Lab 3.1. Install Kubernetes, 21. Initialize the cp

rafael.urrutia.silva

I trying to init the CP doing according to Lab 3.1, page 6 and 21. with:
kubeadm init --config=kubeadm-config.yaml --upload-certs | tee kubeadm-init.out

But I get error show this:

[init] Using Kubernetes version: v1.29.1
[preflight] Running pre-flight checks
error execution phase preflight: [preflight] Some fatal errors occurred:
        [ERROR Port-10250]: Port 10250 is in use
[preflight] If you know what you are doing, you can make a check non-fatal with `--ignore-preflight-errors=...` 

I also stopped the kubelet process which was using port 10250, but without better result.

rafael.urrutia.silva

Hi Chris,

I'm doing the LFS258 running localy on Ubuntu-22-04 VMs based on the image I got from https://cm.lf.training/VIRTUAL_MACHINE_IMAGES/Ubuntu-20-04.tar.xz. FW are disables by now.
In did I run multiple times kubeadm init --config=kubeadm-config.yaml --upload-certs | tee kubeadm-init.out but I also did the kubeadm reset and the pkill kubelet before every time.

I will restart the labb using the same baseimage (VM from your site) but this time I will proceed directly to Lab 3.1. Install Kubernetes and I will not do any update or install before that.
Let me come back with the result then.

Regards
Rafael

chrispokorni

Hi @rafael.urrutia.silva,

I am moving this discussion from the LFD259 forum to its intended forum - LFS258. Thank you for clarifying.

The virtual machines you are referencing were not intended for this course. Please do not use them.

Ubuntu 22.04 LTS has not been tested as a lab environment for this course. From what your peer learners have reported, there are unresolved dependencies that will eventually need to be addressed.

Please provision your VMs locally with the official Canonical provided Ubuntu 20.04 LTS image or cloud hosted with their respective cloud images. The recommended OS for the lab environment is still Ubuntu 20.04 LTS.

Please ensure your hypervisor, or cloud VPC respectively, allows all traffic to your VMs - that is all protocols to all port destinations from all sources. For local VMs I'd recommend a single bridged network interface.

Regards,
-Chris

chrispokorni

Hi @rafael.urrutia.silva,

Are you enrolled in LFS258 or LFD259?
There are major differences between the two courses, and there are inconsistencies between the forum you are posting (LFD259) and the lab content you are referencing (that seems to be LFS258).

However, going back to your error, this is typical for a system where the kubeadm init command (or the kubeadm join command) was executed several times in a row. If it failed the very first time, running it again and again will likely produce more errors and will not resolve to a functional cluster node. Unlike a typical installation command that can be executed several times in a row, upon failure, the init command makes permanent changes to the system that will prevent it from running again.

What type of infrastructure is hosting your cluster? Local hypervisor or cloud? What is the guest OS of the VMs? What firewalls are there in place to filter/block ingress traffic to your VMs?

Regards,
-Chris