On February 25, 2026, between 15:05 UTC and 16:34 UTC, the Copilot coding agent service was degraded, resulting in errors for 5% of all requests and impacting users starting or interacting with agent sessions. This was due to an internal service dependency running out of allocated resources (memory and CPU). We mitigated the incident by adjusting the resource allocation for the affected service, which restored normal operations for the coding agent service.We are working to implement proactive monitoring for resource exhaustion across our services, review and update resource allocations, and improve our alerting capabilities to reduce our time to detection and mitigation of similar issues in the future.
On February 23, 2026, between 21:01 UTC and 21:30 UTC the Search service experienced degraded performance, resulting in an average of 3.5% of search requests for Issues and Pull Requests being rejected. During this period, updates to Issues and Pull Requests may not have been immediately reflected in search results. During a routine migration, we observed a spike in internal traffic due to a configuration change in our search index. We were alerted to the increase in traffic as well as the increase in error rates and rolled back to the previous stable index. We are working to enable more controlled traffic shifting when promoting a new index to allow us to detect potential limitations earlier and ensure these operations succeed in a more controlled manner.
Between 2026-02-23 19:10 and 2026-02-24 00:46 UTC, all lexical code search queries in GitHub.com and the code search API were significantly slowed, and during this incident, between 5 and 10% of search queries timed out. This was caused by a single customer who had created a network of hundreds of orchestrated accounts which searched with a uniquely expensive search query. This search query concentrated load on a single hot shard within the search index, slowing down all queries. After we identified the source of the load and stopped the traffic, latency returned to normal.To avoid this situation occurring again in the future, we are making a number of improvements to our systems, including: improved rate limiting that accounts for highly skewed load on hot shards, improved system resilience for when a small number of shards time out, improved tooling to recognize abusive actors, and capabilities that will allow us to shed load on a single shard in emergencies.
On February 23, 2026, between 15:00 UTC and 17:00 UTC, GitHub Actions experienced degraded performance. During the time, 1.8% of Actions workflow runs experienced delayed starts with an average delay of 15 minutes. The issue was caused by a connection rebalancing event in our internal load balancing layer, which temporarily created uneven traffic distribution across sites and led to request throttling. To prevent recurrence, we are tuning connection rebalancing behavior to spread client reconnections more gradually during load balancer reloads. We are also evaluating improvements to site-level traffic affinity to eliminate the uneven distribution at its source. We have overprovisioned critical paths to prevent any impact if a similar event occurs before those workstreams finish. Finally, we are enhancing our monitoring to detect capacity imbalances proactively.
On February 23, 2026, between 14:45 UTC and 16:19 UTC, the Copilot service was degraded for Claude Haiku 4.5 model. On average, 6% of the requests to this model failed due to an issue with an upstream provider. During this period, automated model degradation notifications directed affected users to alternative models. No other models were impacted. The upstream provider identified and resolved the issue on their end. We are working to improve automatic model failover mechanisms to reduce our time to mitigation of issues like this one in the future.
On February 20, 2026, between 17:45 UTC and 20:41 UTC, 4.2% of workflows running on GitHub Larger Hosted Runners were delayed by an average of 18 minutes. Standard, Mac, and Self-Hosted Runners were not impacted. The delays were caused by communication failures between backend services for one deployment of larger runners. Those failures prevented expected automated scaling and provisioning of larger hosted runner capacity within that deployment. This was mitigated when the affected infrastructure was recycled, larger runner pools in the affected deployment successfully scaled up, and queued jobs processed. We are working to improve the time to detect and diagnose this class of failures and improve the performance of recovery mechanisms for this degraded network state. In addition, we have architectural changes underway that will enable other deployments to pick up work in similar situations, so there is no customer impact due to deployment-specific infrastructure issues like this.
On February 20, 2026, between 07:30 UTC and 11:21 UTC, the Copilot service experienced a degradation of the GPT 5.1 Codex model. During this time period, users encountered a 4.5% error rate when using this model. No other models were impacted.The issue was resolved by a mitigation put in place by the external model provider. GitHub is working with the external model provider to further improve the resiliency of the service to prevent similar incidents in the future.
This incident has been resolved. Thank you for your patience and understanding as we addressed this issue. A detailed root cause analysis will be shared as soon as it is available.
On February 17, 2026, between 17:07 UTC and 19:06 UTC, some customers experienced intermittent authentication failures affecting GitHub Actions, parts of Git operations, and other authentication-dependent requests. On average, the Actions error rate was approximately 0.6% of affected API requests. Git operations ssh read error rate was approximately 0.29%, while ssh write and http operations were not impacted. During the incident, a subset of requests failed due to token verification lookups intermittently failing, leading to 401 errors and degraded reliability for impacted workflows.The issue was caused by elevated replication lag in the token verification database cluster. In the days leading up to the incident, the token store’s write volume grew enough to exceed the cluster’s available capacity. Under peak load, older replica hosts were unable to keep up, replica lag increased, and some token lookups became inconsistent, resulting in intermittent authentication failures.We mitigated the incident by adjusting the database replica topology to route reads away from lagging replicas and by adding/bringing additional replica capacity online. Service health improved progressively after the change, with GitHub Actions recovering by ~19:00 UTC and the incident resolved at 19:06 UTC.We are working to prevent recurrence by improving the resilience and scalability of our underlying token verification data stores to better handle continued growth.
On February 13, 2026, between 21:46 UTC and 22:58 UTC (72 minutes), the GitHub file upload service was degraded and users uploading from a web browser on GitHub.com were unable to upload files to repositories, create release assets, or upload manifest files. During the incident, successful upload completions dropped by ~85% from baseline levels. This was due to a code change that inadvertently modified browser request behavior and violated CORS (Cross-Origin Resource Sharing) policy requirements, causing upload requests to be blocked before reaching the upload service.We mitigated the incident by reverting the code change that introduced the issue.We are working to improve automated testing for browser-side request changes and to add monitoring/automated safeguards for upload flows to reduce our time to detection and mitigation of similar issues in the future.