OpenAIWhat happened on Wednesday and the history of the company one of the longest blackouts The major disruption was caused by a new construction is due to the telemetry service he explained. This outage caused major disruptions to ChatGPT, Sora, and developer API services.

In its latest status report, the company stated that the outage was not caused by a security incident or a new product launch, but by a telemetry service collecting Kubernetes metrics that was activated on Wednesday.

OpenAI apologized to everyone

Kubernetes is an open source program that helps manage application packages and related files in isolated environments. The telemetry service inadvertently caused resource-intensive Kubernetes API operations, which disabled the Kubernetes control plane. New telemetry service, many of OpenAI’s services DNS resolution It was said to affect the Kubernetes processes it relies on.

OpenAI DNS caching, allowed the spread to continue before its full extent was understood, delaying its visibility. OpenAI explained that they discovered the issue minutes before customers started experiencing it, but were unable to quickly resolve it due to overloaded Kubernetes servers. The company stated that this incident was the result of many systems and processes failing simultaneously and interacting in unexpected ways.

OpenAI announced that it will take several measures to prevent similar incidents in the future, including better monitoring of infrastructure changes, improvements to phased deployment, and new mechanisms to ensure its engineers can access Kubernetes under all circumstances API servers. OpenAI for this glitch ChatGPT apologized to its users and admitted that they did not meet his expectations.