I think it's hilarious that so many huge companies can't figure out how to run a simple, static web page that can be updated independently of the rest of their infrastructure. It's something that a kid with a Raspberry Pi and an intro to Python book could do in an hour.
Yet here we see example after example of status web pages that break when the status they're supposed to report is anything other than good.
Not even the fact that it's simple - it's the obvious and de facto way that a status page should be run: an architecture that is entirely separate from the main product so that an outage of the main service doesn't affect the status page itself
Yet here we see example after example of status web pages that break when the status they're supposed to report is anything other than good.