GitHub is rolling out significant updates to its status page to improve how it communicates service health during and after incidents. The changes are guided by transparency, accuracy, and timeliness. Three key enhancements include a new incident severity level, per-service uptime metrics, and more granular insights—starting with a dedicated component for Copilot AI model providers. These improvements aim to give developers a clearer, more precise understanding of platform reliability.
What three changes is GitHub making to its status page?
GitHub is introducing three major updates to its status page to boost transparency about service health. First, they are adding a new incident classification called Degraded Performance, which sits alongside the existing Partial Outage and Major Outage states. Second, they are now publishing per-service uptime percentages over the last 90 days directly on the status page. Third, they are offering more granular insights into service disruptions, starting with a dedicated “Copilot AI Model Providers” component that clarifies availability issues related to model providers. These changes are part of GitHub’s ongoing effort to improve incident communication and provide developers with more accurate and timely information about platform health.
What is the new “Degraded Performance” incident state and why was it introduced?
The Degraded Performance state describes a situation where a service is operational but impaired—for example, users may experience elevated latency, reduced functionality, or intermittent errors affecting a small percentage of requests. Previously, even minor disruptions were classified as at least a “Partial Outage,” which inaccurately suggested widespread unavailability. This new tier allows GitHub to more precisely reflect the spectrum of issues that can occur, giving users a more accurate understanding of actual impact. It also does not count toward uptime calculations, unlike Partial and Major Outages. By adding this intermediate level, GitHub aligns its incident classification with industry best practices and provides clearer communication during minor incidents.
How does the new three-tier incident classification system work?
The three-tier system consists of Degraded Performance, Partial Outage, and Major Outage. Degraded Performance means the service is functional but impaired, such as elevated latency or occasional errors for a small user group. Partial Outage indicates a significant portion of the service is unavailable or severely impacted for many users. Major Outage means the service is broadly unavailable, affecting most or all users. Each level carries a specific downtime weight for uptime calculations: Major Outage counts as 100% downtime duration, Partial Outage as 30%, and Degraded Performance as 0%. This classification helps GitHub communicate severity more accurately and helps users understand the real impact of incidents.
What per-service uptime metrics are now available on the status page?
GitHub now publishes uptime percentages for each individual service over the past 90 days directly on its status page. These percentages are calculated based on the number of incidents, their severity, and their duration for each service. The calculation uses industry-standard methods, weighting each incident type differently: Major Outage counts the full duration as downtime (100% weight), Partial Outage counts 30% of the duration, and Degraded Performance counts 0%. For example, if a service experienced a 1-hour Partial Outage over a 90-day period, it would count as 18 minutes of effective downtime in the uptime calculation—not the full hour. This gives users a clearer view of each service’s recent reliability record.
How is GitHub improving granularity for services like Copilot?
To provide more precise information during service disruptions, GitHub is adding dedicated components for specific services on the status page. The first example is a “Copilot AI Model Providers” component, which will separately report availability issues related to the AI models that power GitHub Copilot. This means that if a model provider experiences an outage, users will see it reflected distinctly rather than as part of a broader Copilot service issue. Such granularity helps developers quickly identify whether a problem stems from GitHub’s own infrastructure or from a third-party provider. GitHub plans to extend this granular approach to other services over time, ensuring clear communication about the specific source of any disruption.
What principles guided GitHub’s approach to these status page updates?
GitHub’s enhancements are driven by three core principles: transparency, accuracy, and timeliness. By introducing the Degraded Performance tier, they improve accuracy by avoiding over-classification of minor issues as partial outages. Publishing per-service uptime metrics adds transparency, allowing users to assess reliability of each service individually. Adding granular components like the Copilot AI model provider component ensures timeliness by making it clear exactly where a problem originates. These changes are part of broader reliability investments following past availability issues. GitHub aims to give developers better insight into platform health, helping them trust the status page information and make informed decisions during incidents.