Installing Kubernetes with Rancher in 2024!

>> Hi. I'm Chris Mutchler,

Principal Enterprise
Architect and

VCDX 257 from
virtualelephant.com.

And in this video,
we're going to

once again deep dive
into the world of

Rancher and show
you how you can

leverage it as
a management

tool for deploying,

configuring and operating

Kubernetes
clusters at scale.

Let's get started.

[MUSIC]

In this video,

we were going to deep dive

into Rancher once again.

Back in 2023, I

recorded my first
video on Rancher

showing you how you
could deploy Kubernetes

clusters inside of a
vSphere environment,

leveraging the Rancher
management tool.

Now, since that time,
it's become one

of my top viewed
videos on the channel,

and I thought
it was time for

an update for 2024.

One of the strengths
of Rancher

as a management tool is

its capability to be able to
deploy Kubernetes clusters

across a myriad of
environment types.

From vSphere environments,

which I'll show you

once again in this
video to Azure,

AWS, GCE, and several
other Cloud providers,

Rancher can become that

single management tool

for all of your
Kubernetes clusters,

thus allowing you to
be able to create

an enterprise grade
service offering

for Kubernetes
based on Rancher.

One of the things
that I appreciate

most about Rancher

is its simplicity and
ease of operations.

Anyone can get a
Rancher management tool

running up in their
environment in

just a few moments
once they have

a machine or operating

system available to them,

and its ability to almost

instantaneously
begin deploying

Kubernetes clusters
makes it one of

the easiest
management tools

out there to be
able to configure,

install, and
then manage from

an SRE or operations
perspective.

So let's dive
right into the UI.

Let me show you what's
changed and how

I began leveraging Rancher

once again with these
updates to be able to

deploy clusters within

a lab environment
very easily.

Once you've got

the Rancher docker
image running,

and I have it running
here inside of

a virtual machine
that I've

deployed inside of
my VCR environment,

and I've actually created

a persistent volume or

static storage for
the docker container

to be able to survive
after reboots,

the first thing
that you'll see in

the UI screen here is

instructions on how to get

the boot password from
the container itself.

Now once you've done that,

you can copy the password

straight into the UI here,

and then once
you've done that,

it will prompt you to
change the password

or accept one that they

randomly generate for you.

And once you've
done that, it

will log you straight into

the main dashboard
or UI of Rancher.

The first thing that you

want to do is actually

click on the menu and

select Cluster Management.

And from there, you want

to select Cloud
Credentials.

Now you can see you

have multiple
options here.

I'm going to go
ahead and create

a VSR credential
that points

to my vCenter Server that

I have running in the lab.

I'm going to specify
the URL for vCenter.

I'm going to make
sure that I tell it

Port 443 and then
I'm going to

enter the username and
password information

for that vCenter Server.

Once that's done, we're

actually ready to start to

pulling a cluster of
Kubernetes via Rancher.

So once we get back to

the cluster
management screen,

we can go ahead
and click Create.

This new cluster create
screen is one of

the key differences
between

the old version of

Rancher and the
new version.

You can see here that

before we had to create
some templates to

use specifically to point

to certain pieces of
information that were

relevant to our
vSphere environment

for an RKE cluster
to be deployed.

In this new version
of Rancher,

you no longer
have to do that.

And so you can
go ahead and

click Create, come
to this screen,

and you can see that
it's prepopulated with

some information
that is specific

to deploying Kubernetes

in a vSphere environment.

So as you would
typically expect,

you can give it
a cluster name,

you can give it
a description.

And then from
there, you can

see that it's created

this default pool
and this pool is

going to be a set
of virtual machines

that it's going to create.

And you can see the
default actually selects

SCT control plane as
well as worker nodes.

Now we're going to want to

separate out the control

plane from the data plane,

and so we're
going to unselect

the worker for this pool.

We're going to give
it a different name

and then we're going
to go through and

specify some information

specific to how I want

the controllers
to be deployed

for Kubernetes within
my own environment.

Now one of the
key things that

I've done prior
to getting to

this screen is
that I've already

downloaded an Ubuntu
Cloud Image OVA

and I've imported
that OVA into

my vCenter Server and

I've converted that
to a template.

And that's going
to let me specify

here on the screen which

template I want to use.

If you haven't
done this before,

just Google Ubuntu
Cloud Image.

You'll see the
first result

there will be the one
that you want to click.

You go through, you select

the version of a Ubuntu
you want and then

download the Cloud
image that you're

choosing to deploy
for your platform.

In my case,
vSphere OVA file.

So you're going
to go through.

You can give it
a pool name,

you can give it
the number of

virtual machines
that you wanted to

deploy. So I'm
going to do three.

Three is the
typical amount

of controllers that I
recommend deploying.

Have to make
sure because of

SCD and core Arm,

it needs to be
an odd number,

so 1, 3, 5, 7, etc.

And then you
can go through

and you can change

certain things like the
data center object,

the data store that it's
going to deploy to.

You can also specify
the folder that

you want the
virtual machines

to be deployed to.

And then once
you've done that,

you can go ahead and edit

the CPU and the
memory size to

be able to make
sure that you have

enough available capacity

for those controllers.

And then you can see here

in the drop down menu for

template where
I've actually

installed the Cloud
image template,

you can see that
I can select it.

Now, make sure you
select the right image.

Otherwise, when
you go to create

this cluster, it's
not going to work.

Now, once I've selected
the cluster image,

the next thing that
you can edit here is

the Cloud config
YAML area.

And so one of the
things that I

like to customize
here for all

of my nodes is
I like to add

a rancher user to
every virtual machine.

I make it part of
the wheel group,

I give it pseudo
privileges,

and then I add
an RSA key for

SSH to be able to log into

those virtual machines
once they've deployed.

And you can add all
different things here in

the Cloud config depending

on your needs and
requirements.

Like I said, at a minimum,

I recommend at least
adding an SSH key to

a user so that you

have access into the
virtual machines.

Now once that's done,

you can go ahead and
select the network.

This is the port group in

the vSphere environment
that you're going

to be attaching the
virtual machines to.

Make sure they have DHCP.

That is a requirement
that I found with

Rancher so that
the machines

automatically come up with

IP address information so

that there's
no requirement

on you to try and modify

IP configs somewhere in

the middle of the
boot process,

which typically
just doesn't work.

And then once
you've done that,

you want to go ahead and

create another pool by

clicking the plus sign

there on the lower
left screen.

Now we're going to
create our workers and

we're going to fill out

that very same
information,

including that Cloud
config file and

making sure that we choose

the right
template as well.

The last thing that
I want to do is you

scroll down on the
UI screen prior

to clicking Create is

actually to select
the CNI selection.

So you can see here in

the dropdown menu
that there's

multiple choices now for
CNIs within Rancher,

and you can see, again,

my favorite Cilium
is now available.

So we're going
to go ahead and

deploy this cluster with

Cilium to see how Cilium

is integrated through

Rancher and
what components

are there for us to
be able to start

leveraging straight
out of the box.

Now once we've gone
ahead and selected that,

we can review
everything that we've

already inputted
into the UI

just to make sure that
everything looks good.

And then we can go ahead

and click that
Create button and

it's going to
start deploying

the Kubernetes
cluster inside

of our vSphere
environment.

So once Rancher has

started creating
the environment,

we can go and we
can see that it

goes into this
provisioning state.

We can click on
the cluster name

to be able to see the
detailed information,

as well as
switching over to

the vSphere
environment to see

the virtual machines
being created.

We can watch them
from that perspective

or we can switch back to

the Rancher UI to
be able to watch

the cluster deploy and
the various states.

As before, your
mileage may vary for

how long it takes
to actually

deploy the
Kubernetes cluster.

Within my homelab
environment,

it generally takes
about 20 minutes

depending on how long it

takes to clone
the template,

as well as start up
the virtual machines,

get a DHCP address,

and then for Rancher
to take over and start

installing the
various Kubernetes

bits and services.

Typically, within
20 minutes,

I'll have a cluster that's

online and ready
for me to start

further configuring
additional services

within the Kubernetes
cluster itself,

such as persistent
storage and NFS.

Other things like the
Ingress Controller

is obviously a
critical one as well.

And so at that point,

we'll have a cluster that

we can start leveraging.

So you can see here
on the screen,

the cluster is online.

We can go ahead and
click it and we'll

actually be brought to
the cluster dashboard,

where we can see some
very basic information

around what's going on
within the cluster.

We can also download the

Kube config
file to be able

to connect to the
cluster via CLI.

I do that and I actually

run the CLI
straight off of

my Rancher virtual machine

where Rancher's actually

running as a
docker container.

And that just becomes my

central point for all of

my future configurations

and deployments
and whatnot.

So one of the things

here you can
see is there's

a lot of information

about the cluster itself.

You can see what the
nodes are doing,

you can see what
pods are running,

you can see
what additional

tools are
available to you.

And then we can
go through and

we can actually
start installing

Prometheus and Grafana as

a monitoring tool first

thing within the
environment.

Now, the first
thing that I do and

that you can see here
on the screen is

I actually install
an NFS pod

inside of the cluster
that connects back

to an NFS server that
I have running within

my lab that I can

then create persistent
volumes on.

So I install
the NFS piece.

If you want to see
how to do that,

I'll link in
the description

below a link to my
GitHub repository for

my Kubernetes sample app

where I use this same pod

to be able to do persistent
storage for RabbitMQ.

And then I go ahead
and I've created

a storage class
specific that I want

to leverage off of

my NFS server within

this Kubernetes
environment.

And then once
I've done that,

then we're ready to
actually install

Prometheus and Grafana and

make sure that we have
persistent storage for

those metrics so that they

maintain state over a
long period of time.

And so you can
just click on

the Install Monitoring
inside of Rancher.

Again, this is
one of the things

that Rancher does
really well for us

is that it bundles a lot

of these things that
are commonplace

within Kubernetes
clusters directly

through the Rancher UI
for us to leverage,

so we don't
have to go hunt

down other YAML files
or figure out what

other people have
done for how they've

deployed some of the
common applications.

And so you can see all
of the charts within

Rancher for the
various things

that we want to be
able to install.

And again, they
link right there in

the top right corner or

install monitoring
on a fresh cluster.

We then select the
monitoring chart,

and then we get here to

the screen for Prometheus
and being able to

configure Prometheus and
Grafana and customize

that configuration
for our environment.

So the first
thing that we're

going to do is
we're going to

select where we're
going to install it,

so we'll install it in
the default project,

and then we're
going to go through

here and we're
going to modify

Prometheus to be able

to leverage
persistent storage.

So we're going to
check that box

there on the screen for
persistent storage,

and then you
can see we have

an expanded set of

options that we
want to be able

to leverage specific to

persistent storage and
persistent volumes.

You can see when we click

on that storage
class name,

you can see that
there's the NFS client.

That's what default
comes with the pod,

and then I've created

that standard
storage class

for my Kubernetes cluster.

So we go ahead and
click Standard,

that's now hooked back
into my NFS server.

You can specify or

modify the size
if you need to.

And then we're going to

do a similar thing
for Grafana.

We're going to select that

enable with PVC
template radio button,

and then it's going to

populate some
information on

the screen force that's

specific to
persistent volumes.

Again, pick that standard
storage class name,

read write once many,

and then we can
specify a size,

and then we can go ahead
and click Install.

And Rancher is going to

automatically install
Prometheus and Grafana.

It's going to preconfigure

it with a wide variety of

dashboards right out of

the gate for us to
be able to leverage.

And it's also
going to open up

this mini shell window

within the UI as it shows

us how it's installing

Prometheus and Grafana

within the
Kubernetes cluster.

And then once that's done,

we'll be able to
see some tie ins to

Grafana immediately
in the branch or UI

for our cluster without
actually having to

go out to the Grafana
service that's running.

Although there
is a hyperlink,

and you'll see us
hit it here on

the screen in
just a moment.

To be able to select
the Grafana UI,

it opened a new window
for us in our browser.

Once we click that
Grafana UI hyperlink

and it opens up
the new browser,

then we're going
to see a myriad of

other dashboards that
instantly become

available for
us to be able

to leverage inside of

our Rancher Kubernetes
cluster that it's

deployed and is
managing for us.

And it's really this
streamlined operation that

differentiates
Rancher from several

of the other Kubernetes
management tools.

So it's had a little bit
more time to bake in

these features to get

this UI streamlined
force and to

make it very easy for

operators and Cloud
architects to be able

to manage and operate

a Kubernetes service

offering directly
through Rancher.

And so there you have it.

Once again, we can
see how streamlined

Rancher is as a management

tool for Kubernetes.

We can see how
quickly you can

spin up a Kubernetes
environment.

In future videos, I'm
going to show you how

to leverage
Rancher for Azure,

as well as deep
dive more into

the Cilium integrations
that we now see

within Rancher so that we

can understand what
is built in from

a Cilium stack perspective

via the Rancher installer,

and how we're going
to manage Cilium and

Rancher Kubernetes
clusters from

a life cycle perspective
going forward,

as far as doing
version upgrades,

deploying additional
tools like Hubble,

being able to leverage

service mesh via Cilium,

as well as the Ingress
Controller for Cilium

as well so that
we can it be

our simplistic CNI
stack that runs

everything for us from

a networking perspective

within our
Kubernetes clusters.

Now if you're enjoying
this content,

I encourage you to
subscribe to my channel,

turn on those
notifications

and hit the like button

and let me know
what you thought

of this video in
the comments below.

I hope this
updated version of

Rancher is
helpful to all of

you out there who are

striving to learn more
about Kubernetes.

As I've talked about in

previous videos this year,

I think that Kubernetes is

the one thing that
all of us should be

focused on in 2024 from

a skill set perspective

so that we can broaden

our capabilities
to be able to

make ourselves
more marketable

out there in the
job market in

these difficult times that

we find ourselves in.

So I hope that you're
enjoying the series,

I hope that you're

finding these
videos useful.

Please leave me
a comment below.

Please reach out to me
on X, @chrismutchler.

Let me know what
you think and

until next time, I'll
talk to you soon.