# Kong prometheus plugin setup

**URL:** <https://discuss.konghq.com/t/kong-prometheus-plugin-setup/7745>\
**Category:** Questions\
**Tags:** kubernetes\
**Created:** [January 15, 2021, 5:34am UTC](https://discuss.konghq.com/t/kong-prometheus-plugin-setup/7745 "2021-01-15T05:34:05Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![piyush\_gupta](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.konghq.com/piyush_gupta/32/1931_2.png) [@piyush\_gupta](https://discuss.konghq.com/u/piyush_gupta)\
**Post date:** [January 15, 2021, 5:34am UTC](https://discuss.konghq.com/t/kong-prometheus-plugin-setup/7745/1 "2021-01-15T05:34:05Z")

</div>

Hi ,

we are planning to use the prometheus plugin for kong running in kubernetes .  
if we extract data from kong admin service /metrics endpoint , will i get the combined metrics from all the pods in or we need to extract /metrics from each of the kong pods .

Need some suggestion in the setup.

---

<div class="post-metadata">

**Author:** ![shane](https://yyz2.discourse-cdn.com/flex036/user_avatar/discuss.konghq.com/shane/32/2433_2.png) [@shane](https://discuss.konghq.com/u/shane)\
**Post date:** [January 19, 2021, 2:09pm UTC](https://discuss.konghq.com/t/kong-prometheus-plugin-setup/7745/2 "2021-01-19T14:09:31Z")

</div>

Hi @piyush_gupta,

The metrics endpoints themselves do not aggregate each other, and so each endpoint only provides metrics for itself.

However if Kong is deployed via the [chart](https://github.com/Kong/charts/tree/main/charts/kong) and the `serviceMonitor.enabled` flag is set to true, then the [Prometheus Operator will be automatically configured to pull from each endpoint](https://github.com/prometheus-operator/prometheus-operator/blob/master/Documentation/api.md#servicemonitorspec) and the data from each endpoint will be aggregated within Prometheus.

Let me know if this is helpful to you?