Affected services and resolution
GitHub marked its September 13 incident resolved at 10:44 UTC. The affected products were API Requests, Issues, Pull Requests, Actions and Pages.
About 89 minutes elapsed between the first investigation notice and resolution. That is the public incident window; the exact start of customer impact and the number of affected users were not established in the reviewed updates.
Official update timeline
All times below are UTC on September 13, 2026.
| Time | GitHub update |
|---|---|
| 09:16 | Investigation opened for API Requests, Issues, Pages and Pull Requests. |
| 09:25 | Actions degradation added. |
| 09:36 | Database replication lag linked to authorization errors. |
| 10:26 | Reduced cluster load accompanied early recovery. |
| 10:28 | Pull Requests still degraded. |
| 10:44 | Incident resolved. |
GitHub described mitigation by reducing internal load. A detailed root-cause analysis was still pending when this report was reviewed. The early explanation should not be treated as the final postmortem.
Check your work after recovery
A provider recovery notice and a successful deployment are separate checks. Before retrying a job, inspect its result, logs and deployment destination so you do not repeat work that already succeeded.
GitHub supports rerunning failed or specific jobs. Choose the required retry after checking the previous attempt. If a workflow remains queued, use our Actions troubleshooting guide.
For a problem happening now, start with current GitHub status or GitHub Actions status. Keep this dated report as context when comparing your own error timestamps.