How to create Kubernetes Cluster & Node Groups in AWS using Managed EKS?

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
hello guys welcome back to my channel in this video I'll show you how to deploy or create Amazon eks cluster in AWS so we will call it as eks as elastic kubernetes service so what is Amazon eks so eks is in a manager service that you can use to run kubernetes on AWS without need to install operate or maintain your own kubernetes control plans or nodes so it will automate everything so it is an open source system for automating the deployment scaling and your containerized applications so it will be calling as container orchestration tool and you can it will be up-to-date versions of the open source kubernetes software so you can use all of the existing plugins or tooling tools from the kubernetes community so the applications that are running on Amazon eks are fully compatible with applications running on any standard kubernetes environment and no matter when they are running in on-premises or data centers or public clouds or private clouds so the it will be easily migrate any standard kubernetes application to Amazon eks without any code modification so this is where uh the brief introduction about eks in the AWS so let's go to in the Amazon AWS console and we can deploy the case cluster and we will attach a worker node group and we can deploy two servers so for this we need to install AWS CLI in my console I have a AWS install I have my AWS version is already installed so these are the few steps to deploy the ecase cluster so this is the first step install AWS error so I have put this command so I have AWS version and next in you need to install Google from the kubernetes for Windows so this is the command you need to copy and run this command new windows just hit enter to install the cube CTL in your local machine so this is a install group CTL the second step and this is a command you need to check whether your constitial command whether it is working or not it will show the version as we 1.2 some point three so it is my version so it is already installed and next you need to set up your environment variables as your group CTL you need to understand where it is located so for that you need to go to your file explorer [Music] environment variables variables you need to I have already given these users menu so here this will be Cube CTL is installed so you can see here and make sure in users which is installed now 253bm exam so this is where you need to give the environment variables here so I add new and you can give it so as of my I have already the environment variable path so as the same as AWS CLI version also so click on cancel so I have already it is so before that you need to create access scan secret key for close to Creation user so for which cluster you want to create that is a user you want to get the access key and secret keys so I am using a third user so I'll go for security credentials so that we will be creating the eks cluster so let me see the credentials so let me delete this credentials so let me create access again second case so these are my access games I got this if you close the space your access screen Security will never see again so you need to download the CSV so in my case I don't want to use this access case whenever I want you can create it so these are the access screen security Keys you need to configure AWS CLI using this configure and I'm creating an exist one and the next step is to create a VPC and two public submits for your cluster and after that you need to create a keypad so let's get back to the console and go to PPC so if you just if you see here and there is no vpcs here let me go to create VPC and I'm doing automate uh the product AKA CBC 10.80 I'll go for two available distance and two public subnets no need of private subnets points and create VPC so if you see here it is automatically taking the internet gateway creation and it is attaching to the VPC and it is attaching to the routable loads also so we have VPC you can see here my VPC is created and you can see here amazing 1B you can see here one a and one b row tables so so it is it will be an one dot table only so it's a public Dot it is internet get attached and the substance to associated so this is my internet gateway which is in case VPC so let me delete this I don't want to use this which is a cost occurring so we have created our VPC and you can close this and you need to create a key pair also for connecting to your nodes after creation of cluster so go to GPS so I have four kbps uh I have the laptop keypad so don't worry about that so if you don't have you can create a keypad Pam or dpk whatever you want so I have my own keypad and go to the IM rows you need to create create two IM roles for eks and work node so go to IM so create a role so choose here it is it will be service type as eks here so you need to choose case cluster because this URL will be creating a eks cluster will be managed and you can see here it is cluster pulse is attached and create role it will be created after that you need to create a work or not role also so you need to attach these policies to the worker node policy so create worker node it's very easy to so through the policies first one is policy so it will connect your awcc2 from your eks clusters so it's easier if you want to pull images so this is container resist to read only and this is CNA policy so this is it will automatically install plugins from your CNA policy so click on next case um create all it's already exists so it is saying that it's already exists so let me go to the roles whether it is ready or not so it is already there let me check the UK's cholesterol also so I'm deleting the existing eks IM row which I have uh created earlier so you can see here EK Circle node any case cluster pulses has been created and the next one is to create the E case cluster in the console you go to the elastic Opening Service you can see here there's no clusters in my console so if you want to add a cluster or you can register your customer so I am creating a cluster parameters equals cluster North Virginia or else I'll give us case managed poster or else AWS it is this is fine and kubernetes is 1.7 default and the role is automatically selected so click on next which is your VPC I have only one VPC so it is public subnets and Security Group I don't have any security room so I need to create one Security Group let me go back easy to create a security group okay as close to hello all so I am going to allow a little traffic so this won't be in real time so for the testing purpose I have allowed all so it will be application create Security Group so it is created so refresh here and choose the second group and it will be if your ipv4 address family so no need to configure service iPad automatically so my endpoint access will be in public time you need to give these those options public or private or public so open your requirement and I don't want to enable any logs so control pin there won't be any server locks or other blocks and you need to defaultly enable this because add-ons so vpcc and I go process record DNS so this will be installed by default so this is where you can preview this cluster while creation and role in case cluster where it is delay creating and networking to secure one Security Group and two subnets on VPC and public and point access and no Logs with three add-ons and add-on versions and create a cluster so it will take some time to create the cluster I mean we'll check the group CTL version I have given here version short hyphen client it will draw another that uh your cluster is not created and your you have not connected to your cluster from Cube serial so this is the error Security Group API Group so it is actually refuses it so let me create after this cluster so let me check the status it is still in creating if you see here there is no worker node group or nodes or server profiles in my compute so you can add your forget profiles from other infrastructure to the cluster or if you want to add only one node here you can add it or you want to create nodes into a group and you can add a notebook so in this case we are creating a notebook here so it will automatically attach a auto scaling and it will deploy the auto scaling activity and it will manage the workload between those two servers you can see here networking so this is my VPC and the service ipu for range and it will submit Security Group it will be attached additional security groups this is my Security Group and there is public server endpoint access and you can see here three add-ons will be there Cube Roxy Amazon VCC and a coordinates will be automatically installed and creating which is installing status and there is no authentication here there is no logs I have enabled so if you have any updates of the cluster you can have it here and I don't have any tags here and you can see here resources there is no parts running in my cluster in the notebook so the part is using modest and simplest kubernetes object a power represents a set of running containers on your cluster simple pod is nothing but a collection of containers running in your nodes and registered in creating stage Let me refresh my console so this is a warning like manage notebook and forget profile cannot be added while the cluster AWC case is being created so I have not added anything it's like a warning message let me pause the video after the classroom camera so if you see here my case clusters came up and it is inactive so let's create a note group and attach to this cluster so it is installed with 1.27 kubernetes version and providers in case so this is my AWS case cluster name and you can see here over you and API server endpoint certificate Authority Secrets encryptions off and chemist guide I didn't given anything if you want to enable you can do it and it is created nine minutes ago and plus Ray run uses in in case three platform version and you need to go to compute and you need to go to resources whether your code DNS are came up or not so I have not added any work or not so it is not showing here so it never was attached not group it will show the idea resources like to work and also have been created and let me connect to my cluster using the cluster uks name AWS in case this is AWS command so I actually copy this command so in the place of cluster name you need to give AWS and the reason name is you can see here in the region section here is and copy this and using AWS CLI command you can write in case so enter so it is saying that update group config name uwc case using your system augmented operational choices let me check whether it is giving any sorry it is my bed I need to give the Extra Spaces here so this is error so copy this and hit so you can see here I have given the space here so it is not connected the AWS command so let's update hyphen config is a namespace uh sorry in each new context so it will add a new context in your Aaron so let me show you this and then username and you can see here kill people is installed and you can see here dot Cube and the config file so it will be appeared like this so to get Authority and all these clusters your details about this cluster here in the context so it is shown up here so this is even better version one so let me try Cube Ctrl kit and nodes so there is no resources one so let me create a notebook inside the E case cluster so go to compute and add notebook so my in case this in case I cannot group and select a node rule which you have created earlier so no need of levels 10 stacks and launch template next next so I'll go for Emit IPad Samsung Linux 2 this updated version and capacity type is on demand select instance video medium so it will be just taking as default selectively or else you can want to decrease it or go for Max has three nodes this is two so it will automatically install a two AWS launcher to ECT instances in your node group scaling activity the auto scaling configuration and whenever the node update configuration it will be unavailable should be one so if there is any update uh one port should be down and one one node will be up and it will be working fine the two subnets and configured here if you want to go inside the worker node and you need to set up something so you need to have the key pair so previously I have created a laptop keypad so yeah in this video I have shown you 30 groups and this is a hello all and you want to add this is also en edit is attached so if you go here ec2 and you can see here and network interfaces the two network interfaces are created in 1G and 1B and the cluster Security Group is attached and you can see here elastic network interface and this is it primary IP addresses and that's it select security groups and this is a default Security Group so no need to address is and click on next and create a worker node group so it will take some time so by default there is no instances here if you can refresh here you'll add your two instances launching up without a scaling group activity so if you see here notebook configuration kubernetes equation is 1.27 MM release version is 1.27.1 and my type is Amazon nx2 x86 60 43 medium instance type and status is creating is not yet configured or created it will post this it will create it I may not group Aaron and this is I am role which you have created notebook and capacity type is on demand these are two minimum do not should be available and Max on three nodes if there is any load it will be pop-up one and the two subnets and if you want to access the nodes remote access you need to have a laptop keypad so this is my laptop VPN and these are the security groups to attached and these are the nodes it will appear here no health issues no levels no update config Now kubernetes update no tags so let me refresh so it is saying that health issue one or more Amazon is represents notebook does not automation public IP if you want enable Okay so you need to add a VPC publicity for your two subnets so let me go to the UPC and choose the subnet for which I have enabled and created so these are my two subnets go to actions view details edit subnet settings and enable auto as an IP for a black IP address save and I have edited 55 f and choose the other one edit subnet settings enable and Save so try it now so it will automatically add it or not so we need to delete this and we need to create again so there won't be any launching initial instances here in the console so if you see here there is no instances running here so let me add for the time being I need to add a node group so I guess node group so that I am roll node Rule and launch template next create so if you see here there is no health issues in this notebook previous notebook it is already deleted so if you see here you can register to cancel so two instances are going to initialize and the status X are in initializing so you can see here in the notes section it will come up with the two nodes so it is still an initializing so go to the auto scaling groups here visit console so you can see here Auto scaling group is also created so you can see here just created now itself and this is this time and this is activity so it will launch it to easier instances the data is coming up today let me go to easy to install console instances page and refresh here so you can see here two instances 63 medium I just created a mini table so it will be two instances in your notebook and this is auto scaling work on book node name group name and these two subnets so this is where your worker node need to create and attach to your request cluster so you can see here the nodes will be available here this goes down it will create automatically the one instance using the auto scaling activity and go to this and what is a short form for nodes so if you see here my 10.80.11.66 is my server which is the one and the other one is 3.40 17.40 which is attached to the second opener Group which is in one day one b and the first one is in one level create in high availability zones so this is where QC will get nodes and you can describe mode copy and paste it will list all the information about the node so you can see here the whole information it has been shown up here and IP address that has created process CPU memory pods how many ports and it will create an animation ID systematic board and all but it can appears like uh there is any uh no CPU memory or limits it will show here the error message like this in the events section so with this you can easily find out how the node is not able to create or the Pod is not able to create in my notes so this is where you can create it easily internet 10 minute steps 10 steps for the creation cluster using this 9 to 10 steps using the cube CTL command using install and the windows so this is my case cluster creation video so hope you guys like the video yeah thanks for that's it guys thanks for watching my video please do like share and subscribe for more videos let me know if you have any questions please post in the comment box thank you
Info
Channel: Tarun Technophile
Views: 2,648
Rating: undefined out of 5
Keywords: #aws #cloud #tarun #technophile
Id: fnWNPLEbGZ8
Channel Id: undefined
Length: 32min 20sec (1940 seconds)
Published: Thu Jul 06 2023
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.