A live tracker counts consecutive days GitHub has gone without a status incident, pulling from GitHub’s public status history.
Key Takeaways
The site tracks GitHub uptime streaks using GitHub’s own status history as the data source.
Current 2026 high score is displayed as a leaderboard-style metric, implying incidents are frequent enough to reset often.
The project frames GitHub reliability as a measurable, gamified stat rather than a vague SLA promise.
Hacker News Comment Review
Self-hosting alternatives (Forgejo, Tangled/AtProto Knot, bare PHP Git viewers) are the dominant response, with several commenters actively migrating away from GitHub.
Commenters debate whether a single uptime number is meaningful for a platform as large as GitHub, comparing it to rolling all AWS services into one metric.
Business continuity concern is real: at least one commenter reports considering GitHub Enterprise on-premises migration due to recurring cloud incidents.
Notable Comments
@reilly3000: GitHub cloud incidents are pushing at least some Enterprise customers toward on-premises evaluation.
@kedihacker: Argues a single incident counter obscures which GitHub services actually failed.
