support icon indicating copy to clipboard operation
support copied to clipboard

add Prow metrics dashboard to grafana

Open goern opened this issue 4 years ago • 9 comments

Is your feature request related to a problem? Please describe. Prow is exposing a lot of metrics for example at http://gh-proxy-metrics-opf-ci-prow.apps.smaug.na.operate-first.cloud/metrics please make them available as a grafana dashboard. https://github.com/kubernetes/test-infra/tree/d917e4a1e773ccf701ac58424a0b4ee3490e8dd1/config/prow/cluster/monitoring/mixins/grafana_dashboards might be helpful

Describe the solution you'd like A grafana dashboard

Describe alternatives you've considered n/a

Additional context n/a

goern avatar Sep 02 '21 11:09 goern

/kind metrics /kind data-engineer /kind data-scientist /area monitoring

schwesig avatar Aug 01 '22 12:08 schwesig

@goern @harshad16 can this issue be closed because of

  • https://github.com/thoth-station/thoth-application/pull/2082
  • https://github.com/thoth-station/thoth-application/pull/2143

schwesig avatar Aug 01 '22 12:08 schwesig

The dashboard for prow is still not yet built. This could be a good issue for someone interested in building dashboard. I would suggest making this issue frozen and in the backlog.

Removing all assignees as this is no longer in active

harshad16 avatar Aug 01 '22 22:08 harshad16

https://www.kubermatic.com/blog/monitoring-prow-resources-with-prometheus-and-grafana/ is another writeup to give some more context

durandom avatar Aug 30 '22 10:08 durandom

  • https://console-openshift-console.apps.smaug.na.operate-first.cloud/k8s/ns/opf-ci-prow/routes
  • https://deck-metrics-opf-ci-prow.apps.smaug.na.operate-first.cloud/metrics
  • http://ghproxy-metrics-opf-ci-prow.apps.smaug.na.operate-first.cloud/metrics
  • https://hook-metrics-opf-ci-prow.apps.smaug.na.operate-first.cloud/metrics
  • grafana query up for UWM

but I cant figure out how to query prometheus for the actual prow component metrics

durandom avatar Aug 30 '22 10:08 durandom

https://github.com/operate-first/apps/pull/2393 fixes user workload monitoring but also removes the above public routes

durandom avatar Sep 08 '22 12:09 durandom

I imported all dashboards from https://github.com/kubernetes/test-infra/tree/d917e4a1e773ccf701ac58424a0b4ee3490e8dd1/config/prow/cluster/monitoring/mixins/grafana_dashboards to https://grafana.operate-first.cloud/dashboards/f/XfwHK1M4k/tmp

 ~/src/misc/test-infra/config/prow/cluster/monitoring/mixins |    master  |                                                                                                mhild
❯ make grafana-dashboards
 ~/src/misc/test-infra/config/prow/cluster/monitoring/mixins |    master  | ↵ SIGINT(2)                                                                                     mhild
❯ cd dashboards_out
 ~/src/misc/test-infra/config/prow/cluster/monitoring/mixins/dashboards_out |    master  |                                                                                 mhild
❯ perl -i -pe 's/prometheus/moc-smaug/' ./*
 ~/src/misc/test-infra/config/prow/cluster/monitoring/mixins/dashboards_out |    master  |              

durandom avatar Sep 08 '22 13:09 durandom

/assign @durandom

durandom avatar Sep 28 '22 13:09 durandom

image

goern avatar Sep 28 '22 14:09 goern