Understanding GitHub's April 2026 Availability: Key Incidents and Lessons
In April 2026, GitHub experienced ten separate incidents that led to degraded performance across various services. While most issues were minor, two significant events on April 1 captured widespread attention: a prolonged code search outage and a brief but impactful audit log failure. This Q&A breaks down what happened, why it happened, and how GitHub is responding to prevent similar problems in the future.
How many incidents affected GitHub in April 2026, and what was the overall impact?
Throughout April, GitHub recorded ten distinct incidents that caused service degradation. The most severe was a code search outage lasting nearly nine hours on April 1, during which search queries completely failed for over two hours and returned stale results for several more. A separate audit log incident on the same day left history inaccessible via the API and web UI for about 28 minutes. Despite these disruptions, no data was permanently lost—GitHub’s core Git repositories remained intact, and all audit log events were eventually written and streamed successfully. The company has since published a detailed blog post covering the major April 1 events and enhanced the information shown on the GitHub status page.
What caused the code search outage on April 1, 2026?
The code search failure stemmed from a routine infrastructure upgrade to the messaging system that powers search indexing. An automated change was applied too aggressively, causing a coordination failure between internal services. This halted search indexing, and results quickly became stale. While engineering teams worked to recover the messaging infrastructure, an unintended service deployment inadvertently cleared internal routing state. That small mistake escalated the staleness problem into a complete outage—100% of search queries failed between 14:40 and 17:00 UTC. Notably, the underlying Git repositories were never affected; the search index is a secondary, derived data source that depends on continuous updates from the messaging system.
How did GitHub restore the code search service?
Recovery involved several coordinated steps. First, the team performed a controlled restart of the messaging infrastructure, reestablishing proper coordination between services. Next, they reset the search index to a point-in-time snapshot taken before the disruption began. Because the index is derived from Git repositories (which remained healthy throughout the incident), no source data was lost. Re-indexing commenced immediately, but it took until 23:45 UTC for search results to reflect the current state of all repositories. During the intervening period (17:00–23:45 UTC), search returned results, but those results did not include any repository changes made after approximately 07:00 UTC that day. Full current data was restored by late evening.
What happened with the audit log service on April 1, 2026?
Between 15:34 and 16:02 UTC, GitHub’s audit log service lost connectivity to its backing data store because of a failed credential rotation. During that 28-minute window, historical audit log data was unavailable through both the API and the web UI, resulting in 5xx errors for 4,297 API actors and 127 github.com users. Additionally, events generated during the outage were delayed by up to 29 minutes in the main interface and event streaming. However, GitHub confirmed that no audit log events were lost—all events were eventually written and streamed successfully. Customers using GitHub Enterprise Cloud with data residency were not impacted. The team was alerted six minutes after the failure and took immediate action to restore the connection.
What steps is GitHub taking to prevent similar incidents in the future?
GitHub has outlined several specific improvements targeting the root causes of both the code search and audit log incidents. For code search, they are implementing more gradual upgrades with better health checks that can catch problems before they cascade. They are also adding deployment safeguards to prevent unintended changes during active incidents, faster recovery tooling to reduce the time needed to restore service, and better traffic isolation to limit the impact of unexpected spikes. For the audit log issue, the credential rotation process will be reviewed and hardened to prevent premature authentication loss. Collectively, these measures aim to make infrastructure changes more controlled and to shorten the window of disruption when failures do occur.
Were any customer data lost during these incidents?
No. GitHub has explicitly stated that no repository data was lost during the code search outage. The search index is a secondary, cached index derived from the primary Git repositories, which remained fully operational throughout the event. Similarly, in the audit log incident, all events were eventually written and streamed successfully—none were lost, even though access was temporarily blocked. This distinction is important: while service availability was degraded, the durability and integrity of customer data were preserved. The incidents highlight a failure in the systems that provide access to data (search, audit log retrieval), not in the core data storage itself.
For ongoing updates, check the GitHub status page and the April 2026 availability blog post.