Forum

Same issue here.

Failed to install K3s: failed to configure certificates: rpc error: code = Unavailable desc = K3s API not serving expected certificate

Hello,

Upon review, your Wonder Mesh device (deb13-vm) currently has an abnormal client credential status, which is causing K3s to fail during the certificate configuration phase.

Please log in to the device via SSH and try the following operations to rotate the credentials:

# Stop k3s
sudo systemctl stop k3s

# Rotate certificates
sudo k3s certificate rotate

# Restart k3s
sudo systemctl start k3s

# Wait about 30 seconds and check node status
sleep 30 && sudo k3s kubectl get nodes

If the above commands return an error or have no effect, you can try clearing the old TLS directory to let K3s regenerate it:

sudo systemctl stop k3s
sudo rm -rf /var/lib/rancher/k3s/server/tls
sudo systemctl start k3s
sleep 60 && sudo k3s kubectl get nodes

Once completed, please click "Reinstall Zeabur Service" on the server page of the Zeabur dashboard to trigger a re-registration. The certificate configuration should pass successfully this time.

If the issue persists, please reply and let us know, and we will assist you from the backend.

I don't see "Reinstall Zeabur Service", only the "Install K3s immediately" button. Clicking that button doesn't solve the issue, and it shows the error: Failed to install K3s: failed to configure certificates: rpc error: code = Unavailable desc = K3s API not serving expected certificate.

Here is my operation log:

liulang-deb-test:~/projects$ # Stop k3s
sudo systemctl stop k3s

# Rotate certificates
sudo k3s certificate rotate

# Restart k3s
sudo systemctl start k3s

# Wait about 30 seconds and check node status
sleep 30 && sudo k3s kubectl get nodes
[sudo] liulang's password:
INFO[0000] Server detected, rotating agent and server certificates 
INFO[0000] Rotating dynamic listener certificate 
INFO[0000] Rotating certificates for admin 
INFO[0000] Rotating certificates for cloud-controller 
INFO[0000] Rotating certificates for controller-manager 
INFO[0000] Rotating certificates for etcd 
INFO[0000] Rotating certificates for api-server 
INFO[0000] Rotating certificates for auth-proxy 
INFO[0000] Rotating certificates for scheduler 
INFO[0000] Rotating certificates for supervisor 
INFO[0000] Rotating certificates for kube-proxy 
INFO[0000] Rotating certificates for kubelet 
INFO[0000] Rotating certificates for k3s-controller 
INFO[0000] Successfully backed up certificates to /var/lib/rancher/k3s/server/tls-1778737553, please restart k3s server or agent to rotate certificates 
NAME STATUS ROLES AGE VERSION
100.64.1.247 Ready control-plane 20h v1.35.4+k3s1
liulang-deb-test:~/projects$ sudo systemctl stop k3s
sudo rm -rf /var/lib/rancher/k3s/server/tls
sudo systemctl start k3s
sleep 60 && sudo k3s kubectl get nodes
NAME STATUS ROLES AGE VERSION
100.64.1.247 Ready control-plane 20h v1.35.4+k3s1
liulang-deb-test:~/projects$ 

There is a 'Reinstall Zeabur Service' option on the settings page, but clicking it results in an error:

Error: Invalid input Code: INVALID_ARGUMENT Description: kubernetes API unreachable: get server version: Get "https://100.64.1.247:6443/api/v1/pods": tls: failed to verify certificate: x509: certificate signed by unknown authority Trace ID: bc7a1459680f1a0f4cf68a877a927f19

I just received an email saying it was fixed, but after testing, installing k3s still results in the same error.

Hello, after further investigation, this is a known issue with our system on the Wonder Mesh server. The certificate verification step is not correctly passing through the internal network connection, causing the installation process to fail.

This is not an issue with your server or operations; it is a bug on our end, and the team is currently working on a fix. We will notify you once it is resolved, at which point you can reinstall.

We sincerely apologize for the inconvenience.

Received