tree bab5f4ab1a055613c8f7627d420ef86ee5e88fab
parent 6813b84a99f6d2d62abde358a1a9d73dbbd35042
author Thomas Draebing <thomas.draebing@sap.com> 1613032113 +0100
committer Thomas Draebing <thomas.draebing@sap.com> 1613134036 +0100

Add healthcheck ping and dashboard for Gerrit

The healthcheck plugin for Gerrit provides a convenient way to determine
the health of different functionalities and components of Gerrit. If
the endpoint provided by the plugin is pinged, it will execute a set
of checks and return either 200 if all checks passed or 500 if at least
one failed. It will also provide metrics that can be scraped by
Prometheus.

This change adds the option for Gerrit installations outside of Kubernetes
to install a sidecar container in the Prometheus deployment that every
30 s pings the healthcheck plugin's endpoint, thereby triggering the
checks. This is not provided for kubernetes, since there the ping should
be the task of the Kubernetes liveness probes.

The change additionally adds a dashboard displaying the status of the
healthcheck for each Gerrit instance over time.

Change-Id: Ieeedc4406b642e542c89679a8314d771ca0928af