[rdo-list] Overlap between dashboards/monitoring and future status aggregation tool

[David Moreau Simard]

Hello,

We've discussed several times that we need a public place to
centralize relevant status information for the community in an
user-friendly way.

The status information we have right now are available either from the
dashboard [1] or the monitoring [2].
Ultimately, the dashboard is monitoring things that is either already
monitored by our Sensu instance or could be.

I'm working on status.rdoproject.org [3] right now (not yet fed with live data).
The software for it, Cachet [4], is an open source version of statuspage.io.
It's meant to be driven automatically through it's API but there is
also an administrative interface to manually create/edit/resolve
events if need be.

So, for example, if Sensu detects a problem, it coud create an event
on status.rdoproject -- and resolve it when it's resolved.
The existing scripts for dashboards could be adapted to work with
status.rdoproject.
We could even have jenkins jobs post updates on status.rdoproject if we wanted.
It also supports displaying basic metrics, like shown in the demo [5].

There will clearly be an overlap between status.rdo and
dashboards.rdo. Are we okay with this ?
I'm not looking to create yet another standard to rule them all (cue
XKCD) but to use the right tool for the right job.

Thoughts ?

[1]: https://dashboards.rdoproject.org/rdo-dev
[2]: http://uchiwa.monitoring.rdoproject.org/ (creds: rdo/rdo)
[3]: http://status.rdoproject.org/
[4]: https://cachethq.io/
[5]: https://demo.cachethq.io/

David Moreau Simard
Senior Software Engineer | Openstack RDO

dmsimard = [irc, github, twitter]