6 ms·
For the last three years, I've been working on https://onlineornot.com https://onlineornot.com It's uptime monitoring (and status pages) for software teams. I
by rozenmd 3y ago
For the last three years, I've been working on https://onlineornot.com https://onlineornot.com
It's uptime monitoring (and status pages) for software teams.
In my words, the aim is "monitoring that doesn't suck" - I've worked at companies with proactive monitoring like OnlineOrNot before, and was surprised how little the incumbents are innovating in the space. One customer once told me "f*k <vendor>, all their system ever did was alert us when we weren't down".
I'm currently working on a self-documenting (OpenAPI, rolled it myself: https://developers.onlineornot.com/ https://developers.onlineornot.com/) API that'll let folks use terraform (or even just the API itself) to setup their uptime checks, cron job monitoring, status pages, even their teams.
- mattbrewsbytes 3y agoThis is a super simple IT problem to solve technically (ping a URL, provide a status HTML page, etc.) but really hard to get right, like your customer comment about a vendor. If done wrong, people will go to their "you had one job" card. How do you handle hosting of your own service and isolation from larger "cloud" or internet issues?
- rozenmd 3y agoI replicate the service across several AWS regions, and Cloudflare Workers. At the moment, it's really good at answering "am I down everywhere?", since I can just double check in several other regions. I recently taught it to answer "Am I down just in this region?" by monitoring across cloud providers in the same location, though it's more of a niche use case (for the people I chat to, anyway)