GitHub outage history
HostTracker's checkpoints recorded 1 outage of 5 minutes or longer on github.com in the last 90 days, 10 minutes of downtime in total. Every outage below has its own page with the checks that observed it. GitHub also posted 50 incidents on its own status page that our checks did not record as an outage; they are listed too, marked Official.
- Outages
- 1
- last 90 days
- Total downtime
- 10m
- summed across all outages
- Longest
- 10m
- single outage
Recorded outages
Last 90 days · 5 min and longer · UTC
- Official onlyIncident with Git Operations 1 hour 44 minutes
- Official onlyIncident with API, Git, and Copilot requests 31 minutes
- Official onlyElevated errors on GPT-5.4 Nano 38 minutes
- Official onlyIncident with Dependabot 2 hours 30 minutes
- Official onlyIncident with Actions, Pages, and Webhooks 1 hour 33 minutes
- Official onlyDisruption with GitHub services 32 minutes
- Official onlyDisruption with Copilot Cloud services 2 hours 30 minutes
- Official onlyDisruption with some Copilot services 1 hour 15 minutes
- Official onlyElevated 404s for GitHub App s2s tokens in git/API 1 minute
- Official onlyIncident with Actions, Issues, API Requests and Pull Requests 9 minutes
- Official onlyGrok 4.6 and Grok 4.7 Service Issues 42 minutes
- Official onlyDisruption with Dependabot Alerts 3 hours 53 minutes
- Official onlyDelayed Actions status check results 36 minutes
- Official onlyProject updates delayed 11 hours 2 minutes
- Official onlyIncident with Actions 1 hour 1 minute
- Official onlyDelays for Pull Requests and Webhook deliveries 4 hours 30 minutes
- Official onlyDegraded availability for GitHub Actions, GitHub Pages and Webhooks 1 hour 8 minutes
- Official onlyCCA Automations are delayed when starting an Automation 24 minutes
- Official only[Retroactive] Actions workflow run logs delayed 1 minute
- Official onlyIncident with Actions, Pull Requests and Issues 7 hours 54 minutes
- Official onlyIncident with Git Operations, Issues, Actions and Pull Requests 48 minutes
- Official onlyIncident with Git Operations, Pull Requests and Actions 1 hour 10 minutes
- Official onlySeveral services are degraded 3 hours 43 minutes
- Official onlyDisruption with some GitHub services 1 hour 45 minutes
- Official onlyIncident with Actions 3 hours 38 minutes
- Official onlyActions Job Delays 3 hours 9 minutes
- Official onlyElevated request latency 20 minutes
- Official only[Retroactive] Actions workflow run failures after deployment gate approvals 1 minute
- Official onlyCopilot Code Review is unable to complete reviews 52 minutes
- Official onlyDisruption with billing information updates 3 hours 50 minutes
- Official onlyIncident across several services 18 hours 44 minutes
- Official onlyIncident with Pull Requests 1 hour 9 minutes
- Official onlyElevated rate of errors for OpenAI models provided by Copilot 50 minutes
- Official onlyDegradation with Gemini 3.8 Flash 10 hours 28 minutes
- Official onlyDisruption with some GitHub services 49 minutes
- Official onlyDisruption with some GitHub services 1 hour 30 minutes
- Official onlyActions Larger Runner Jobs for some customers may be slow to start 55 minutes
- 10 minutes
- Official onlyIncident with several GitHub Services 1 hour 29 minutes
- Official onlyDegradation in repos contents API 21 minutes
- Official onlyDisruption with Copilot Code Review 1 hour 48 minutes
- Official onlyIncident with Grok Copilot AI Model Provider 2 hours 54 minutes
- Official onlyDelays in commit processing 1 hour 1 minute
- Official onlyElevated rate of errors for OpenAI models provided by Copilot 42 minutes
- Official onlyIncident with Copilot AI Model Providers 2 hours 8 minutes
- Official onlyDisruption with GitHub Billing 20 hours 7 minutes
- Official onlyIncident with Actions and Pull Requests 1 hour 30 minutes
- Official onlyIncident with Actions 2 hours 50 minutes
- Official onlyDisruption with some GitHub services 59 minutes
- Official onlyActions delays in starting runs 38 minutes
- Official onlyElevated errors on Fable 5 due to upstream provider 47 minutes
Rows tagged "Official" carry the incident GitHub posted on its own status page, quoted and linked on the incident's page; an "Official only" row is one our checkpoints did not record as an outage. Every other outage was measured by HostTracker's own checkpoints and confirmed from several locations before it was recorded. Open any row for its timeline, or see the live GitHub status and the 90-day uptime chart.
Independent measurement by HostTracker. Not affiliated with GitHub.
Know before your users do when GitHub breaks
Start a free trial and get an alert the moment a service you depend on goes down - confirmed from several locations first.