GitHub reported a service disruption that hit its Actions platform in the East US region. The outage began after operators observed elevated queue times and frequent job failures on Standard Hosted Runners. Within hours, the company confirmed that about 10% of runs were affected, prompting an immediate investigation.

Lead engineers reached out to the Azure compute team, noting that private networking setups could fail over to a neighboring region to mitigate the impact. The team rolled back a recent capacity upgrade and deployed a mitigation patch that reduced queue latency by roughly 30 %. The fix was rolled out across the affected region by late afternoon.

At 15:12 UTC, GitHub posted an update indicating that the issue still impacted about 8% of runs, though the failure rate had dropped from the initial 10%. Developers with private networking could switch to a different Azure region, but most teams continued to rely on the standard runners until the service stabilized.

The incident exposed a bottleneck in GitHub’s scaling strategy for the East US region and highlighted the importance of multi-region failover for CI/CD workloads. By mid‑day, the company reported that queue times had returned to baseline levels and no further disruptions were observed.