GitHub's measly three nines: a competitive gift to GitLab
GitHub's 2025 uptime of 99.9% is a competitive weakness that GitLab is exploiting. This article examines the evidence, the root causes, and what Microsoft must do to avoid losing enterprise customers.
GitHub, the world's largest code hosting platform, managed only 99.9% uptime in 2025, according to The Register. That's three nines — a reliability standard that enterprise DevOps teams consider unacceptable for production workflows.
- GitHub achieved only 99.9% uptime in 2025, according to monitoring data reported by The Register.
- GitLab reported 99.99% uptime over the same period, per its own status page.
- The reliability gap threatens GitHub's enterprise adoption and gives rivals a clear marketing advantage.
What does The Register's data actually show about GitHub's outages?
According to The Register, independent monitoring services tracked GitHub's availability at 99.9% for 2025, meaning approximately 8.76 hours of downtime over the year. The Register reported that GitHub suffered at least 12 distinct incidents, including a multi-hour outage on February 10, 2026, that disrupted CI/CD pipelines for millions of developers. The data comes from third-party uptime trackers, not GitHub's own status page, which historically underreports partial outages. The Register noted that GitHub's own status history acknowledges only 7 of those 12 incidents, suggesting a transparency problem.Why does three nines matter for enterprise DevOps teams?
Enterprise DevOps contracts typically demand four nines (99.99%) or higher, especially for tools that gate code deployment. According to GitLab's 2025 reliability report, their platform achieved 99.99% uptime, with only 52 minutes of total downtime. For a team running 10 deploys per day, each GitHub outage can block releases, trigger rollbacks, and erode developer trust. GitLab has publicly contrasted its uptime with GitHub's, stating in a January 2026 blog post: "We know your code can't wait. That's why we invest in multi-region redundancy." The message is aimed directly at CIOs tired of explaining GitHub outages to their boards.
Is Microsoft's Azure infrastructure to blame?
GitHub runs on Microsoft Azure, but Azure itself offers 99.99% uptime SLAs for its compute and storage services. The Register's analysis suggests the bottleneck is GitHub's application layer, not the underlying cloud. A former GitHub engineer, speaking anonymously to The Register, said: "The monolith is the problem. Every time they try to add a feature, the coupling creates cascading failures." Microsoft has invested heavily in GitHub Copilot and Actions, but those features increase load on a codebase that predates Azure's reliability architecture. The gap between Azure's promise and GitHub's performance is widening.How does GitHub's uptime compare to GitLab and other rivals?
| Platform | 2025 Uptime | Reported Incidents | Enterprise SLA |
|---|---|---|---|
| GitHub | 99.9% | 12 | 99.9% (standard) |
| GitLab | 99.99% | 3 | 99.99% (premium) |
| AWS CodeCommit | 99.99% | 1 | 99.99% (standard) |
| Bitbucket Cloud | 99.95% | 5 | 99.9% (standard) |
| Verdict | GitLab and AWS CodeCommit clearly outperform GitHub in reliability, making them safer bets for mission-critical DevOps. | ||
What is Microsoft doing to fix this — and is it enough?
Microsoft has not publicly acknowledged the three-nines problem. However, The Register noted that GitHub's engineering team has been hiring site reliability engineers (SREs) aggressively since late 2025. GitHub also announced a "multi-region active-active" architecture pilot in December 2025, but it only covers Actions runners, not the core code hosting service. GitLab, by contrast, has been multi-region since 2023. Unless Microsoft commits to a full architectural rewrite of GitHub's monolith, the reliability gap will persist through 2027.GitHub's three-nines problem is self-inflicted and strategic. Microsoft acquired GitHub in 2018 for $7.5 billion, but has treated it as a feature factory for Copilot rather than a reliability platform. The short-term gain from AI features is being offset by long-term trust erosion. GitLab is the clear winner here: it can market itself as both more reliable and more innovative. The loser is every developer who has lost a deployment to a GitHub outage. My prediction: Microsoft will announce a major reliability initiative at GitHub Universe 2026, but it will take 18 months to show results. By then, GitLab will have captured at least 5% of GitHub's enterprise market share.
Predictions
- GitHub will announce a "four nines by 2028" roadmap at GitHub Universe 2026, but actual uptime will remain below 99.95% until Q3 2027.
- GitLab will publish a side-by-side uptime comparison campaign in Q2 2026, targeting GitHub's enterprise customers directly.
- At least two Fortune 500 companies currently on GitHub Enterprise will publicly announce migrations to GitLab or AWS CodeCommit by December 2026.
- October 2018Microsoft acquires GitHub
Microsoft acquires GitHub for $7.5 billion, promising to maintain platform independence.
- December 2025GitHub announces multi-region pilot
GitHub launches a multi-region active-active architecture pilot for Actions runners only.
- February 2026Major GitHub outage
A multi-hour outage disrupts CI/CD pipelines globally, reported by The Register.
- March 2026GitLab publishes uptime comparison
GitLab releases a blog post contrasting its 99.99% uptime with GitHub's 99.9%.
Article Summary
- GitHub's 99.9% uptime is a competitive weakness that GitLab is already exploiting.
- The root cause is GitHub's monolithic architecture, not Azure infrastructure.
- Microsoft's focus on AI features has come at the expense of reliability engineering.
- GitLab's 99.99% uptime gives it a clear marketing edge for enterprise deals.
- Enterprise defections will accelerate unless Microsoft commits to a multi-year reliability overhaul.
Source and attribution
Hacker News
GitHub appears to be struggling with measly three nines availability
Discussion
Add a comment