Disruption with some GitHub services
This summary is created by Generative AI and may differ from the actual content.
Overview
On April 28, 2026, from ~12:41 UTC to 17:09 UTC, GitHub Actions jobs using the Standard Ubuntu 22 and Ubuntu 24 hosted runners experienced start‑up delays. About 8 % of those jobs saw delays longer than five minutes or failed to start. Larger and self‑hosted runners were unaffected. The issue was traced to a performance regression in the VM reimage process, which reduced the pool of available runners. A rollback to a known‑good image version restored capacity and resolved the problem.
Impact
Approximately 8 % of hosted Ubuntu 22 and Ubuntu 24 runner jobs experienced delays >5 minutes or failures. The degradation was limited to these runner types; other hosted labels and self‑hosted runners continued to operate normally. No broader service outage was reported.
Trigger
A performance regression introduced in the VM reimage process caused the reimage step to take longer than expected, thereby lowering the overall capacity of runners available to pick up new jobs.
Detection
Monitoring systems reported capacity constraints and elevated wait times for hosted ubuntu‑latest and ubuntu‑24.04 runners. The degraded performance was observed through these metrics, prompting the on‑call team to investigate.
Resolution
The team rolled back to a previously validated VM image, which immediately increased runner capacity and unblocked job starts. Additional mitigations were applied to unblock running actions, and the service was monitored until full recovery was confirmed.
Root Cause
The root cause was a regression in the VM reimage process that degraded performance, reducing the speed at which new runner instances could be provisioned. This regression lowered the available capacity for Ubuntu‑based hosted runners, leading to the observed delays and failures.