GitHub favicon

GitHub outage on 2026-10-21

github.com
October 2026 21 Major incident

GitHub's own status page logged 4h 30m of delays to Pull Requests, Webhooks and Actions on 21 October 2026, caused by overload on GitHub's event streaming platform. If your product relies on webhook deliveries or pull request updates from GitHub, this was on GitHub's side, not yours.

Started

14:52 UTC

Duration

Lasted 4h 30m

Source

IsDown

Next time

GitHub is degraded. Is your product working?

An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.

50 monitors for your own websites. No card.

Affected components: Webhooks, Pull Requests, Actions

What happened?

GitHub opened the incident at 14:52 UTC, reporting degraded performance across Actions, Pull Requests and Webhooks. By 15:36 GitHub identified increased load on its event streaming platform as the contributing factor, creating hotspots, and said it was shifting traffic and adding processing capacity. Pull request updates and webhook deliveries began recovering from around 16:17, and GitHub marked Pull Requests as operating normally at 16:33. Actions and webhook deliveries followed, both confirmed normal by 16:57 and 16:59. Audit log events lagged well behind the rest of the system: at 16:25 GitHub reported delays of up to 100 minutes, rising to 110 minutes by 16:34, and still running close behind at 100 minutes by 16:59. GitHub estimated at 17:09 that clearing the backlog would take two to three hours. Delays dropped below 90 minutes by 18:18 and below 70 minutes by 18:52, with GitHub projecting full recovery in about 40 minutes at that point. The incident was marked resolved at 19:22, 4 hours and 30 minutes after it opened, with GitHub saying a root cause analysis would follow once available. No such analysis has been published since. No user reports were filed during this window on UptimeRobot's own records.

Learning

UptimeRobot's own probes against GitHub's api-auth and raw-cdn components, run from the EU and North America every 60 seconds, showed 100% uptime over 24 hours and 7 days, and 99.999% over 30 days, with no registration against this incident. That is not a contradiction. Those probes check whether GitHub's authentication and content delivery endpoints respond, and they did, throughout. The failure sat deeper in GitHub's infrastructure, in the event streaming platform that feeds webhook deliveries, pull request update notifications and audit log processing, a layer that simple endpoint checks cannot see because the endpoints themselves stayed up and reachable. If you were only watching your own webhook receiver or your own API calls, you would have seen requests succeed while the events behind them queued up for hours, particularly audit logs, which lagged over an hour after everything else recovered. This is the shape to recognise: your side green, the provider's internal processing backed up, and nothing in a simple reachability check to flag it. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.

4.7
stars out of 5
284+ reviews on

Start monitoring in 30 seconds.

There's nothing to install. No credit card required. 50 monitors for free.