Reputation: 819
It looks like it happens randomly, but when I bring up Minikube and start working with it, after a while (can't say exactly how long, it changes every time) I suddenly can't use commands with kubectl and get an error -
Unable to connect to the server: dial tcp 192.168.1.20:8443: connect: no route to host
I tried to find a solution on the Internet but couldn't find the reason why this was happening, and I can't fix it other than deleting a minikube with
sudo minikube delete
And then restarted with
sudo minikube start --vm-driver=none
Anyone come across this and can help?
Upvotes: 1
Views: 9293
Reputation: 11055
I had similar behaviour when my minikube was overloaded. I was spinning up a minikube with default settings, but then deploying too much workloads on it. At some point I get the same error.
Try tracking the load on your node with:
kubectl top nodes
or
kubectl top pods --all-namespaces
I found that running minikube with more resources helped me.
My new default setup is:
minikube start --cpus 4 --memory 8192
Upvotes: 4
Reputation: 647
Another possible cause of this issue is that the local proxy settings may have changed. I had a local proxy server temporarily set but my terminal session had picked up the server configuration change and set the http_proxy and https_proxy environment variables. If you are having connection issues to the minikube cluster, remember to also check your local system network configuration and check there isn't something that is causing your traffic to be re-directed elsewhere.
Upvotes: 2
Reputation: 1577
Delete your minikube:
minikube delete
and try to augment ram and cpu resources for minikube in this way:
minikube config set memory 8192
minikube config set cpus 4
minikube start
minikube should start with ne new settings
Upvotes: 4