GitHub Outage History

GitHub reported degraded performance for Git Operations and Actions beginning at Jul 29, 2025 10:46 AM UTC as the status page noted an ongoing investigation into increased 429 errors. Users attempting Git pushes and running Actions workflows could experience slower operations or intermittent failures while the issue was investigated. The incident lasted about 1 hour and 20 minutes and included two official status updates as the situation evolved.

The incident began around Jul 28, 2025 5:31 PM UTC when Git Operations entered a degraded state. Over the next hours, the impact expanded to API Requests, Issues, and Pull Requests as additional components were degraded. By 10:40 PM UTC, API Requests, Issues, and Pull Requests were listed as degraded and investigations continued. A related note at 1:26 AM UTC indicated GitHub Enterprise Importer migrations were stalled. The affected components gradually returned to normal, with all four core areas reported back to up by Jul 29, 2025 2:06 AM UTC. The outage lasted about 8 hours and 35 minutes, during which users could have experienced slower responses or partial functionality in Git operations, workflow-related API calls, issues, and pull requests. The provider issued 4 official updates during the incident.

GitHub experienced a brief degradation in Actions Hosted Runners, with provisioning delays causing jobs to start later than expected. The incident began at 3:36 PM UTC when the Actions component moved to a degraded state, with users seeing delays of over 5 minutes for jobs starting. The issue was addressed and the service returned to normal by 4:31 PM UTC, for a total duration of about 55 minutes. The provider posted two status updates during the incident documenting investigation and subsequent recovery.

GitHub Copilot experienced a brief degradation affecting the Claude Sonnet 4 model across Copilot Chat, VS Code, and other Copilot products, while other Copilot models remained available. The incident began at 6:36 PM UTC and was resolved by 6:51 PM UTC, lasting about 15 minutes. The issue was attributed to an upstream model provider, with no impact to other Copilot models. The status page logged two updates during the event, describing the degraded availability and restoration.

An incident at GitHub began around 7:15 AM UTC with degraded performance reported for Issues and Webhooks. The impact expanded to additional GitHub features over the next couple of hours, including Codespaces, Pull Requests, Packages, and Copilot, as the status page published continuous updates. Users could experience slower issue tracking and webhook processing, and some workflows (for example queued builds) did not start promptly.

The outage lasted about 2 hours and 36 minutes, with the GitHub status page issuing nine updates as services gradually recovered. By the end of the window, Issues, Pull Requests, Packages, Copilot, Codespaces, and API Requests had returned to normal operations, with the final updates confirming a full recovery around 09:51 UTC. User reports were present but limited during the incident.

GitHub Actions experienced degraded performance for a short period, impacting automated workflows that rely on Actions. Users could have seen slower builds or intermittent failures while the issue was active. The incident began around Jul 8, 2025 4:05 PM UTC with an investigation into degraded performance for Actions, as noted on GitHub's status page. The service issued two updates during the incident and returned to normal around Jul 8, 2025 4:46 PM UTC, for a duration of about 40 minutes.

GitHub Copilot experienced degraded performance for about 35 minutes, from Jun 18, 2025 10:41 PM UTC to Jun 18, 2025 11:16 PM UTC. The GitHub Status page noted a disruption affecting some GitHub services and confirmed Copilot was experiencing degraded performance as engineers investigated. The issue mainly affected Copilot's AI code-completion features, with users potentially seeing slower or less responsive suggestions. The status page issued a single update during the event, and Copilot returned to normal after the incident. User reports were limited, indicating only a small number of users observed Copilot-related issues.

GitHub reported partial degradation of the Issues experience, with users likely seeing slower performance when creating, viewing, or managing issues. The outage began on 4:21 PM UTC and was resolved by 5:46 PM UTC, lasting 1 hour and 25 minutes and affecting the Issues component. The incident was communicated via the GitHub Status page as investigators worked to diagnose the degraded behavior.

An outage began around 7:42 PM UTC with Copilot experiencing degraded performance. The issue then expanded to affect multiple GitHub services, including Copilot, Pages, API Requests, Pull Requests, Issues, Webhooks, and Codespaces. Users reported errors and degraded experiences across developer workflows, with some pages returning errors and PRs failing checks. The incident lasted roughly 44 minutes and recovery progressed over the next several minutes, with all listed services returning to normal by around 8:26 PM UTC. The incident triggered broad user reports from around the world, indicating widespread impact on coding, builds, and collaboration.

Recovery progressed through a series of status updates from GitHub, with API Requests recovering first and other services following. The provider issued eight status updates during the incident, reflecting ongoing investigation, degraded performance, and eventual restoration across seven components: Copilot, Pages, API Requests, Pull Requests, Issues, Webhooks, and Codespaces.

The GitHub status page reported a partial outage affecting GitHub Actions and Copilot chat models. Users were unable to generate attestations from public repositories due to a broader outage with partner services, and Copilot requests experienced degraded performance. The incident began around 6:56 PM UTC and a formal status update to a partial outage was posted at 7:02 PM UTC. The disruption lasted about 2 hours and 15 minutes, with Copilot services returning to normal by 9:11 PM UTC. The issue appeared linked to a broader disruption impacting related partner services.