# GitHub Blames 8-Hour Outage on Autoscaling Fail and VS Code Retry Storm GitHub explains the causes behind the prolonged service disruption. - **Source:** [www.theregister.com - Articles](https://brifing.si/source/2c34ea50-d341-4a10-925d-12c012a9ad10) - **Published:** 2026-08-19 - **Original article:** - **On Brifing:** https://brifing.si/s/7c38b81a-0d06-4c7d-adb4-e3fe0d2602aa ## The Beginning of the Outage GitHub experienced an outage that started at 13:28 UTC on August 17 and lasted for 7 hours and 47 minutes. The issue began with the overload of load balancers in the company's Central US facility. This caused errors in core services like Issues, Pull Requests, APIs, Actions, and Copilot. The initial trigger was an Istio sidecar reaching its concurrency limit. ## Autoscaling Misconfiguration GitHub's autoscaling policy was misconfigured. The policy monitored the main service but did not check the sidecar's concurrency limit. This error led to a cascading failure. Engineers temporarily mitigated the issue by reducing gateway retry counts through a code change and configuring load balancers to reject Copilot Token Service requests. ## VS Code's Role GitHub explained that a bug in Visual Studio Code contributed to the prolonged outage. Delayed responses to a single internal endpoint triggered a latent retry bug in VS Code, amplifying traffic by approximately 10x. This significantly delayed the recovery of Copilot Token Service. Most services recovered by 16:36 UTC, but Copilot Token Service didn't come back online until 21:02 UTC. ## Complicating Factors GitHub cited additional factors that hindered recovery efforts, including numerous scraping attacks on codeload endpoints. The company announced plans to correct autoscaling policies, review retry limits, audit Istio concurrency settings, and fix VS Code's behavior that amplified Copilot token traffic. ## Competitive Solutions The outage prompted some developers to seek alternatives. CloudBees CEO Moritz Plassnig noted in a LinkedIn post that Cursor, OpenAI, and several other startups are developing competitive solutions. Plassnig stated, "GitHub won't be the default solution going forward, and we are looking at a much more bifurcated ecosystem (which is good and bad)." ## GitHub's Reliability Issues GitHub's reliability problems aren't limited to this week. The company has acknowledged similar issues in the past. Such outages affect a critical piece of infrastructure that many organizations depend on. Developers now have more alternatives, and GitHub's value proposition is becoming increasingly challenging. ## Cursor's Move During GitHub's outage, SpaceX-owned Cursor announced an early beta of Origin Code Hosting. This move signals increasing competition in the developer ecosystem. As Plassnig pointed out, GitHub may no longer be the default solution, and the ecosystem could become more fragmented. ## Implications GitHub's outage has raised significant concerns among engineers. A misconfiguration and retry storm rendered critical infrastructure unusable for hours, impacting many organizations. This incident could weaken GitHub's position against competitors and lead to significant shifts in the developer ecosystem. ## Sources covering this story - https://www.theregister.com/saas/2026/08/19/github-blames-8-hour-outage-on-autoscaling-fail-and-vs-code-retry-storm/5289547 — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._