# Is there a way get Traefik latency metric?

**URL:** <https://community.traefik.io/t/is-there-a-way-get-traefik-latency-metric/14633>\
**Category:** Traefik v2\
**Tags:** metrics, kubernetes-ingress, kubernetes-crd\
**Created:** [June 1, 2022, 5:13am UTC](https://community.traefik.io/t/is-there-a-way-get-traefik-latency-metric/14633 "2022-06-01T05:13:09Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![shettyh](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/shettyh/32/5411_2.png) [@shettyh](https://community.traefik.io/u/shettyh)\
**Post date:** [June 1, 2022, 5:13am UTC](https://community.traefik.io/t/is-there-a-way-get-traefik-latency-metric/14633/1 "2022-06-01T05:13:09Z")

</div>

Hi,

Wanted to get the traefik induced latency for each request. Is there any metric exposed for the same ?

Could not find anything here [Traefik Metrics Overview - Traefik](https://doc.traefik.io/traefik/observability/metrics/overview/)

Thanks

---

<div class="post-metadata">

**Author:** ![Dojo](https://sea2.discourse-cdn.com/flex020/user_avatar/community.traefik.io/dojo/32/7478_2.png) [@Dojo](https://community.traefik.io/u/Dojo)\
**Post date:** [August 27, 2023, 10:23am UTC](https://community.traefik.io/t/is-there-a-way-get-traefik-latency-metric/14633/2 "2023-08-27T10:23:50Z")

</div>

Some Grafana Dashboards are displaying an average responsetime by issuing the following promql: `sum(traefik_service_request_duration_seconds_sum) by (service) / sum(traefik_service_request_duration_seconds_count) by (service)`.

Imo this is at best an approximation for latancy - there are many downsides to this method.

I would love to have a real latency metric in treafik!