Incident with Pull Requests
This summary is created by Generative AI and may differ from the actual content.
Overview
On April 23, 2026, between 16:05 UTC and 20:43 UTC, the Pull Requests service experienced a regression affecting merge queue operations, resulting in incorrect merge commits when the merge group contained more than one PR, with 2,092 pull requests affected.
Impact
2,092 pull requests were affected, with changes from previously merged PRs and prior commits inadvertently reverted by subsequent merges, but the issue did not affect pull requests merged outside of merge queue, nor merge queue groups using the merge or rebase methods.
Trigger
The regression was introduced by a new code path that adjusted merge base computation for merge queue ref updates, which was intended to be gated behind a feature flag for an unreleased feature, but the gating was incomplete.
Detection
The issue was identified through customer reports, as it was not detected by existing automated monitoring, and it took approximately 3 hours and 33 minutes to identify the issue.
Resolution
The incident was mitigated by reverting the code change and force-deploying the fix across all environments, and after resolution, affected repositories were identified and sent targeted remediation instructions to repository administrators with step-by-step recovery guidance.
Root Cause
The regression was caused by the incomplete gating of a new code path, which was intended to be gated behind a feature flag for an unreleased feature, and the lack of automated checks to validate merge correctness for multi-PR squash groups, which allowed the defect to surface only in production.