Microsoft's Azure OpenAI Service in the Sweden Central region experienced a full-day outage caused by cascading backend failures and out-of-memory errors in Kubernetes pods. The incident began at 0900 UTC and wasn't resolved until 1612 UTC, affecting GPT-5.2, GPT-5 Mini, and GPT-4.1 models. Microsoft attempted multiple

2m read timeFrom go.theregister.com
Post cover image

Sort: