A deep dive into Cloudflare’s September 12, 2025 dashboard and API outage

According to The Cloudflare Blog, there was an outage in Cloudflare’s Tenant Service API which led to a broad outage of many APIs and the Cloudflare Dashboard. The immediate trigger was a bug in the dashboard that caused repeated calls to the Tenant Service API, because a problematic object in a React useEffect dependency array was recreated on every state change, making the effect run repeatedly.

As the Tenant Service became overloaded, API request authorization failed and 5xx errors propagated, impacting the dashboard and other APIs. The response included installing a global rate limit on the Tenant Service and increasing pods to boost capacity, with a patch later deployed that inadvertently worsened the situation before being reverted. The post notes the outage followed a pattern known as a Thundering Herd, and that a hotfix and planned dashboard changes aim to spread out retries and reduce contention. Cloudflare emphasises learning from the incident and improving visibility and release processes to prevent a recurrence.