.max(http_request_duration_seconds_count) * 0.99D.histogram_quantile(0.99, sum by (le) (rate(http_request_duration_seconds_bucket[5m])))

Discussion 0

Join free to post in the discussion.

p99 latency across all instances

Discussion 0

Related questions