Status
Status is important. It might be a status report, or sometimes it's "network status" (are we up?) You can setup custom monitoring services and playbooks for failover and recovery. However, sometimes a simple 3rd party service might suffice. A newcomer in the field is https://statuspage.io
Another project is Graphite [1], which can be used with the built-in statsd collection in MediaWiki [2]
In the Cloud-Native world, Prometheus is an active project. It's a time-series database and metrics monitoring solution, with many exporters including haproxy, memcached, mysqld, etc..
Prometheus is able to alert on any metric, so long as the data is available. This data comes from your code Instrumentation. See their docs for information on Instrumenting your code. https://prometheus.io/docs/practices/instrumentation/
And for a great dashboard, there is Netdata
