Feature request: Internal Sensu, etcd and Postgres monitoring, and alert integration

[jhenderson-pro]

Feature Suggestions

As a Sensu operator, it would be helpful to have a one-stop shop catalog integration to monitor Sensu with Sensu, Including Sensu backends, embedded or external etcd, and Postgres health monitoring and alerts.

Context

Working with a customer, they brought up the point that we do not currently have complete integration to 'monitor Sensu with Sensu.' However, we currently have pieces of this to roll your own and accomplish the same result, but they're requesting an integration to have them rolled together.

[portertech]

What documentation do we have around monitoring sensu with sensu?

[portertech]

Working on requirements.

We have documented monitoring sensu with sensu, using /health.
https://docs.sensu.io/sensu-go/latest/operations/monitor-sensu/monitor-sensu-with-sensu/

CE stated that there is a gap in regards to monitoring etcd, we need one or more key metrics to monitor.

[portertech]

We will need a big disclaimer that a sensu cluster cannot monitoring itself, it needs a partner (another cluster).

[portertech]

@jhenderson-pro please add customer etcd issue context.

[jhenderson-pro]

Initially, a customer had an issue with unbounded etcd growth (internal ref in the CE priority queue). This had a significant impact on Sensu's health. This feature is requested to facilitate simpler configuring of monitoring and alerts for critical Sensu components in an integration with the goal of some warning before the environment becomes unresponsive.
Initially, internal or external etcd, Postgres, and the backends. Etcd disk quotas are a common support request and may be good to include in an integration.