Hello everyone, welcome back to my

YouTube channel. So in this video we are

going to discuss about Argo CD. Argo CD

is a GitHubsbased

continuous delivery tool and this tool

is specifically designed to manage your

Kubernetes cluster. What does it mean?

So basically you can deploy your

Kubernetes resources on Kubernetes

cluster using Argo CD but you cannot use

Argo CD to manage deployment on your

Azure web app, Azure container app.

Similarly for AWS and GCP so it is

specifically designed to make your life

easy while you are deploying your

resources or managing your resources on

Kubernetes cluster. So let's understand

in more detail.

So basically Argo CD follow Githops

principle and as we know Githops

basically follows DevOps best practices

and what's the principle of GitHubs? So

in GitHubs basically Git repository is a

single source of truth for your

application. What does it mean

basically? So whatever the configuration

or whatever the state you have in your

GitHub repositories same state same

configuration will be available or will

be deployed on your Kubernetes clusters.

So it basically continuously monitor uh

your kubernetes cluster state and same

time it also monitor your g repositories

and it's always sync whatever the

configuration or whatever the code you

have in your github repos same code will

be available or deployed on your

kubernetes clusters. So it basically

shrink the changes. Suppose you have

made some changes on your GitHub

repositories. Maybe you have changed

your docker image version like from

v_sub_1 to v_sub_2. As soon as you

commit your changes, Argo CD

continuously monitor your GitHub

repositories. It will see okay there are

some changes it will automatically

shrink that changes whatever you have

made on your GitHub repositories and it

will be deployed on your Kubernetes

clusters. So we can also say Argo CD is

a auto deployment tool because you are

not going to change or update anything

manually. You are just going to update

your changes on GitHub reposities and

your job is done. Argo CD will take care

and same thing if you are going to make

any changes on Aros or sorry on your

Kubernetes cluster directly then it will

automatically fix your GitHub code as

well. So it is vice versa like either

you are going to make changes on your

GitHub it will automatically shrink or

automatically deploy your changes on

coortis clusters and same time if you

are going to change anything directly on

kubert clusters it is going to fix those

changes on your github as well.

So now what problem basically aroc solve

if suppose you are not using argo cd

then what all the challenges you are

going to face. So for the very first is

no single source of truth because maybe

some of your engineers going to make

changes directly on Kubernetes clusters

and you are not aware who did what

changes. So there are no proper tracking

systems. So if you are using Argo CD

which is a GitHub tool it always track

who made what changes on your GitHub

repositories and if you want to do a

roll back it is quite easy to do a roll

back as well that we will see in

upcoming uh demonstration maybe after

few minutes manual update cause errors.

So if you are manually updating your

manifest file directly on your cluster

maybe in that case you might make

mistake and there are no tracking what

changes you have applied. So there are

chances if you are going to make

anything uh manually that definitely

uh those changes can cause uh issue on

your cluster on your application.

Git and cluster go out of sync. Maybe

you have some uh the current state

whatever state you have in your

kubernetes state or sorry on kubernetes

cluster same state you don't have in

your github so there is a mismatch in uh

in your manifest file maybe so there are

no mechanism to auto shrink your changes

so that's problem is also going to solve

by argo cd rollbacks are hard if you are

going to make suppose if you are not

using aroc then definitely you are going

to do manually update or maybe you have

set up a CI/CD pipeline then the whole

procedure you have to follow but if you

have to make a roll back roll back in

case of any issues but if you are using

Argo CD you are simply go to your Argo

CD dashboard and you can simply revert

the changes to the previous commit which

was working.

So these are the some benefits if you

are going to use Argo CD and these are

the drawbacks if you are not going to

use Argo CD.

All those things we are going to see in

demonstration as well. Now let's talk

about architecture of Argo CD. So there

are three main components in Argo CD.

The very first is API server. We call it

as a brain of Argo CD because it handle

all the communication.

error and another component we have

repository servers which is basically it

the the job of repository server is to

read your GitHub repos then only it is

going to make changes on your Kubernetes

clusters. Another one we have

application controller which basically

continuously compare your configuration

whatever you have in git and whatever

you have in co clusters. So it will it

will continuously monitor and sorry it

will continuously compare the

configuration if it find any difference

it is going to update those changes like

so whatever the current state you have

in cluster same state you are going to

have in your GitHub repos so these are

the three main components which used by

the Argo CD

now let's talk about the features of

ArgoD beautiful UI dashboards so you

will get a clear view of your

applications in real time like what

what's the health in case if as if is

there any issues everything you are

going to manage using a single dashboard

selfhealing if someone changes the

cluster manually auto or sorry Argo CD

uh can basically bring it back like

definitely you are going to make or

update those changes in GitHub as soon

as you update and Argo CD continuously

monitor your GitHub repos

and they find any differentiation in

your GitHub repo and in your Kubernetes

cluster, it is going to automatically

update those changes on your Kubernetes

clusters. So the state will remain same

in both in your GitHub and your

Kubernetes clusters. Automatic

deployment we already discussed

rollbacks also we have discussed and now

it uh you can not only use ML manifest

if you have helum charts that also

supports by argo cd if you have

customize json net or whatever manifest

you have they are going to support all

these manifest types multicluster

supports so you can deploy your

application on multiple cluster as well

using aroc CD.

Now let's see how we can install Argo CD

on Kubernetes clusters. So the very

first requirement to set up a Argo CD is

you should have up and running

Kubernetes clusters. So in our case we

have a cluster which we have enabled in

docker desktop. So like for learning you

can or hands-on basically you can

definitely use docker desktop. You

should not set up a AKS cluster or EKS

cluster which definitely going to cost

you. So for learning or hands-on purpose

you can use docker desktop.

So

let me open my docker desktop.

So we have Kubernetes cluster. You can

see the cluster status is active and

it's running. And this is a blank

Kubernetes cluster. If I will go to

command prompt and let me quickly check

if we have any pod deployed. UCTL get

pods. So we don't have any parts or any

application up and running. Now

let's verify one more thing. ubectl get

pods - a.

So you can see here we don't have any

pod up and running for argo cd. So these

are the default pod which basically used

by a kubern cluster when you enabled it.

So now let's set up a argod first. So

there are few steps which you have to

follow. So the very first thing you have

to set up a separate name space for

argod. So let me uh copy it.

So it is going to create a name space

called Argo CD. Next thing you have to

use this ML man ML file which you can

easily find on GitHub or you if you

simply uh type install

aru CD

and you can go on this documentation and

you can follow the steps as well. So the

very first thing is you have to create a

name space. Second you have to

download and install this ML file.

So let me open my command prompt. Let me

clear it.

So it is going to take couple of

seconds.

Okay. So you can see uh Argo CD

application controller for not this one.

This is network policy. Argo CD server,

Argo CD repo server, notifications. a

lot of uh things it is going to set up

for you.

So now let's see what next we have to

run. Let me go to my presentation. Okay.

Now we need to access this uh Kubernetes

sorry Argo CD dashboards. So for that

what we can do either you can expose or

you can use a load balancer as a service

type or you if if you want to test

locally you can simply do a port

forwarding. So let me do this port

forwarding.

So let's verify first uh the pod status.

Maybe it is going to take couple of time

sorry a couple of minutes. So cubectl

get pods a sorry not a small caps a

you can see the argo cd1 running and the

this is initializing. Okay, let's wait.

Pod is initializing. Okay, let's wait a

few minutes. Once all the pod of Argo CD

is uh the status will be up and running,

then we will do a port forwarding and

then we will browse the dashboard of

Argo CD.

Okay. So now you can see the status all

Argo CD ports are running in running

state. Now let's do a port forwarding.

So let me copy this. So basically we are

going to run on port 80 sorry 8080.

Okay. Uh cubectl port forward does not

have a named port

443. Okay. Let's verify the services

first. QTL get sbc.

Sorry.

Okay. So now we need to use a name space

as well which is Argo CD.

Okay. So here we have cluster IP which

is fine.

Let's change it to 8082 not 2 is also

there.

Sorry 8081 only because the Argo CD repo

server is okay. Let's try with 8081

first. So services argo CD- server does

not have a name port 443.

Okay. So, let's see. Port 80

that's 290.

And here we will use port 480.

Okay. So, let me copy this

and

okay. So, this is the Argo CD URL which

you will get and the username will be

default admin and password you have to

decrypt it. So for that let me open my

power cell.

Okay. Now let me if you will see here

one thing uh let me go to my uh

okay let me open new one otherwise it if

I will cancel it it is going to stop the

port forwarding cubectl get

sorry hyphen n and

so we have admin secret okay so now what

we need sorry argod secrets. So now what

we need we need to uh because whatever

the password we will have here that is

in encrypted form. So we need to decrypt

the password for that what I'm going to

use not this one.

So I'm going to use pass sorry a power

cell. So let me copy this

and let me paste it.

So this is the password you will get it.

You can see we have used arcd initial -

admin secret. So if I will go to uh

command prompt. So this is the one which

is going to have the password.

So let me copy the password first.

Okay. So this is the dashboard of Argo

CD. Now what you have to do the very

first thing you should have your GitHub

repositories

and you need to create a application. So

let me create application first. Let me

create any name maybe app one and

project name we will use default

synchronization policy. So we have two

options manual and automatic. We will go

with automatic. And here you we have

selfheal option as well. In case if you

are going to make any changes on

kubernetes cluster directly then it is

going to autoheal like it is going to

make the changes in your GitHub as well.

Okay. So now we need to define here

source. So source is going to your

GitHub repository. So let me open my

GitHub.

Okay, I will go with this repository

which is K K at fundamental and let me

go to in any one of day three. Let's go.

Okay. So here we have only one ML file.

Let me see another one.

Uh let's check A5.

Okay. So here we have multiple ML file.

So let me copy this and here I will

provide the GitHub repo link. So

you can see the revision head. It means

whatever the latest commit you will have

on this repositories it is going to take

uh that particular changes. If you have

any custom path, you can define it here

as well. Now in cluster details, let me

use my okay path is required. So let me

do one thing. Let me copy this. Sorry

name space. So nameace is must like you

should have nameace in your manifest

file. Without this it is not going to

work. So let me use default name space

here. I'm not sure if I have a name

space in manifest file. Let me do one

thing. Let me use day four because I

know on day four I have uh

name spaces defined in template. Let me

go to day four

and let me open deployment one. Okay. So

here we have name space. So okay let's

go with day four one.

Now let me quickly save it.

Sorry, let me quickly hit on create.

Unable to create application.

Application server for app one is

invalid. Okay, maybe let me remove this

name.

Okay. Now let me click on create.

Okay. So the problem was with GitHub

URL. Uh I should give a URL till my

repository which is create fundamental

because the day four folder path I have

already defined in path location. So you

can see as soon as I save you can see

the status here missing progressing and

synchronize. So as soon as what we did

we just simply define our GitHub repo

URL and it is going to automatically

deploy our resources on Kubernetes

cluster.

If I will click on this,

you can see this is the project which we

have created app and under this we have

two manifest file NX - deployment NX -

replica set. Now you can see in NX

deployment we have four number of

replica sets. So if I will go to

configuration here I have defined four

replicas of NX deployment. So if I will

go to app now we have here four replicas

of NGNX deployment and similarly for

here NX replica sets three

because in this configuration I should

have replica count three

three so you can see everything we can

see uh in single dashboard here you can

see the app health which is healthy

synchronization status is synchronized

and the head head is a last commits

which is 4759.

If I will go to my repos and uh let me

see what's the last commit I have made

4759. So the same commit it used 4759

and if you need to check like more

details you can simply go to particular

uh part and you see all the live

manifest here

and similarly if you need to check the

events and if you need to check the logs

so everything you are going to see in a

single dashboard. So if you are going to

use Argo CD definitely it is going to

benefits and this what we did is simply

we have connected to GitHub repos but in

real time definitely you are going to

use Argo CD which is a continuous

delivery tool along with any uh DevOps

tool either Jenkins or maybe Ashure

DevOps I will create the video on that

as well like we will set up a complete

end to end pipeline using Argo CD but

for now just want to give a quick brief

what is Argo CD and how we can set it up

and the simple application we have tried

to deploy it. Now let's see uh one more

thing here. Let me make some changes on

deployment hyphen NGNX file. Let me

directly edit it. Now what I'm going to

do I'm change going to change the

replica account 3. Now let's see what

happen

as soon as I will commit it. Let me go

back

and let me click on refresh.

So you can see it is going to shrink the

changes because you can see stats out of

sync. Now the status is shrink again.

Now you can see one pod is going to

delete because now we have specified

replica count three in deployment

manifest file. And if you will see a

last sync which is head or we can say a

commit ID E9. If I will go and check we

have a last commit ID which is E9. So

what we did we just simply make changes

on Kubern sorry on GitHub repos and it

is automatically deployed those changes

on Kubernetes cluster. We did we did

nothing here. So if you need to verify

it, let me open it and let's verify the

pod. QTL get ports. So you can see we

have a ngnx deployment and replica set

three uh deployment port three replica

sets. It means Argo CD is working fine.

So in and if you want to see

synchronization status, you can see all

these things here and health status also

you can simply monitor here.

And similarly you can create multiple

applications and uh multiple repos. So

the main thing here you should have a

separate repositories

for your manifest file. You should not

have a single repos where you have

application source code and your

manifest files as well. So it is not a

best practice. always create a separate

repositories for your manifest files

which you can directly link with uh or

you can directly add on your Argo CCD

dashboards and for security concerns you

can use SSOS OIDC everything you can

configure on Argo CD for security point

of views and you can use certificate as

well you can see for now I'm getting SSL

warning because I have not used any

certificates but if you want you can use

certificates as well so I hope it's

clear uh how Argo CD basically work and

it is very simple and straightforward

just you need to understand how you are

going to make changes while using a

CI/CD pipeline definitely we are going

to cover in upcoming videos so that's

all for today see you next video thank

you so