Forum archive › Kubernetes & Helm Charts

Error from server (ServiceUnavailable): the server is currently unable to handle the request (get pods.metrics.k8s.io)

Read-only archive. These discussions are from Tracston's earlier technical forum and are kept online because the answers are still useful. New replies are closed. Need help with something like this? Talk to us.

Question 8 Dec 2021

Problem

User is trying to run the command

kubectl top pod --all-namespaces

or

kubectl top node --all-namespaces

and get the following error

Error from server (ServiceUnavailable): the server is currently unable to handle the request (get pods.metrics.k8s.io)

Solution

first make sure metrics-server is installed

kubectl get po -n kube-system |grep metrics

if not add metrics server

kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml

if the command " kubectl top node --all-namespaces " still returning error then add configuration to policy

kubectl edit deployment.apps -n kube-system metrics-server

search for line with dnsPolicy: ClusterFirst

and add below the line

hostNetwork: true

Need this fixed properly?

We do this for a living.

Kubernetes, DevOps and platform engineering, delivered in days — not another weekend lost to a stack trace.

Start a project →