Day 34/40 - Step-By-Step Guide To Upgrade a Multi Node Kubernetes Cluster With Kubeadm

[@narsinghsaini3626]

Very Easy to Understand. Thanks for the amazing content.

[@TechTutorialswithPiyush]

Thank you

[@codinggeeg]

amazing content! following it from day 1 and it's really amazing

[@TechTutorialswithPiyush]

Appreciate your kind words!

[@harshgupta5113]

Thanks for the video

[@abc-edm]

your video makes it more enjoyable on learning K8s, thanks Piyush!!!!

[@TechTutorialswithPiyush]

Awesome to hear! Glad you found it helpful! 🚀

[@vikashswain7555]

on youtube I have watch many videos related to k8s but none of them explained related to upgrades. thanks for this amazing content.

[@TechTutorialswithPiyush]

Thank you for the feedback :) I am glad, I was able to help

[@SinghBalraj102]

Great demonstration! How to upgrade the cluster. I had seen a couple of videos, but no one explained in detail like you. You are amazing and explain each task in detail, which is easy for a beginner like me to understand. God bless 🙌 keep up the great work.👍

[@TechTutorialswithPiyush]

You're welcome, appreciate your kind words!

[@nareshv7797]

Thanks for dedicating your valuable time....very informative and helpful

[@TechTutorialswithPiyush]

You're welcome 🙏😊

[@GrigTech]

Thank You

[@suhasdhoke1735]

Thank you 😌

[@TechTutorialswithPiyush]

Welcome buddy 😊

[@giteshghure212]

Greate Content

[@TechTutorialswithPiyush]

Thanks a ton

[@movieshub4530]

Great content sir. ❤

[@TechTutorialswithPiyush]

Thank you 🙏

[@jhonsen9842]

No one Explained this Drain and Cordon conecpt . Exellent.

[@TechTutorialswithPiyush]

My pleasure always 😊

[@pankajtoconnect]

Thanks for your great efforts, could you please make a video on eks upgrade using terraform or eksctl without downtime

[@TechTutorialswithPiyush]

Yes, will do for sure

[@vikashswain7555]

great explanations!. do we have to use same procedure for EKS upgradations .

[@TechTutorialswithPiyush]

EKS is a managed Kubernetes service so many of these steps are already automated with that, its relatively simple and straight forward

[@pankajtoconnect]

Thanks for making this video. Sir could you make a video on eks upgrade using terraform or eksctl . Today I given a interview they have asked this question,also in this video you have discussed to upgrade automation could you please suggest, how we can do that, because if I have 100 worker node then each will take more time.

[@baivabmukhopadhyay8970]

Please also make a video doing installation of kubeadm and updating it in azure vms

[@TechTutorialswithPiyush]

Will do in the future

[@IqbalShaikh-k8j]

Need video regrading Upgrade Kubernetes Cluster with Kubeadm from 131.1 to 131.2

[@TechTutorialswithPiyush]

It's the same process brother, just change the version details as described in the video

[@hamzaerrahma9858]

Thank you for this video. For example, when working with a cluster where we don’t know if there are static pods (like MySQL in your case), and we want to drain a node, won’t we lose the static pods?

[@gsp237]

oh sure, thanks

[@AlitafSd]

Please do video on EKS cluster upgrade. And OS image patching

[@TechTutorialswithPiyush]

This series is not Cloud dependent so I'd probably do that once we are done with this series

[@jayasuriyap8748]

Please post an video on cert manager and let's encrypt in kubernetes cluster for custom domains

[@TechTutorialswithPiyush]

Will cover those topics once I'm done with this series

[@ajitsahoo4875]

Hi Piyush, another question for this point. Lets we so have couple of pod running a Node and we are using Nodeaffinity for scheduling the pod. So that those Pods can run on specific node only. If we go for a upgrade of a worker node and those pods are drained out of the nodes then there will be be massive outage for the application that running on those pods. Right? How to come up this issue?

[@TechTutorialswithPiyush]

Good questions Ajit! Yes, that could be an issue but for that we use a combination of Node Affinity, Taints and Tolerationa nd Node Anti-affinity. Anti-affinity is used to spread your workloads on multiple nodes so that the same workload is not scheduled on the single node, it spreads evenly or as per the settings

[@Kk-rl7nv]

Please please provide video on eks upgrade

[@TechTutorialswithPiyush]

Maybe after this series

[@youtubepremimum-x8w]

Please do video on EKS cluster upgrade. In interviews so many people are asking about this please do a video on it Thanks 😊

[@TechTutorialswithPiyush]

Upgrades to any managed cloud service is pretty simple and straight forward, You just have to perform a few checks and then upgrade the control plane using eksctl once that is done you run the upgrade command on worker nodepool, that's is

[@srikanthparisineni5562]

Please do one video on KOPS to manage clusters in kubernetes

[@TirumalaiahG]

@@TechTutorialswithPiyush Could you please share the steps that we have to perform before upgrading a managed cluster like EKS?

[@srikanthparisineni5562]

Please do video on KOPS

[@TechTutorialswithPiyush]

Will do for sure, once I am done with this series

[@srikanthparisineni5562]

@@TechTutorialswithPiyush Thanks

[@ajitsahoo4875]

What will happen if you upgrade the master node? Means we are running multinode cluster not a HA clustering In that scenario our application will be down for certain time?

[@TechTutorialswithPiyush]

Let's understand when your upgade in progress, your control plane will be down for a brief moment - APi server would be down meaning, you cannot issue a command on your cluster and other components wont be able to communicate with APiserver but your workloads will have no impact - ETCD will be down meaning, your etcd db wouldnt have the latest cluster state but your existing workload still be running - controller manager would be down meaning, if a pod fails/crashes or need of a scaling , that wont be possible but your existing pods keep on running if they are healthy - kube schedular will be down meaning news pods will not be scheduled, existing pods will be running

[@ajitsahoo4875]

Thanks Piyush for explanation That Clears my Doubts

[@guntreddirajesh]

🤩

[@josephrajareddy4606]

So the benefit of using ECS fargate is we don’t need team of ppl to manage upgrades?

[@TechTutorialswithPiyush]

That's one benefit yes

[@dr.hemantchauhan2613]

❤

[@jaskaransoni4624]

Hi piyush can you please able to provide me training about DeVops

[@TechTutorialswithPiyush]

Sorry bro, I replied earlier as well, I don't provide one on one or paid training

[@suhasiniaddanki7304]

If a node is assigned as daemonset, cant do we Cardon that node?

[@TechTutorialswithPiyush]

Yes, we can cordon the node by passing --ignore-daemonsets as the argument to the cordon command.

[@suhasiniaddanki7304]

Thank you @@TechTutorialswithPiyush