►
From YouTube: From Zero to Jakarta EE in Kubernetes in Under 40 Minutes | Jakarta EE @ KubeCon NA 2020
Description
Ondrej Mihalyi presents as part of the Eclipse Foundation virtual booth at KubeCon NA 2020
To get started with Jakarta EE, the best place to look at is the https://jakarta.ee page. I suggest looking at tutorials in the Resources Documentation page https://jakarta.ee/resources/#documentation.
The second best place to look at is the Payara documentation, especially the Ecosystem section with info about using Payara Jakarta EE runtimes with Maven, Gradle and many popular IDEs: https://docs.payara.fish/community/docs/5.2020.6/documentation/ecosystem/ecosystem.html
If you want to study more about running Jakarta EE applications in Kubernetes, I recommend to first learn how to use Payara Docker images: https://docs.payara.fish/community/docs/5.2020.6/documentation/ecosystem/docker-images.html. And then study some of these guides about using Payara Micro in Kubernetes: https://www.payara.fish/solutions/containers-and-the-payara-platform/the-payara-platform-and-kubernetes/.
Get Social!
Subscribe: https://www.youtube.com/user/EclipseFdn
Follow us on Twitter: https://twitter.com/JakartaEE
Like us on Facebook: https://www.facebook.com/JakartaEE/
Join us on LinkedIn: https://www.linkedin.com/groups/13597
Tag us: #JakartaEE
A
A
A
A
A
A
A
A
A
B
A
B
A
A
A
So
if
you
want
to,
if
you
want
to
scale
the
front
end,
we
will
also
create
a
load
balancer
here
and
we'll
have
multiple
multi-performance
services.
So
in
case
there
are
multiple
users.
Each
of
them
can
be
routed
to
a
different
front-end
service,
and
each
front-end
service
will
then
go
through
load,
balancer
and
access.
B
A
A
A
A
A
B
A
A
So
first
I'll
show
you
how
I
got
it
running
on
pyro.
In
the
first
place,
it's
not
a
big
deal.
The
project
is
normal
war
war,
maven
project
to
be
able
to
run
it
just
from
command
line
or
from
net
news
id,
which
basically
uses
maven
configuration
to
launch
the
application
we
use.
Spire
micro,
maven
plugin
specify
the
version
of
the
repair
runtime.
A
A
We
disable
clustering
features
because
they,
just
if
we
don't
use
clustering,
later
slow
down
the
startup
and
we
specify
which
application
to
deploy
for
for
development
purposes
will
deploy
the
exploded
application
so
that
changes
in
turning
the
ide
automatically
propagate
and
by
our
microbial
will
redeploy
the
application
and
we
specify
the
context
mode.
So
everything
is
at
the
top
level.
A
B
A
A
So
the
the
side
effect
is
that
if
you
have
docker
d
money
installed,
it
won't
be
in
your
local
repository
in
the
criminal
demon
to
launch
it
locally.
You
have
to
pull
it
from
the
from
the
repository
which
chip
uses
to
deploy
to
and
then
you
can
run
the
run.
The
the
container
locally
and
use
the
image
locally.
A
This
is
mainly
for
development,
because
in
continuous
integration
systems
you
would
most
probably
want
to
deploy
to
a
docker
repository.
It
also
allows
building
a
tar
or
a
compressed
compressed
image
that
you
can
import
to
any
repository,
so
you
can
build
locally,
even
if
you
don't
have
a
docker
demon
and
then
import.
This
image
to
any
local
repository
manually,
if
you
want
in
practice,
you
would
rarely
use
this
option
so
I'll
start
with
using
this
plugin.
A
A
B
A
If
you,
if
you
are
going
to
use
this
in
production
like
environments,
you
would
probably
use
or
increase
this
variable
make
it
configurable
outside
of
the
maven
bomb
file.
So
we
would,
for
example,
use
version
one
or
version
zero,
one
or
just
use
a
variable
image
version
and
specify
it
externally.
When
you
run
the
maven
build,
but
now
we
will
work
with
only
the
latest,
which
is
the
default,
and
then
we
need
to
specify
some
configuration
for
the
pyro
docker
container,
which
is
specific
to
paradox.
A
So
first
thing
is
still
for
for
jib.
This
is
algebra,
treats
docker
containers
which
deploy
war
applications,
so
it
it
knows
that
you
are
working
with
a
web
application,
because
the
type
of
your
maven
project
is
war,
so
it
knows
how
to
handle
the
files
generated
by
the
maven
build
and
it's
gonna
provide.
A
So
this
up
root
variable
where
you
can
we
can
we
specify
where
to
copy
the
resulting
artifacts
to
into
the
docker
container,
so
for
docker,
compiler
docker
container
according
to
the
documentation,
the
deployments
are
by
default
in
a
prepared
environment,
pre-created
environment
opted
by
our
deployments,
and
if
we
deploy
our
application
as
root.or,
it
will
automatically
create
use
context
roots,
slash
the
top
level
context
root.
We
don't
need
to
specify
anything
else.
A
Then
you
can
see
the
pattern
here.
It's
the
same,
no
cluster.
We,
we
use
the
no
cluster
option
here.
It
doesn't
matter
if
I
use
capital
c
or
small
c
parameter,
doesn't
care,
and
then
we
deploy
the
web
application
that
we
copied
to
this
uproot
this
one.
This
time
it's
a
folder,
so
jeep
plugin
will
selectively
copy
only
changed
files
from
the
explosive
folder
and
to
save
some
some
uploads
when
you
did
redeploy
your
docker
image.
This
is
nice
about
jib,
which
contains
a
lot
of
intelligence.
A
A
You
only
now
have
to
know
maven
to
use
it
and
to
configure
you
don't
need
to
use
the
docker
file.
Everything
is
here
a
lot
of
what
you
need
is
generated
for
you
and
it
supports
natively
web
applications
so
that
it
optimizes
how
the
docker
image
is
layered,
how
the
files
are
layered
and
it
doesn't
always
overwrite
the
whole
war
file,
only
the
files
that
that
changed.
A
A
A
A
A
A
A
A
A
A
A
A
A
A
This
is
everything
we
need
to
to
really
build
a
docker
image
and
put
it
to
the
repository
if
it
was
empty.
You
would
now
see
a
new
image.
We
see
that
the
images
are
already
there,
but
they
are
refreshed,
or
specifically
this
one.
If
we
go
to
the
parent
project
and
run
the
same
command
to
build
also
for
the
other
service,
you
will
see
that
both
are
built.
A
That's
very
easy
second
step
is
to
specify
the
kubernetes
deployment.
So
now
we
have
a
talker
image.
We
can
run
each
docker
image
separately,
but
it's
fairly
easy,
but
I
won't
do
it
because
it's
it
would
require
some
time
we
don't
have
and
we
we
would
also
need
to
define
how
services
see
each
other,
how
they
connect
to
each
other
I'll
jump
down
to
the
code.
A
The
only
important
line
in
the
code
is
the
the
line
where
the
front
end
discovers
or
connects
to
the
duke's
h
service.
So
we
have
a
duke's
h,
connector,
which
is
a
wrapper
around
the
rest
call
or
all
the
rest
calls
to
the
duke's
age
service
and
it
needs
to
know
the
url
which
it
needs
to.
It
should
call
so
in
in
this
case.
In
our
case,
we
will
read
the
base
url
from
an
environment
variable
and
dxhurl
by
default.
A
A
That's
the
main
starting
point
where
you
can
find
now
now.
After
some,
while
the
the
jakarta
e
page
contains
a
lot
of
resources
that
weren't
there
and
the
ecruz
foundation
is
doing
very
good
job
and
all
the
jakarta
all
the
people
involved
in
jakarta,
he
are
doing
very
good
job
in
aggregating
all
these
resources.
A
Sometimes
they
are
not
easy
to
find
because
there's
a
little
simply
a
lot
of
them.
But
I
would
point
you
to
this
resources.
Documentation
page,
which
links
to
jakarta
e
tutorial,
and
here
is
the
jakarta
first
cup
tutorial,
which
I
based
my
application
on,
and
then
there
is
a
full-fledged
cartage
application
called
cargo
tracker,
which
demonstrates
a
lot
of
jakarta
apis,
so
I
think
almost
all
of
them
in
just
one
application.
A
So
I
would
start
by
studying
all
these
three
and
there's
another
question:
how
do
I
set
it
up
on
cloud9
cloudline?
Is
I
think,
and
cloud
ide,
that's
always
specific
to
every
ide,
but
I
think
it's
good
to
start
with
the
maven
projects
and
which
most
ides
support
in
some
way
or
another
or
and
when
you
want
to
start
with
maven.
There's.
A
A
A
A
Our
dependencies-
yes
here
so
basically
every
every
jakarta
maven
project
should
only
have
one
single
dependency
and
there
will
be
either
jack
at
the
web.
Api
or
jakarta
ee
api
depends
on
which
profile
which
subset
of
apis.
You
want
to
use
in
your
application,
either
it's
a
full
one,
api
or
a
subset
of
which
is
called
web
profile.
A
Some
implementations
of
jakarta.
You
only
support
this
subset
like,
for
example,
player
micro,
which
is
somewhere
in
the
middle.
It
supports
all
of
the
web
profile
and
a
lot
from
the
full
profile,
but
not
all
of
it.
So
it's
safe
to
start
with
web
profile
web
api
and
then
add
individual
apis
to
it.
Actually,
there's
player
micro
archetype.
A
A
I
would
start
with
the
the
chakartay
tutorial
in
the
documentation
section
and
all
the
other
tutorials
and
then,
if
you
want
to
code
in
an
ide,
it's
probably
best
to
start
with
an
example
application
from
these
tutorials
or
if
you
want
to
start
from
scratch,
just
use
some
generator.
The
standard
one
is
maven
archetype.
There
are
several
archetypes.
A
There
are
multiple
implementations
of
the
chakarta
api.
So
for
each
implementation
there
is
a
an
archetype
or
some
generator,
but
there
are
also
generators
for
the
vanilla,
plain
jakarta,
ee
well
and
they
will
generate
basically
a
product
that
contains
only
one
single
dependency
and
all
the
rest
is
specific
for
for
prior,
like
like
this
pill
of
materials,
which
defines
versions
of
all
the
components
of
fire.
If
you
want
to
include
them
in
your
project
and
compile
against
them,
okay,
so
I'll
continue.
A
I'll
continue
with
how
to
deploy
to
kubernetes.
Traditionally
you
know
for
kubernetes
you
would
write
yamlo
files,
which
are
yeah
quite
quite
easy
to
use
for
a
simple
service
for
simple
configuration
you
can
see.
There
is
a
definition
of
the
service.
There's
a
lot
of
boilerplate
in
there
that
just
defines
api
versions
and
metadata.
A
A
A
Then
you
need
to
specify
which
instances
or
where
are
the
instances
of
your
service.
This
is
by
applying
a
selector
here,
it's
a
pathfinder,
but
we
will
see
it
will
be
for
for
our
duke's
age
service.
It
will
be
duke's
age
and
and
then
basically
just
a
port.
That's
that's
a
service
with
a
load
balancer.
It's
very
simple!
A
A
It
contains
two
main
goals,
or
it
also
contains
a
build
call
which
basically
does
the
same
thing
as
chip,
but
in
a
different
way,
and
I
already
started
using
jib
and
like
it
how
it
builds
the
toker
images
and
uploads
them
to
a
docker
repository
in
one
step.
So
I
chose
to
not
change
that
and
only
use
what
I
need
from
the
kubernetes
maven
plugin.
B
A
A
First,
we
need
to
specify
the
the
docker
repository
we
already
did
it
for
for
the
jib
plugin,
but
to
do
it
we
did
it
in
in
the
maven
parent
project,
but
to
specify
it
also
for
the
kubernetes
plugin.
We
need
to
redefine
it
as
a
jq
docker
registry
property,
which
points
to
the
docker
registry
property
that
we
defined
before
so
now.
Both
are
the
same.
A
A
We
needed
to
do
just
to
be
able
to
generate
the
kubernetes
files
that
points
to
correct
repository
and
correct,
correct,
docker
images,
so
we
only
need
to
specify
the
docker
repository
because
kubernetes
maven
plugin
is
so
nice
that
it's
able
to
derive
all
the
rest
from
the
even
configuration
so
by
default.
It
does
the
same
what
we
did
for
jip.
A
There
are
you,
can
you
can
override
this,
but
this
is
the
same
default
that
works
most
of
the
time,
at
least
for
demo
purposes.
So
we
won't
need
to
change
this.
We
only
need
to
point
it
to
the
correct
docker
registry,
so
we
will
really
find
this
variable,
as
the
one
is
that's
expected
by
the
communities.
Maybe
plugging
here.
A
A
Probably
not
where
you
would
expect
them,
but
it's
pretty
pretty
nice
and
they
will
be
in
the
classes
directory,
so
they
will
end
up
as
meta
information
inside
your
web
application.
So
anybody
who
will
open
your
web
application
bar
package,
we'll
we'll
see
the
intended
deployment
and
can
use
it
to
deploy
the
companies.
A
Let's
see
there
is
a
deployment
automatically
generated,
contains
a
lot
of
metadata.
That
is
not
very
interesting.
What's
interesting,
when
what
you
really
needed
is
this
specs
part?
It
specifies
how
many
replicas,
so
how
many
instances
of
our
service
we
want
to
use-
and
it's
specified
selector,
which
is
upduke's
age,
is
the
most
important,
but
you
also
use
all
these,
so
there
are
several.
These
are
like
tags
that
are
used
to
find
this
deployment.
When
you
create
some
other
kubernetes
component
that
needs
to
refer
to
to
a
deployment.
A
A
This
is
like
a
template,
so
this
container
defined
here
will
be
used
to
start
any
port
any
container
when
defined
by
the
deployment.
So
if
you
scale
up
to
three
services,
it
will
launch
three
docker
containers
based
on
this
metadata
and
we
will
provide
it
with
this
configuration.
So
what
is?
Does
it's
again
specifies
some
environment
variables?
A
What
is
what
is
not
very
nice
for
us.
Is
this
image
pool
policy
which,
because
in
development,
it
would
mean
that
if
we
redeploy
just
this
comparative
configuration
and
we
have
a
new
docker
image
under
the
same
tag,
the
latest
stack.
So
if
we
just
override
the
existing
docker
image
in
the
docker
repository
and
with
this
configuration,
the
document
will
be
downloading
only
the
first
time
it's
needed
and
then
it
won't
get
updated
or
even
if
we
change
our
application.
A
A
A
A
A
A
A
A
So
when
we
go
to
the
kubernetes
dashboards
now
we
see
we
have
first
cup
bar
in
duke's
age
services
and
let
me
go
to
deployments
each
of
the
services
points
to
the
respective
deployment
and
then
there
are
just
two
ports,
one
port
for
for
each
service
and
deployment,
but
we
can
have
multiple
parts,
which
means
multiple
instances
of
the
service.
We'll
come
back
to
this
so.
A
A
Inhale
hello,
voila
everything
is
working
now
it
the
front
end
calls
the
back
and
service
the
duke's
h
service.
It
will
retrieve
the
age
of
duke
will
print
the
ip
of
the
service
you
see,
if
you
remember
previously
on
when
we
were
lucky
running
locally.
It
was
one
two,
seven
point,
zero
point,
one
one
now
it's
ten
point
something
and
most
mostly
incoverities
ip
addresses
start
with
ten
and
now
one
thing,
one
more
thing
we
did.
A
I
did,
which
I
didn't
tell
you
is,
I
pointed
pointed
the
front
end
to
the
back
end,
and
this
is
by
host
name
and
on
localhost.
It
was
always
using
localhost
to
point
to
the
to
the
external
service,
but
now,
if
it
would
point
to
localhost,
it
wouldn't
find
anything
because
there's
no
dxh
service
running
on
locally
on
the
same
in
the
same
container
needs
to
use
something
in
kubernetes
to
find
the
other
container
and
that
something
is
the
service.
So
also
this
internal
duke
service
has
a
service
in
kubernetes.
A
A
A
A
A
A
A
A
A
B
A
A
A
A
A
Images
that
are
available
for
you
and
there's
documentation
how
to
use
them
and
how
you
can
specify
the
configuration
for
these
images.
If
you
want
to
get
the
money
in
comparisons
you
basically
just
dockerize
your
application,
using
player,
micro
or
or
any
other
jakarta
implementation
and
then
study
how
to
just
get
running
with
kubernetes.
A
A
Okay
insert,
so
this
talk
will
be
recorded,
so
probably
just
try
to
go
through
it
and
I
will,
or
the
eclipse
foundation
will
attach
my
demonstration
project
to
to
the
recording.
So
I
hope
that
you'll
be
able
to
go
through
it
and
see.
I
I
have
some
documentation
in
in
the
project
and
that
would
probably
show
you
how
to
get
do.
Do
the
steps.