On June 17, 2026, between approximately 03:35 UTC and 04:44 UTC, GitHub Copilot was degraded and most of its frontier chat models were temporarily unavailable across all regions. During this window, affected models either disappeared from the model picker in the web, editor, and CLI experiences, or returned a "model not available" error when selected. Customers could continue using GitHub Copilot by selecting one of the models that remained available. The incident occurred during off-peak hours, which limited the number of customers affected.This was due to a configuration change that our production system deemed invalid. We mitigated the incident by reverting the configuration change, after which the affected models returned automatically as the service reloaded the previous configuration.We are working to roll out configuration changes gradually with stronger validations, alerts on sudden drops in the number of available models, and automatically rolls back configuration changes that produce these alerts.
On June 16, 2026, between 17:20 UTC and 18:15 UTC, the Opus 4.8 model experienced degraded availability in GitHub Copilot. During this window, some requests to Opus 4.8 failed or errored. Other Copilot models were not affected and remained available as alternatives. This was caused by an issue with an upstream model provider. The upstream provider resolved the issue, and we monitored Opus 4.8 until success rates returned to normal. The incident is fully resolved.
Between 17:38 UTC and 18:22 UTC on June 15, 2026, approximately 83% of requests to the analytics endpoint serving the /chronicle feature failed. The cause was an internal feature-flag service that encountered a transient error and failed to recover, causing feature flag checks to fail. The analytics endpoint was gated behind one of these flags, resulting in requests being rejected. We restored service health by removing the feature flag gating the analytics endpoint and deploying that change. To avoid recurrence of similar incidents, we have changed the feature-flag client so that errors that are not known to be permanent are retried, and we are improving alerting and startup behavior so this class of failure is detected and recovered from faster.
On June 15, 2026, between 15:27 UTC and 16:23 UTC, GitHub webhook deliveries were delayed. During this window, webhook events were delivered later than normal, with average end-to-end delivery latency peaking at approximately 8.8 minutes. No webhook deliveries were lost — delayed events were queued and delivered once processing recovered.This was caused by a temporary throughput constraint in an internal event-processing system that moves webhook events through GitHub's delivery pipeline. The rate at which events were processed for delivery dropped below the incoming volume, creating a backlog. We restarted the affected pipeline service, after which throughput recovered and the backlog fully drained by approximately 16:29 UTC. Webhook delivery latency returned to normal, the incident was mitigated at 16:39 UTC, and fully resolved at 17:37 UTC.To reduce the likelihood and impact of similar incidents, we are working on improving the accuracy of the utilization metrics used to scale our delivery worker pools, reviewing connection and capacity headroom in the delivery pipeline.
On June 11, 2026, between 19:28 UTC and 21:06 UTC, GitHub webhook deliveries were delayed. Average delivery latency peaked at approximately 3.4 minutes, with some deliveries delayed by as much as 62 minutes at the 99th percentile. No events were lost — delayed events were queued and delivered once processing caught up.This was due to a change in how webhook traffic was distributed across regions: to relieve load on one region, a portion of processing was shifted to another, where higher latency prevented our delivery workers from keeping pace with incoming volume, creating a backlog. We mitigated the incident by rebalancing webhook traffic distribution; as load returned to normal levels, processing caught up and the delivery backlog fully drained.We are working on improving the accuracy of the utilization metrics used to scale our delivery worker pools, and reassess how we distribute webhook traffic across regions, to reduce our time to detection and mitigation of issues like this one in the future.
Between 15:05 UTC and 16:25 UTC, GitHub API services experienced degraded availability due to sporadic authentication failures affecting approximately 9% of requests. Customers experienced intermittent "logged out" behavior as erroneous 401 responses triggered repeated authentication flows in app integrations. Affected requests also experienced approximately 800ms of additional latency. A memcached proxy service rollout to our internal API infrastructure caused our authentication service to pick up an incorrect memcached host configuration, leading to intermittent authentication lookup failures. We mitigated the incident by deploying a configuration change to memcached to use the correct host. To prevent similar issues in the future, we plan to migrate our authentication system to the new memcached infrastructure to improve resilience and strengthen overall reliability posture.
On June 8, 2026, between 14:49 and 14:54 UTC, a subset of requests to GitHub.com, the REST API, GraphQL API, and Webhooks UI/API experienced elevated error rates due to a transient infrastructure capacity issue that self-resolved within approximately 5 minutes.
Users experienced HTTP 500 errors and timeouts when accessing GitHub.com, the REST API, GraphQL API, and Webhooks UI/API for approximately 5 minutes, with the REST API taking up to 12 minutes to fully recover.
On May 25, 2026, between 09:02 UTC and 09:11 UTC, Git push operations over HTTPS and SSH experienced elevated failures. During this window, an average of 31% and a peak of 43% of push requests failed.
The incident was caused by a recently enabled code path that issued an unexpectedly expensive database query against a primary database. The resulting load exhausted the database's connection pool, which caused the push failures above. The acute impact resolved automatically as in-flight work completed. We mitigated the incident by disabling the feature flag controlling the new code path. To prevent recurrence, we have updated the affected background workflows to route reads to replica databases instead of the primary, removing the specific code pattern that caused this incident; broader follow-up work is underway to apply the same safeguard to similar workflows across GitHub.
On June 8, 2026, between 08:40 UTC and 09:30 UTC, the Claude Opus 4.7 model experienced degraded availability with error rates peaking at 8.4% and averaging 1.9%. This was due to an upstream provider issue that caused temporary unavailability and rate limiting on secondary failover systems. Users selecting Auto or alternative models were unaffected. We are improving provider failover mechanisms and monitoring to prevent similar issues.
On June 8, 2026, between approximately 06:30 UTC and 08:36 UTC, signed-out users experienced sustained elevated HTTP 504 errors when accessing Pull Requests, Issues, releases, patch diffs, and other related GitHub.com pages. During the incident, approximately 17% of unauthenticated requests to the affected GitHub.com endpoints returned gateway timeout errors, peaking at roughly 34% of requests at around 06:50 UTC. Some GitHub Actions workflows were also affected when they depended on release downloads or related GitHub.com endpoints. The impact lasted approximately two hours and was isolated to unauthenticated traffic; signed-in users were not affected. The issue was caused by a significant increase in abusive traffic to specific GitHub.com endpoints. This degraded our ability to respond to unauthenticated requests, causing requests to queue beyond timeout thresholds and return gateway timeout errors. We mitigated the incident by identifying the anomalous traffic pattern and applying targeted blocks at the load balancer and application layers. Error rates returned to normal and affected services were fully restored by 08:36 UTC. To reduce the likelihood and impact of similar incidents in the future, we are improving automated detection and blocking for these traffic patterns, improving our emergency traffic-blocking deployment path, and evaluating routing changes for endpoints used by both signed-out users and automated workflows.