Post #960510

2026-04-01 09:07 UTC

reposting with source and alt: https://damrnelson.github.io/github-historical-uptime/ note that the scale is between 99.5% and 100% uptime on the first graph, and actually the real data is worse since it's averaging the uptime of all their services and then the data gets even worse when you don't average uptime across entire months

Replies (1)

@clarfonthey@toot.cat 2026-04-01 09:19

like on one hand, the absolutely abysmal performance of Actions is crushing GitHub's uptime statistics but on the other, every single GitHub service is still going on and offline post-Microslop to give a bit of perspective on monthly uptime statistics, in terms of how much time you're allowed to be down (in a 30-day month) before failing to meet the quotas: 99.9999%: 2.5 seconds 99.999%: 25 seconds 99.99%: 2 minutes, 19 seconds 99.9%: 43 minutes, 12 seconds 99%: 7 hours, 12 minutes 90% 3 days and keep in mind, Microslop has every incentive to lie about these numbers wherever possible. if the page refreshes in two minutes, hey, that means the site isn't down! the status page for them is a way to let people know they're "working hard to resolve the issue" whenever stuff crops up, and if they think you won't notice, they won't report it

Open ##1486017