GitHub Outage History

GitHub experienced intermittent authentication failures affecting GitHub App server-to-server tokens, GitHub Actions authentication tokens, and git operations. Some customers may have encountered intermittent API request failures when using these tokens, impacting automated workflows, CI/CD pipelines, and code operations that rely on token-based authentication. The incident began at Feb 17, 2026 5:51 PM UTC and lasted for about 1 hour and 15 minutes, with recovery observed by the final update at 7:06 PM UTC. Affected components included Actions and Git Operations.

An outage briefly affected GitHub file upload services, impacting users attempting to upload files. The disruption began at 10:36 PM UTC and lasted about 25 minutes, with services restored by 11:01 PM UTC. GitHub posted two status updates describing the disruption and the subsequent restoration. Early user reports indicated the issue started before the official acknowledgment, but the impact was limited to file upload workflows rather than a full platform outage.

GitHub experienced degraded performance impacting a subset of services. Users reported slower operation and intermittent issues across affected features. The incident began at Feb 12, 2026 6:36 PM UTC when the GitHub status page flagged a disruption with some services and indicated investigators were looking into reports of impacted performance. The outage lasted about 2 hours and was resolved by Feb 12, 2026 8:36 PM UTC.

During the incident, GitHub issued 2 status updates to communicate the situation and recovery. User reports came from multiple regions, including Boston, United States, and Le Vaudreuil, France, suggesting international reach. While not all GitHub services were affected, the degradation disrupted workflows for affected users during the window.

GitHub Copilot completions and inline suggestions experienced degraded availability in some regions. Users could see intermittent or missing Copilot suggestions during the incident. The outage began on 2:11 PM UTC and was resolved by 4:51 PM UTC, lasting approximately 2 hours and 40 minutes. The issue affected Copilot features across multiple regions.

GitHub issued two status updates during the incident, describing intermittent disruption and work to resolve the issue. The service returned to normal following the final update as Copilot features stabilized.

The GitHub service experienced a brief outage that prevented users from downloading repository archives containing Git LFS objects. The disruption began at Feb 12, 2026 10:42 AM UTC and was resolved by Feb 12, 2026 11:16 AM UTC, lasting about 34 minutes. User reports were present around 9:30 AM UTC prior to the incident, indicating some early visibility from the community. The impact appeared limited to archival downloads rather than a full platform outage.

GitHub’s status page published two official updates during the incident, acknowledging the disruption and that engineers were investigating, followed by confirmation of restoration. The outage primarily affected attempts to download archives with Git LFS objects rather than indicating a broader service failure.

The GitHub Codespaces outage began at Feb 12, 2026 7:56 AM UTC with degraded availability across the Codespaces environment. The issue escalated to degraded performance, and by Feb 12, 2026 9:46 AM UTC the status page listed Codespaces as degraded while teams continued to investigate. The outage lasted about 2 hours, ending with a recovery around Feb 12, 2026 9:56 AM UTC.

GitHub issued three status updates during the incident to communicate the progression: the initial notification of investigation, a follow-up noting degraded performance, and the final recovery confirmation. The disruption affected the Codespaces component and impacted users worldwide, with reported access and usage issues across multiple regions.

The GitHub status page reported a disruption affecting some services starting Feb 11, 2026 7:01 PM UTC, leading to degraded performance and intermittent access for a subset of GitHub features. Users in Ottawa and Kowloon reported slower responses and occasional access issues, indicating a broad yet partial impact. The incident persisted for about 6 hours and was resolved by Feb 12, 2026 1:01 AM UTC.

During the incident, the GitHub Status page issued two updates describing the disruption and its resolution. The event drew limited user feedback, reflecting a small but noticeable impact on a portion of users.

GitHub experienced increased latency in GraphQL traffic, degrading API Requests and Copilot for some users. The incident began at Feb 11, 2026 3:31 PM UTC and prompted status changes on the affected components. User reports started around 3:00 PM UTC, describing slower GraphQL responses and Copilot-related messages.

Copilot recovered by 3:51 PM UTC, while API Requests returned to normal later at 5:16 PM UTC. The GitHub status page issued 3 updates during the incident, documenting ongoing investigation and eventual restoration. The outage lasted about 1 hour and 45 minutes and affected GraphQL-dependent workflows during the window.

The GitHub Copilot feature experienced failures as requests could not be processed due to an unsupported API version. The issue began at 1:54 AM UTC and affected Copilot usage globally, with error messages indicating the API version is no longer supported. Users reported Copilot errors and related failures across multiple regions, reflecting a broad impact on how Copilot interacted with GitHub services. The incident was observed through user reports rather than official status updates.

GitHub's Pull Requests component experienced degraded performance from Feb 10, 2026 3:11 PM UTC to Feb 10, 2026 4:01 PM UTC, impacting PR creation, reviews, and related checks for about 50 minutes. Users across multiple regions reported issues with PR workflows and CI steps during the incident, including sporadic failures in some actions steps and unusual page behavior.

GitHub communicated via two status updates: the first at Feb 10, 2026 3:11 PM UTC noting Disruption with some GitHub services and that Pull Requests was experiencing degraded performance while investigations continued, and the second at Feb 10, 2026 4:01 PM UTC indicating the Pull Requests component had returned to normal. The incident affected 1 component and included 2 status updates.