Incident with Actions, Issues, API Requests and Pull Requests
On October 26, 2026, between 17:52 and 18:49 UTC, some customers experienced slower response times across the Checks API, Issues, Pull Requests, GraphQL, Notifications, and related services. Impact was limited to requests served from one of our datacenter sites. Some GraphQL requests failed, and some notifications were delayed. Among GraphQL requests served from the affected site, the average error rate was 5.68%, with a peak five-minute error rate of 14.32%. <br /><br />Unusually large requests for contribution history overwhelmed part of our database infrastructure, causing a backlog. Requests continued reaching the slow component rather than being redirected to healthy systems. Restarting the component restored its performance. The incident was marked mitigated at 18:46 UTC and resolved at 18:57 UTC. <br /><br />We are adding limits to these requests, improving alerts to detect overload sooner, and strengthening automatic redirection to healthy systems.
We saw elevated latency affecting REST API requests, including Checks, and some Actions, Issues, Pull Requests, and GraphQL functionality. Some requests may have been slow or timed out, and notifications were delayed from 17:59 UTC - 18:40 UTC.<br /><br />We mitigated the issue and all services have recovered.
The degradation has been mitigated. We are monitoring to ensure stability.
We are investigating reports of degraded performance for API Requests, Actions, Issues and Pull Requests
關於這些資料
GitHub 的事件資料來自哪裡?
資料來自GitHub 官方狀態頁,每 5 分鐘同步,保留最近 90 天。詳情頁提供官方記錄連結。
如何解讀元件狀態列?
每格代表一天。綠色表示無異常記錄,黃色表示部分降級,紅色表示較大故障,灰色表示無資料。可用率統計涵蓋最近 90 天。
為什麼有些服務查不到歷史?
只收錄提供公開官方狀態頁的服務。
持續時間如何計算?
按官方標記的開始至復原時間計算。進行中的事件計算至目前時間。跨天事件會顯示在每個受影響的日期。
完整的服務清單與即時狀態見狀態總覽