Lab 3.1 step 5 , 6 and 8

BacemSOUISSI

Hello,
I have problems in processing the scripts in the lab 3.1 (installing kubernetes cluster).
In the step 5, I don't understand what I have to do : I created the file with the vm command but I don't what I have to do with the line
deb http://apt.kubernetes.io/ kubernetes-xenial main

I tried to put this line in the content of the file kubernetes.list , is that good ?
is there any other command to execute

In the step 6, when I try to execute the command I have an error message saying that - is not a valid option so I tried to execute the commande without the - in the end and it succeed but I don't know what effect it has !

After that, in the step 8, when I execute the command
root@lfs458-node-1a0a: ̃# apt-get install -y \kubeadm=1.18.1-00 kubelet=1.18.1-00 kubectl=1.18.1-00
I have the message saying that I have to use snap install , is this what I have to do or it is a problem with my config ?

I installed the three components with the snap command but then when I execute y=the command
apt-mark hold kubelet kubeadm kubectl
I have an error message !

Thank you in advance for your help !

chrispokorni

Hi @BacemSOUISSI,

On step 5 you are required to add the line you mentioned in the kubernetes.list file, and then save the file.

On step 6 you are required to run the curl command in its entirety. Because it is a multi-line command, depending on your environment, it may not paste correctly in your terminal. I suggest copying the entire multi-line command curl -s ... apt-key add - from the PDF into a simple text editor, remove the backslashes \ and convert it into a single-line command, and only after that paste it into the terminal. The output should only say OK after the command executed successfully.

Upon successful completion of steps 5, 6, and 7, on step 8 you should be able to install the Kubernetes 1.18.1 components with apt-get install command and also set them on hold with apt-mark hold command.

Regards,
-Chris

BacemSOUISSI

Hi @chrispokorni
Thank you for your response ! all fine with those steps now I am on the step 14 to init kubeadm.

I have a question about step 10 : I need to un-comment the lines 5 and 6 in the calico.yaml ? because those lines are commented on my environment !

When I proceed with the steps till step 14 to inti kubeadm (without un commenting the lines) I have those errors

W0529 07:25:23.887294 642 configset.go:202] WARNING: kubeadm cannot validate component configs for API groups [kubelet.config.k8s.io kubeproxy.config.k8s.io]
[init] Using Kubernetes version: v1.18.1
[preflight] Running pre-flight checks
[WARNING Service-Docker]: docker service is not enabled, please run 'systemctl enable docker.service'
[WARNING IsDockerSystemdCheck]: detected "cgroupfs" as the Docker cgroup driver. The recommended driver is "systemd". Please follow the guide at https://kubernetes.io/docs/setup/cri/
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=...
To see the stack trace of this error execute with --v=5 or higher

chrispokorni

Hi,

There is no need to uncomment those 2 lines in the calico.yaml file. The values will be used by default.

The error you are seeing is typical when kubeadm init ... was run multiple times in sequence. If the master node does not initialize properly, and you do not see in the output a confirmation that it initialized and the kubeadm join ... commands, then you would need to run a kubeadm reset on the master node and then run a kubeadm init ... one more time.

Regards,
-Chris

BacemSOUISSI

Hi,

I run kubeadm reset then init but I still getting the same error !

I think it is a port problem

error execution phase preflight: [preflight] Some fatal errors occurred:
[ERROR Port-10250]: Port 10250 is in use

thank you for your help

chrispokorni

You can use a few tools to find out what process or service is listening on that port, and then stop the process or service. You can use netstat, lsof or another tool of your preference to find that process.

Regards,
-Chris

BacemSOUISSI

Thank you !

I think it was a problem related to the first install that I have done with the wrong config which caused a kubelet service to be started and reserved this port.
I tried many solutions from the net to free the port or kill the service but no one was successful so I had created the instance again and no I have succeed installing kubernetes cluster on the master and joining the worker (in lab 3.2)

I have a remark about the lab 3.2 : in section 6 we must edit hosts file to add DNS for the master but the ip given in the example is the worker's one.

chrispokorni

Hi,

The kubelet does reserve that port, and since it runs as a service, it gets started up immediately after a system reboot.

The latest lab book version I have correctly shows the master IP address in step 6 of 3.2, which is consistent with the master IP in step 12 of 3.1. Perhaps you have another version of the lab book. You can always download the latest version from the course.

Regards,
-Chris

BacemSOUISSI

Hi,

I have the latest lab book 2020-04-20 from resources section but I can see only yaml files in it.
Can you please tell me how to get the latest version of course content.

Best regards,
Bacem.

coop

The course content itself was updated and is what you are looking at in the LMS. If you are seeing something else please clear your browser cache. You cannot download the course as a file.

fcioanca

The Lab pdf can be downloaded from the Resources tab in the Menu. The latest yaml files are in the Solutions tar file, which can be downloaded following instructions available in Chapter 1 > Course Information > Course Resources.

Best regards,
Flavia
The Linux Foundation Training Team