GitHub suffered simultaneous degraded availability across Webhooks, Actions, and Copilot on Apr 23, 2026, with root cause identified by 16:52 UTC.
Key Takeaways
Three core services affected: Webhooks, Copilot, and Actions – spanning CI/CD pipelines, AI tooling, and event-driven integrations simultaneously.
Incident timeline spanned ~40 minutes from first report (16:12 UTC) to root cause identification (16:52 UTC), with mitigation still in progress at that point.
Actions degradation was partial, not a full outage – some jobs completed, some failed, with no clear pattern distinguishing success from failure.
GitHub’s status page tracked the incident in real time across four update posts, consistent with their standard incident communication protocol.
Hacker News Comment Review
Commenters flagged that partial Actions failures are worse operationally than a full outage: jobs consuming 10+ minutes of runner time before failing waste quota and delay feedback loops.
There is growing sentiment that GitHub outages are frequent enough that the platform’s reliability posture is becoming a business risk, not just an inconvenience – a few commenters noted they have already migrated to GitLab for self-hosted CI runners.
Discussion touched on SLA math: GitHub would need roughly 16 additional hours of downtime in the 90-day rolling window to breach two 9s, which commenters treated as a meaningful threshold worth watching.
Notable Comments
@AnkerSkallebank: partial failures are harder to handle than clean failures – “kind of wish they would just fail outright, instead of running for 10 minutes and then failing.”
@argee: already migrated to GitLab; cites free self-hosted CI runners as a concrete operational advantage over GitHub.
