Page MenuHomePhabricator

Complete Netbox prometheus scraping
Closed, ResolvedPublic

Description

We left the project to scrape netbox stats into prometheus, which stalled on authenticating prometheus. Need to revisit this task.

Related Objects

Event Timeline

I lost context on that, but https://netbox.wikimedia.org/metrics is not behind authentication.

The previous try was done in https://gerrit.wikimedia.org/r/c/operations/puppet/+/526819 (see comments).

As we rely more and more on Netbox API, it would be useful to track its health and usage.

Not sure however which ones of the 648 metrics are useful to have on a dashboard.

Change 805889 had a related patch set uploaded (by Ayounsi; author: Ayounsi):

[operations/puppet@production] Prometheus: scrap Netbox django metrics

https://gerrit.wikimedia.org/r/805889

Change 805888 merged by Ayounsi:

[operations/puppet@production] Netbox: expose Netbox on the frontend's FQDN

https://gerrit.wikimedia.org/r/805888

Change 805889 merged by Ayounsi:

[operations/puppet@production] Prometheus: gently pull Netbox django metrics

https://gerrit.wikimedia.org/r/805889

Change 806368 had a related patch set uploaded (by Ayounsi; author: Ayounsi):

[operations/puppet@production] Prometheus/Netbox: use netbox.wikimedia.org SNI

https://gerrit.wikimedia.org/r/806368

Change 806368 merged by Ayounsi:

[operations/puppet@production] Prometheus/Netbox: use netbox.wikimedia.org SNI

https://gerrit.wikimedia.org/r/806368

All done thanks to John and Filippo.

Example dashboard can be seen there https://grafana.wikimedia.org/d/DvXT6LCnk/ feel free to add to it if needed.