Kubernetes tutorial: How to install a cluster on Ubuntu with kubeadm?

[@AjayAnand21]

Thanks for step by step instructions. I was able to setup K8s cluster without any issue

[@osamaabou-el-gibeen9704]

man that cgroup driver saved my week. thanks a million

[@LearnDevOpswithSerge]

Glad to hear it. You are welcome!

[@sarvamsahitya]

Thanks. Did try to install the dashboard. Please share if you have done it already

[@LearnDevOpswithSerge]

Hi, here it is: kzbin.info/www/bejne/iYbOkIGjaqechtk

[@ashanp1234]

Thank you very much for this video. Can you share the manifest file you created in 9:09 ?

[@LearnDevOpswithSerge]

Hi, here it is: apiVersion: apps/v1 kind: Deployment metadata: name: nginx-deployment labels: app: nginx spec: replicas: 3 selector: matchLabels: app: nginx template: metadata: labels: app: nginx spec: containers: - name: nginx image: nginx:1.14.2 ports: - containerPort: 80 --- apiVersion: v1 kind: Service metadata: name: nginx spec: type: ClusterIP ports: - targetPort: 80 port: 80 selector: app: nginx

[@ramonriserio]

man, Despite I've defined apiserver-advertise-address flag in kubeadm init command: kubeadm init apiserver-advertise-address=192.168.10.17 the IP of kube-apiserver-master and etcd-master is 10.0.2.15 (NAT interface) and I think this is causing CrashLoopBackOff when intalling Flannel. Do you know how I can fix it?

[@LearnDevOpswithSerge]

Hi, 1. You can check your Kubernetes API server IP address using the command: "kubectl -n kube-system get pod -l component=kube-apiserver -o yaml | grep advert" or find it in the file "/etc/kubernetes/manifests/kube-apiserver.yaml" on the master node 2. Recheck your internal IP addresses (5:55) of the nodes using the command "kubectl get node -o wide". They shouldn't be 10.0.2.15. 3. Recheck the interface in the flannel config (8:19)

[@joshuaedward6893]

Hi there, i cant seem to get my network working at all, I'm using the same manifest as you(edited the necessary things) but its not working, my kube-flannel is currently stuck at terminating phase for more than a day now.

[@LearnDevOpswithSerge]

Hi, try to check the logs (kubectl logs -p POD-NAME).

[@ManjunathKundargi]

Hi, what will be the command at 3:55 if i am using docker instead of containerd

[@LearnDevOpswithSerge]

Hi, Kubernetes has not supported Docker since version 1.24. Here is more information about it: kubernetes.io/blog/2022/02/17/dockershim-faq/

[@richardsrobin_r]

vagrant file for networking setting , which ip should be used for --apiserver-advertise-address =windows ip4 address ?

[@LearnDevOpswithSerge]

Hi, here is my config: Vagrant file: 3.times do |i| config.vm.define "node#{i+1}" do |node| node.vm.hostname = "node#{i+1}" node.vm.network "public_network", bridge: "Intel(R) Wireless-AC 9560 160MHz", ip: "192.168.1.5#{i+1}" end end --apiserver-advertise-address: root@node1:~# cat /etc/kubernetes/manifests/kube-apiserver.yaml | grep advertise kubeadm.kubernetes.io/kube-apiserver.advertise-address.endpoint: 192.168.1.51:6443 - --advertise-address=192.168.1.51

[@RaviKumar-ko2hn]

Hi I am trying to setup k8s Ubuntu 22 jammy using aws ec2 instance. But issue with pod network. In logs i got dockershim.sock no such file directly like. How do i fix this issue.

[@LearnDevOpswithSerge]

Hi, The best choice is not to use Dockershim. Dockershim has been removed as of Kubernetes v1.24. kubernetes.io/blog/2022/05/03/dockershim-historical-context/

[@hamzaerrahma9858]

Thank you for this video, I initialized my cluster and everything was working perfectly. However, after I rebooted my machines, the kubectl get nodes command stopped working. When I run sudo kubectl get nodes, I get the following error message:The connection to the server 192.168.56.101:6443 was refused - did you specify the right host or port?

[@LearnDevOpswithSerge]

Hi, have you disabled swap? stackoverflow.com/questions/56737867/the-connection-to-the-server-x-x-x-6443-was-refused-did-you-specify-the-right Try to execute these commands on the master node: 1. sudo -i 2. swapoff -a

[@hamzaerrahma9858]

@@LearnDevOpswithSerge Thanks You Very Much

[@PratapKumar-sm9if]

Hi Serge, I have followed your tutorial, However, Kube Proxy status shows a Crashloop error Any suggestion Please ? Thanks

[@LearnDevOpswithSerge]

Hi, I guess you mean Flannel when talking about kube proxy. 1. Check logs of the crashed container: "kubectl -n logs -p " 2. Check events of the pod in the 'CrashLoopBackOff' state: "kubectl -n describe pod " 3. Analyze the information from the 1st and 2d list items.

[@magicalvibez2037]

how did u created the master and worker nodes?

[@LearnDevOpswithSerge]

Hi, I run virtual machines using VirtualBox and Vagrant.

[@satishbarnana2267]

how can we configure 3 vms in single machine and can we run master and worker nodes on same machine?

[@LearnDevOpswithSerge]

Hi, I use VirtualBox with Vagrant to create VMs. In Linux, you can use KVM or VirtualBox too. Yes, you can use master for your workloads. To do it, remove taints on the master/control plane node.

[@badex3301]

Thanks for the informative vid. Would be nice to have a google doc stating the command at 3:55

[@LearnDevOpswithSerge]

Hi, containerd config default | sed 's/SystemdCgroup = false/SystemdCgroup = true/' | sed 's/sandbox_image = "registry.k8s.io\/pause:3.6"/sandbox_image = "registry.k8s.io\/pause:3.9"/' | sudo tee /etc/containerd/config.toml

[@badex3301]

@@LearnDevOpswithSerge thank you! Subscribed

[@shang-mokai5428]

Thanks You Very Much !!! working well for me

[@timetraveller2045]

Many Thanks! Very usefull video!

[@IndianSumaira]

did it worked for you ? I am facing some trouble when i try to init cluster, says its: st:10248/healthz": dial tcp 127.0.0.1:10248: connect: connection refused. Unfortunately, an error has occurred: timed out waiting for the condition This error is likely caused by: - The kubelet is not running - The kubelet is unhealthy due to a misconfiguration of the node in some way (required cgroups disabled) If you are on a systemd-powered system, you can try to troubleshoot the error with the following commands: - 'systemctl status kubelet' - 'journalctl -xeu kubelet' Additionally, a control plane component may have crashed or exited when started by the container runtime. To troubleshoot, list all containers using your preferred container runtimes CLI. Here is one example how you may list all running Kubernetes containers by using crictl: - 'crictl --runtime-endpoint unix:///var/run/containerd/containerd.sock ps -a | grep kube | grep -v pause' Once you have found the failing container, you can inspect its logs with: - 'crictl --runtime-endpoint unix:///var/run/containerd/containerd.sock logs CONTAINERID' error execution phase wait-control-plane: couldn't initialize a Kubernetes cluster To see the stack trace of this error execute with --v=5 or higher

[@suryagowda3663]

Hi , i want to learn step by step about kubernetes with syllabus. Can u share here...

[@LearnDevOpswithSerge]

Hi, I'm going to make a video about it.

[@mazahirahmed]

8.20 name of internal network interface is not same for master and slave for me, I am getting error.

[@LearnDevOpswithSerge]

Hi, 1. You can set labels on each node group with different interface names. For example, node1 - label1 node2 - label1 node3 - label2 node4 - label2 2. Make 2 DaemonSets with different args for containers and set different nodeSelectors for each DaemonSet. 3. I suggest you watch my video: kzbin.info/www/bejne/d2rTmaxunteFjrs

[@suryagowda3663]

What is kubeadm? What is the use? When we should use?

[@LearnDevOpswithSerge]

Hi, kubernetes.io/docs/reference/setup-tools/kubeadm/

[@tiagobadalo]

love you Serge

[@knvssandeepbolisetti5008]

Thank You Very Much.

[@LearnDevOpswithSerge]

You are welcome!

[@youtea7324]

at 03:47 you prepared a command, from where do i get it.

[@LearnDevOpswithSerge]

Hi, containerd config default | sed 's/SystemdCgroup = false/SystemdCgroup = true/' | sed 's/sandbox_image = "registry.k8s.io\/pause:3.6"/sandbox_image = "registry.k8s.io\/pause:3.9"/' | sudo tee /etc/containerd/config.toml

[@christopherobinna2140]

@@LearnDevOpswithSerge how did you generate this command. any documentation for it. up till this stage you have explained everything in detail but just copied and pasted this out of nowhere. the tutorial is very great!

[@LearnDevOpswithSerge]

​ @christopherobinna2140 Hi, 1. SystemdCgroup = true # The documentaion about it kubernetes.io/docs/setup/production-environment/container-runtimes/#containerd-systemd 2. The issue with sandbox image: github.com/cri-o/cri-o/issues/6985 discuss.kubernetes.io/t/can-not-install-kubernetes-cluster-with-kubeadm/24079 ---- Note that the version of the sandbox image in each Containerd version can be changed so check it. For example, in my video, the version is 3.6 and now it is 3.8. The message while installing Kubernetes: W0725 06:32:35.829024 14158 checks.go:844] detected that the sandbox image "registry.k8s.io/pause:3.8" of the container runtime is inconsistent with that used by kubeadm.It is recommended to use "registry.k8s.io/pause:3.9" as the CRI sandbox image.

[@sumanth6114]

dude at 3:49 its big command where can i get that?

[@LearnDevOpswithSerge]

Hi, containerd config default | sed 's/SystemdCgroup = false/SystemdCgroup = true/' | sed 's/sandbox_image = "registry.k8s.io\/pause:3.6"/sandbox_image = "registry.k8s.io\/pause:3.9"/' | sudo tee /etc/containerd/config.toml

[@laurentsun4707]

Hello, for the command at 5:35 how to you find the --apiserver-advertise-address

[@LearnDevOpswithSerge]

Hi, it is an IP of your master node. You can find out it by the command: "ip address show" or shortly "ip a". ``` vagrant@master:~$ ip address show | grep 192.168.1.50 inet 192.168.1.50/24 brd 192.168.1.255 scope global enp0s8 ```

[@laurentsun4707]

@@LearnDevOpswithSerge Thank you! Do you have a video to show how to setup the kubernetes dashboard?

[@LearnDevOpswithSerge]

I haven't done it yet. Thanks for the idea!

[@ElliotAlderson-s4l]

@@LearnDevOpswithSerge also, the --pod-network-cidr, where did you find that?

[@LearnDevOpswithSerge]

@@ElliotAlderson-s4l Hi, 2d item kubernetes.io/docs/setup/production-environment/tools/kubeadm/create-cluster-kubeadm/#initializing-your-control-plane-node

[@bololo4988]

why you not have error invalid argumen with command sudo sysctl --system

[@LearnDevOpswithSerge]

Hi, this command is correct: $ sysctl --help | grep system --system read values from all system directories

[@ramankhanna9526]

thankyou

[@LearnDevOpswithSerge]

You are welcome!

[@sathvikbandaru3985]

Thanks you

[@sathvikbandaru3985]

for me Culster set up is ready because of you but little issue my pods are not running properly when im using flannel but when i using calico now my pods are running

[@sandeepthapa9478]

containerd config default | sudo tee /etc/containerd/config.toml >/dev/null 2>&1 $ sudo sed -i 's/SystemdCgroup \= false/SystemdCgroup \= true/g' /etc/containerd/config.toml from where did u get this command in k8s documentation

[@LearnDevOpswithSerge]

Hi, 1. SystemdCgroup = true # The documentaion about it kubernetes.io/docs/setup/production-environment/container-runtimes/#containerd-systemd 2. The issue with sandbox image: github.com/cri-o/cri-o/issues/6985 discuss.kubernetes.io/t/can-not-install-kubernetes-cluster-with-kubeadm/24079 ---- Note that the version of the sandbox image in each Containerd version can be changed so check it. For example, in my video, the version is 3.6 and now it is 3.8. The message while installing Kubernetes: W0725 06:32:35.829024 14158 checks.go:844] detected that the sandbox image "registry.k8s.io/pause:3.8" of the container runtime is inconsistent with that used by kubeadm.It is recommended to use "registry.k8s.io/pause:3.9" as the CRI sandbox image.