[ERROR FileAvailable--etc-kubernetes-manifests-etcd.yaml]: /etc/kubernetes/manifests/etcd.yaml already exists

[ShouZhiDuan]

Here is the translation of your message:

"I am using kubeadm v1.31 to deploy a Kubernetes v1.31 cluster, and I followed the official documentation to deploy an external etcd with 3 nodes. I ran the command:
kubeadm init --config kubeadm-config.yaml --upload-certs
and it was successful. Then I received the following join command:
kubeadm join 192.168.17.100:6443 --token abcdef.0123456789abcdef --discovery-token-ca-cert-hash sha256:c555953eec3c2c2eecee257ac7d874d3f0adde296b29a1c701f5c56d91c030a8 --control-plane --certificate-key 9d95da80b72cd07412352cc16cce5ebc4482da035a85dacc59a231b7261e5ec7

However, when I tried to execute the join command on another control plane node, I encountered the following error:
[preflight] Running pre-flight checks before initializing the new control plane instance
error execution phase preflight: [preflight] Some fatal errors occurred.
[ERROR FileAvailable--etc-kubernetes-manifests-etcd.yaml]: /etc/kubernetes/manifests/etcd.yaml already exists

This error occurred because I deployed one of the external etcd nodes on the same node as a control plane node, which caused a conflict due to limited virtual machine resources. I am wondering if, when joining other control plane nodes, I need to specify the use of the external etcd and prevent the local etcd from being started. However, I couldn't find any instructions for this scenario in the official documentation."

Tasks

Preview Give feedback

No tasks being tracked yet.

[neolit123]

This error occurred because I deployed one of the external etcd nodes on the same node as a control plane node, which caused a conflict due to limited virtual machine resources. I am wondering if, when joining other control plane nodes, I need to specify the use of the external etcd and prevent the local etcd from being started. However, I couldn't find any instructions for this scenario in the official documentation."

it is not an official supported scenario and that's why it's missing in the documentation.
note that the separate etcd external members need to be on separate machines.
you should not have an existing external etcd member machine also be used for a control plane component machine.

best to try getting help on the support channels. try the links from the bot below.

/support

[github-actions]

Hello, @ShouZhiDuan 🤖 👋

You seem to have troubles using Kubernetes and kubeadm.
Note that our issue trackers should not be used for providing support to users.
There are special channels for that purpose.

Please see:

• https://github.com/kubernetes/kubeadm#support

[github-actions]

Hello, @ShouZhiDuan 🤖 👋

You seem to have troubles using Kubernetes and kubeadm.
Note that our issue trackers should not be used for providing support to users.
There are special channels for that purpose.

Please see:

• https://github.com/kubernetes/kubeadm#support