Is Kubernetes down right now?

No. Our checks are finding no problems with Kubernetes right now. That verdict is live, and the banner at the top of this page carries it too. We re-check Kubernetes continuously using its official status feed, direct network probes and reports from visitors, stopping at the first check that gives a clear answer, and the page updates itself as soon as anything changes. A green verdict does not rule out a fault affecting only some users or regions.

The problems visitors report most often on this page are website / app down, API errors, can't connect, no response / timing out, 5xx error pages, lag or disconnects and files won't sync. Those are the options the report button offers for Kubernetes, chosen for how this service actually fails rather than a generic list, and the "user reports" chart above breaks down which ones people are picking right now.

Our checks run from one place on the internet, so a fault limited to your region or network can be real while our probes succeed. With Kubernetes specifically, Kubernetes is software you or your provider run rather than a hosted service, so what breaks is a specific cluster, not the project. kubernetes.io being reachable says nothing about your cluster, and the control plane, the nodes and your image registry all fail separately. That is why this page reports what it observed rather than declaring the service healthy, and why reports from other visitors count toward the verdict.

For Kubernetes, run kubectl cluster-info and then kubectl get nodes. The API server answering while pods fail points at your workloads or nodes. The API server refusing entirely points at the control plane, your kubeconfig or your network path to it. A problem that follows you across networks and devices is very unlikely to be yours.

If this page shows evidence of problems, the fault is most likely on Kubernetes's side. Specifically: avoid deleting stuck pods in bulk to force a reschedule. If nodes are unhealthy or the registry is unreachable, the replacements land in the same state and you have thrown away the running ones you still had. Reporting it here helps other people see they are not alone.

Continuously, around the clock. Kubernetes does not publish a machine-readable status feed, so we cannot read a verdict from them. We probe kubernetes.io directly over HTTP, fall back to an ICMP ping if that is inconclusive, and weigh reports from visitors. That means we can tell you the service is unreachable, but not what Kubernetes says about it. The verification tier box on this page names whichever check produced the verdict you are reading, so you can see how direct the evidence is.

No. status.buzz is an independent monitoring site operated by LuzLab LLC, with no connection to Kubernetes or kubernetes.io. It reads publicly available endpoints and status feeds, and publishes what it observed. For account or billing problems you need Kubernetes's own support, not us.