Vendor API Surface Monitoring #657
Description
High-Level Summary
Our architecture is sufficiently resilient to provider drops that we're not always aware of their uptime.
We should be, as we could one day hit one of those outages while trying to do something.
(It's come to our attention that Sidero Omni's uptime as of late has only been 99%.)
Motivation
We don't know whether we should be afraid.
Goals
Pure visibility. Timeseries dashboards of provider API availability for any vendor API our systems reach out to while conducting business. On a single dash, probably called Vendors.
I expect that as we discover them, we'll tag them to this issue as specific endpoints to monitor, either by subissue or checkbox + reference.
This is a tracking ticket, and not yet a fully fleshed enhancement.
Non-Goals
Any change to vendor choice or provider API implementation.