Applying GitOps principles to your operations
Could we use our Git repository as the source of truth for operational tasks, and somehow reconcile changes with our real-world view?
In December, we did not experience any incidents that resulted in degraded performance across GitHub services. This report sheds light into an incident that impacted customers using GitHub Packages and GitHub Pages in November.
In December, we did not experience any incidents that resulted in degraded performance across GitHub services. This report sheds light into an incident that impacted GitHub Packages and GitHub Pages in November.
November 25 16:34 UTC (lasting 1 hour and 56 minutes)
On November 25, 2022 at 14:39 UTC, our alerting systems detected an incident that impacted customers using GitHub Packages and GitHub Pages. The GitHub Packages team initially statused GitHub Packages to yellow, and after assessing impact, it statused to red at 15:06 UTC.
During this incident, customers experienced unavailability of packages for container, npm, and NuGet registries. We were able to serve requests for RubyGems and Maven registries. GitHub Packages’ unavailability also impacted GitHub Pages, as they were not able to pull packages, which resulted in CI build failures. Repository landing pages also saw timeouts while fetching packages information.
GitHub Packages uses a third-party database to store data for the service and the provider was experiencing an outage, which impacted GitHub Packages performance. The first responder connected with the provider’s support team to learn more about the region specific outage. The provider then mitigated the issue before the first responder could do the failover to another region. With the mitigation in place, GitHub Packages started to recover along with GitHub Pages and the repository landing pages.
As follow up action items, the team is exploring options to make GitHub Pages and repository landing pages more resilient to GitHub Packages outages. We are also investigating options where failovers can be performed quickly and automatically in case of regional outages.
Please follow our status page for real-time updates on status changes. To learn more about what we’re working on, check out the GitHub Engineering Blog.