►
From YouTube: IoT powered by Microprofile – Microservices in practice | Jakarta Tech Talks - August 2019
Description
This talk will show you how easily you can design a good production-ready, secure, modular Java full-stack system with IoT devices based on the Enterprise stack with Microprofile combined with good continuous deployment infrastructure patterns. There will be code, demos, and (buzzword) dragons.
///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
Want more great Eclipse Community webinars? Become a member of our Jakarta Tech Talks Meetup: https://www.meetup.com/jakartatechtalks
A
Hello,
everyone
can,
you
guys
hear
me
we're
now
live
I'm,
hoping
that
everyone
can
hear
us
and
see
us.
We
are
today
here
for
another
Jakarta
Tech
Talk.
Today
we
have
IOT
powered
by
micro,
profile
micro
services
in
practice
and
we'll
see
how
things
are
working
out
and
with
with
us,
we
have
both
Rustom
Mac,
Madero
and
madhawk
Kim.
A
Please
correct
me
with
my
pronunciation
and
I
will
invite
people
to
post
their
questions
in
the
chat
or
there
is
a
low
in
the
middle
of
the
screen.
There
is
a
ask
a
question,
a
place
where
you
can
ask
questions
and
mothers
now
things
are
going
and
with
that
I'll
turn
it
over
to
Rustin
and
mod,
and
if
there
are
any
issues
with
the
crowdcast
with
with
the
broadcasting,
please
let
me
know
in
chat.
You
have
with
this
fabulous
opportunity
to
use
the
crowd
cast
with
us
for
the
very
first
time.
B
A
C
B
C
B
C
B
Yeah
well,
the
thing
was
that
we
it's
it's
kind
of
introductory
I
mean
before,
for
some
of
you.
Who've
worked
with
these
kind
of
things
before
it
might
seem
to
be
kind
of
introductory,
but
in
in
a
way
it's
also
in
a
way
of
trying
out
new
technology
and
trying
to
find
a
kind
of
fun
project
to
to
try
that
thing
out.
B
Instead
of
using
microservice
and
everything
to
write
like
super
hardcore
enterprise
architecture,
kind
of
stuff,
you
can
actually
start
with
something
fun
and
then
it
will
be
much
easier
to
get
started
with
a
few
other
projects.
So
this
was
our
way
of
learning
learning
stuff
and
then
we
kind
of
want
to
share
the
experience.
What
we've
learned
and
how
things
worked
out
with
the
rest
of
here
yeah.
C
And
we
also
when
we
started
this
project,
we
found
that
there
were
not
that
many
example
projects
so
micro
profile.
They
were
like
a
couple
of
examples
on
some
parts
of
it,
a
couple
of
examples
on
other
parts,
but
few
examples
out
there
that
showed
lots
of
micro
profile
in
one
application.
So
we
want
you
to
make
that
as
well,
so
you
can
have.
If
you
want
to
see
some
code
examples
and
such
yeah
well,
we
are
creating
an
example.
Yeah.
C
So
should
we
go
on
to
yeah
what
we're
talking
about
them,
so
yeah
new
office,
new
lights?
Yes,
so
we
just
mentioned
lights
and
currently
we're
in
there
in
a
company's
office.
We
moved
here
the
company
moved
here
a
year
ago
and
that's,
but
about
the
same
time
when
we
started
looking
into
micro
profile.
So
we
thought
oh
new
office.
Okay,
let's
we
need
some
light
lights
in
the
office.
So
let's
well
that's
a
great
opportunity.
So
that's
great
enough
for
that.
Yeah.
B
B
We're
gonna
do
like
bleeding-edge
like
controlling
things.
I
mean
there
is.
There
is
a
kind
of
system,
and
this
kind
of
craziness
as
well,
because
thing
is
that
you
might
want
to
have
when
you're
moving
to
like
brand
new
building
with
all
this
fancy
things
you
might
want
to
to
do
like
to
adjust
lights
depending
I,
know,
weather
and
like
time
of
the
day,
maybe
because
well,
since
we
live
in
Norway,
we
have
a
little
bit
well
slightly
dark
winters.
B
You
know
the
Sun
goes
up
kind
of
late
and
goes
up
goes
down
quite
quite
early,
so
you
know
you
might
want
to
lighten
things
up
a
little
bit
in
an
office
or
like
inside
the
buildings
right.
So
the
idea
was
to
to
create
something
that
will
be
controlled
by
outside
conditions
like
sunny,
cloudy
things
like
that,
or
maybe
temperature
or
time
of
the
year
and
things
like
that
and
then
well,
we
thought.
Maybe
we
should
do
something
about
it.
Yeah.
C
A
B
B
They're
selling,
like
like
big,
like
things
that
simulate
Sun
for
you,
so
in
the
mornings
and
stuff,
so
you
actually
feel
bit
more
awake
and
stuff,
so
it
might
be
kind
of
weird
thing
to
think
about
for
some
countries,
but
I
I
think
some
of
the
countries
like
like
also
we
talk
to
Sonia.
She
lives
in
Canada
and
I.
Guess
it's
kind
of
a
little
bit
of
the
same
thing.
B
B
Well,
actually,
the
enterprise
developers
done
well
mentioning
the
idea
was
again.
We
wanted
to
learn
stuff
I
mean
the
idea
was
not
to
create
a
super
production
already
light
system
and
stuff.
It
was
like
okay,
let's
just
use
it
as
a
fun
project
and
then
learn
lots
of
stuff.
So
we
ended
up
on
purpose
to
some
extent,
I
guess
over
engineering,
stuff,
I'm,
creating
like
microservices
and
crazy
stuff,
because
I
mean,
if
you
would
be
doing
it
like
proper
for
like
production
already
stuff,
then
it
would
probably
maybe
build
it
a
little
bit
simpler.
B
But
then
we
wouldn't
learn
all
the
other
things.
So
we
kind
of
went
all
the
way
and
went
to
bit
overboard
with
because
of
the
enterprise
stuff,
because
this
is
what
we
do
it
every
day
like
developing
software
for
and
to
enterprise
software
ie
kind
of
software
right.
And
then
that's,
that's!
That's
where
the
enterprise
developer
parts
comes
in
yeah.
C
B
B
C
B
And
and
then
you
have
obviously
things
like
it.
Well,
obviously,
you
need
a
time
service
that
will
provide
surprise,
surprise
fine,
so
it
will
be
just
tell
you
what
time
of
the
day
it
is,
and
maybe
what
if
it
is
a
weekend
or
not
things
like
that
right,
you
will.
You
will
have
a
weather
service
that
will
take
care
of
like
outside
temperature
or
Sammy,
cloudy
kind
of
things
like
conditions,
yeah
weather
conditions,
yeah.
B
B
A
C
So
it's
like
you're
the
architecture,
so,
okay
yeah,
but
let's
start
by
saying
what
what
my
profile
is.
I
guess:
I,
guess
that
not
everyone
are
familiar
with
my
profile.
Yet
later
at
least
not
that's
not
my
experience
of
my
experience
is
that
many
people
I
knew
to
make
a
profile
which
is
perfect,
defines
yeah,
that's
great.
It
can't
it
learn
more
so
easily
put
like
if
you
want
to
summarize
make
your
profile
in
one
sentence:
it's
Java
EE
for
microservices
yeah,.
B
And
it's
it's
actually
it's
very
true
because
we
did
but
did
this
talk
quite
a
few
places
and
we
did
kind
of
raise
of
hands.
There
was
a
quick
game.
How
many
of
you
have
used
well
Java
and
then
it
would
be
more
most
of
the
people
would
raise
hands
and
Joey
well
again
a
lot
and
then
you
will
be
like
okay,
how
many
of
you
use
spring
booth
and
then
they
will
still
be
kind
of
majority
of
the
room
and
then
I
was
like
Michael
profile.
B
It
will
be
much
less
because
people-
it's
it's
not
I,
mean
I,
can't
really
say
it's
a
new
thing
anymore.
It's
been
there
for
a
little
bit
of
time,
but
it
still
is
less
known
kind
of
thing.
So
that's
that's
one
of
the
reasons
we're
doing
this
talk
as
well
to
kind
of
spread
the
awareness
of
that
thing
and
explain
and
show
that
it's
actually
very
easy
to
get
hosts
started
and
music
and
everything
yeah.
C
Yeah
yeah,
in
the
same
exactly
the
same
way
well
most
of
the
same
way
as
Sally
or
Jakarta
yeah
I
guess
it
should
say.
Yes,
sir
Katya
micro
profile
is
a
standard.
So
it's
like
it's
not
the
micro
profile
implementation
as
such,
but
it's
a
standard
which
ever
wonders
can
implement,
which
should
come
back
to
yeah.
B
C
C
It
is
I
like
the
short
work
or
like
the
short
way
to
put
at
it.
It's
like.
If
you
are,
if
you
are
familiar
with
Java
EE,
Tecate
II,
then
the
same
concepts.
The
same
things
are
in
micro
profile,
whereas
in
if
you're,
switching
to
spring
butyou
have
to
learn
them,
learn
this
bring
way
of
doing
things.
Yeah.
B
So
the
concepts
will
be
think
you're
used
to
kind
of
the
Java
Jakarta
EE
version
of
doing
things.
It
will
be
constants
called
concepts
and
the
way
of
doing
things
will
be
very,
very
similar.
So
you
don't
really
have
to
learn
much
new
stuff
I
mean
even
though
we
IT
people,
love
learning
and
new
ideas,
new
stuff,
but
something
you
know.
Sometimes
it's
easy
to
stick
to
what
you
know
and
stuff
like
that.
There's.
C
A
B
B
That's
the
next
one,
so
it
is
it's
quite
quite
a
quite
a
well
spread
community
I
would
say:
I
mean
it's
like
it's.
As
you
can
see.
It's
quite
a
lot
of
difference.
Both
companies
that
that
you
provide
application
servers
like
I
know
they
are
ready
at
IBM,
Tommy
tried
yeah
Oracle,
so
you
have
all
cumulus
and
then
you
have
also
Java
communities,
and
then
you
have
other
companies
that
kind
of
like
Microsoft
that
they
don't
have
their
own
application
server,
but
they
still
kind
of
support
this
and.
C
C
B
B
So
this
is
somebody
else
looking
at
the
standards
and
writing
their
codes
to
to
fulfill
those
requirements
in
a
way
right
and
well
there
you
can
see
quite
a
lot
of
I
would
say:
I
would
say
usual
suspects
there,
because
I
mean
there
is
a
lot
of
things
that
you've
seen
before
you
have
a
Liberty.
You
have
told
me:
II
have
thorn
tales
of
corn
pale
in
case.
If
you
wondering
this
is
kind
of
a
spin-off
of
has
been
a
little
bit,
so
it's
like
fork
in
a
way
off.
Well,
what
was
wildfly
a
point.
C
B
C
A
C
It
was
a
release
like
coffee,
ergo,
so
Quercus
is
read,
that's
new
application,
server,
a
drawer
lightweight
and
so
on.
We
we're
going
to
be
using
it
on
an
example.
Later
on,
like
one
of
the
applications
is
running
in
workers,
but
what
you
might
note
you
might
ask
them,
why
aren't
they
shown
here
and
that's
my
Cobra
Quercus
is
implementing
a
most
of
the
micro
profile
standards,
but
not
all
of
them.
A
C
B
C
B
So
it's
kind
of
its
kind
of
pick
and
mix
kind
of
thing
where
you
kind
of
stand
in
front
of
a
big
wall
of
I,
don't
know
like
it
chocolates
whatever,
and
you
just
pick
whatever
you
want
and
put
it
in
your
bag,
so
this
is
kind
of
in
a
way
it
is
like
that-
and
this
is
this-
is
what
the
this
micro
profile
standard.
It
was
kind
of
picked
for
us,
that
is,
that
makes
the
development
of
micro
services
and
stuff
like
that,
much
easier
and
much
more
smooth.
C
It's
like
the
underlying
ID
behind
my
profile,
at
least
as
far
as
that
I
haven't
doing
just
that
the
things
you
typically
need
in
a
micro
service
yeah
nowadays
and
I
think
it's
yeah,
the
usual
suspects,
this
dimension,
but
also
yeah
health
check,
which
is
perfectly
fine
combined
with
kubernetes.
So,
if
you're
doing
cubanía,
this
lightness
probe,
Redden's
probe
perfectly
fits
with
hope,
tech
and.
B
Then
you
have
metrics
that
will
also
fit
well
with
the
other
things
that
are
there
for
measuring
things
and
well,
you
have
obviously
fault
tolerance
and
all
those
kind
of
things
that
you
really
most
of
the
time.
You
wanted
to
really
get
away
with
the
micro
service.
That
does
not
support
those
things
or
you
would
want
to
support
that.
C
B
And
the
thing
is
that
so
what
happens
is
that
for
each
version
that
comes
out
so
now
we're
looking
at
3.0?
That's
been
released
what,
in
month
two
months
ago
and
for
each
version
the
things
are
kind
of
updating.
So
in
most
of
the
case
I
mean
lately,
it's
been
not
as
much
changes
in
inside,
like
what
boxes
are
added
or
removed
most
of
the
most
they
added,
but
then
they
cannot
keep
on
upgrading
versions
of
that
mostly
so
rest
lines
has
been
there
for
quite
some
time,
but
now
it's
version
1.3.
B
So
it's
been
updated
or
healthcheck.
So,
as
you
can
see,
the
blue
ones
has
been
updated
since
last
version,
and
then
you
have
the
PD.
What
or
any
color
is
I
don't
know.
Yellow
orange
kind
of
whatever
that
thing
is
is
that
there
is
no
change
from
last
time
and
it's
getting
I
guess
it's
getting
a
bit
more
stable
now.
So
it's
not
really
that
much
changes
but
yeah
yeah.
C
We
we
started
doing
this
coding
yeah,
basically
before
working
3.0,
and
you
have
to
create
from
two
point
two
to
three
point:
zero
took
us
30
minutes,
not
even
that's
like
you
have
no
diversion
in
some
pom
piles
and
it
was
really
minor
changes
and
afterwards
everything
worked
with
quit,
needs
it
upgrade
so
yeah
it's
starting
to
stabilize
started
it.
A
stable
I,
see
it's
quite.
B
B
I
mean
like
once
you've
done
it
it's
much
easier
to
create
new
ones,
because,
like
again,
when
we
worked
on
this
for
the
first
time,
it
was
a
bit
unusual
and
stuff
like
that.
That
was,
you
know
how
things
work
and
oh,
this
is
how
they
make
it
work
and
stuff
like
that.
But
then,
after
that,
I
think
each
of
us
created
a
bunch
of
like
new
services
just
for
fun
in
microphone,
and
it
was
very
easy
and
very
quick,
but
well
alright.
C
And
so
in
micro
services,
typically,
you
care
about
okay,
how
big
is
it
so
fast
as
it
startup
these
kind
of
things,
whole
lightweight
I,
said
yeah,
just
some
people
claim
the
Java
is
heavy
and
so
on,
which
is
no
longer
necessarily
true.
Yeah
I
would
say
yeah.
This
is
an
example
of
that
this
one's
from
remember
last
year
or
in
the
year
before
no
yeah,
you
can
see
what
form
the
previous
name
but
still
I
guess.
B
B
Yeah
and
then
I
guess
there
is
a
question
about
what
about
Quercus
I,
suppose
that
you're
asking
about
the
startup
times
and
everything
will
get
if
you're
doing
that
I
have
an
answer.
A
very
short
version
of
answer
is
that
we'll
get
back
to
that
and
it's
crazy
I
mean
and
then
that
cliffhanger
I
think
move
on,
but
we'll
get
back
to
that.
If
it
was
not,
this
question
you
were
asking
then
try.
Try
again,
you
know
we'll
try
to
answer
that.
Yeah.
A
B
This
part
and
that's
what
it
does
it
so
we're
using
for
this
demo,
were
using
Philips
hue.
But
the
point
is
that
you
could
use
anything
else
so
that
control
servers
the
service
talks
to
a
library
that
was
created
by
Philips,
so
a
jar
or
kind
of
file
that
is
created
by
them.
So
then
you
can
actually
control
the
lights
and
stuff
through
your
coat.
B
That's
actually
quite
fun,
but
then
yeah.
So
what
we
gonna
do
now
is
to
show
you
a
few
pieces
of
like
basically
it
features
so
we're
going
to
be
showing
off
those
contents
of
those
little
boxes
that
you
saw
earlier,
like
our
dependency,
injection
and
stuff,
like
that,
some
of
them
will
be
very
obvious,
and
some
of
them
will
be
a
bit
more
kind
of
oh.
This
is
cool.
This
is
this
is
something
that
I
want
to
use
in
my
project.
Hopefully,.
C
So,
let's
start
from
the
beginning:
I
guess
yeah
the
most
obvious
I
guess
obvious,
most
obvious
CVI
dependence
injection
it's
exactly
the
same
as
Java
Tecate
II,
but
that's
also
principle
in
Microsoft
micro
profile
like
wherever
there
is
a
standard
from
jar
from
the
cautery,
that's
suitable.
We
just
reuse
that,
instead
of
making
it
over
and
I
get
over
again.
So
it's
the
same
CDI,
okay,.
C
B
Guess
would
be
to
mention
just
well.
I
mean
surprise,
surprise,
you
can
actually
generate
Jason
and
you
can
do
like
you
can
define
URLs
and
stuff
for
create
queries,
yeah,
which
is
probably
not
a
very
surprising
thing
that
you
for
a
microservice
buzz.
Well,
I!
Guess
it's
kind
of
Horner
on
an
honorable
mention
kind
of
section
of
this
talk,
yeah.
C
Under
these
free
features
right
flood,
those
like
make
a
profile
1.0.
The
first
person
contained
these
free
features
which
are
all
taken
directly
from
Joey
and
are
identical,
and
it's
you
basically
need
it
everywhere
and
yeah
for
the
record
we
are.
We
are
aware
that
we
using
a
get
request
to
mutation
of
the
other
states,
so
we're
gonna
fix
that
later
on.
B
I
mean
still
I
mean
if
you
haven't
used
that
or
you
don't
have
an
experience
with
this
kind
of
thing.
This
is
how
you
define
a
URL
to
to
talk
to
things
so
at
path
that
would
be
this
and
and
consumes
and
produces,
is
basically
where
you
define
a
mime
type
of
whatever
it's
going
to
be
taking
in
as
a
parameter
or
returning
back
so
produces.
So
the
thing
will
produce
adjacent,
but
consumes
does
it
expects
some
kind
of
text
payload
or
something
like
that?
Coming
in
yeah,
yep.
C
B
More
complex
than
the
logic,
so
this
is
the
gateway
service.
This
is
the
kind
of
the
brains
of
the
whole
thing
right,
so
that
one
takes
in
the
inputs
from
all
the
other
services,
and
then
it
will
also
tell
us
what
to
do
to
the
life
in
control
server.
So
it
takes
lots
of
things
in
calculates.
It
somehow
and
returns
a
value
to
the
light
so
say
well,
set
the
lights
to
whatever.
C
A
A
B
Properties,
files
and
stuff
like
that,
and
then
you
have
like
the
whole
hierarchy
of
the
property
files,
really
that
will
kind
of
override
each
other,
but
then
you
can
also
do
you
can
also
provide.
The
cool
thing
is
that
you
can
actually
provide
the
default
values
and
stuff
in
the
code.
That's
like
in
case
everything
else
fails
and
all
the
other
properties,
files
and
stuff
are
gone.
You
still
have
something
like
the
last
kind
of
frontier,
with
all
the
default
values
and
stuff
like
that
that
you
can
provide.
C
C
B
And
stuff
like
that,
because
normally
you
would
want
to
do
that
when,
like,
for
example,
you
have
a
you,
have
different
values
for
different
environments,
so
your
production
environment,
we
have
one
value
and
the
development
environment
would
have
another
one
and
test
one
would
have
another
one
and
so
on
so
on.
So
that's
where
you
this
is
where
it's.
It
becomes
really
useful.
So
you
can
actually
override
those
things.
Yeah
a
bit
more
advanced
thing.
I
guess
is
the
next
one
is
where
you
can
do.
C
B
If
you're
running
it
in
a
container
or
whatever
and
using
I,
don't
know,
say
kubernetes
or
something
like
that
to
to
to
control
it.
So
this
is
one
of
the
things
that
we'll
be
using
to
realize.
If
my
service
is
healthy
or
not,
and
if
it's
not
healthy
I'll,
can
you
kill
it
off
and
start
another
container
with
another
one
and
see
if
that
one
is
better,
and
so
all
these
kind
of
pods
and
things
that
will
cover
knives
will
do
it.
B
We
kind
of
try
to
rearrange
things
and
turn
things
off
and
on
and
see
if
things
are
working
and
stable
and
how
many
replicas
it
would
have,
and
all
those
kind
of
things
would
be
partially
based
on
this
health
checks
right
and
the
cool
thing
is
that
you
can
actually
control
those
with
code,
but
we'll
get
back
to
that
later.
So
yeah.
C
B
And
then,
if
you
go
to
the
next,
one
probably
is
the
most
interesting
for
like
yeah.
So
this
is
how
we
define
it.
So
this
is
oversimplified,
obviously
to
fit
a
little
slide
and
to
make
it
very,
very
easy
to
you
to
grasp
in
a
way.
But
so
ok
can
we
connect
to
this
thing.
Yes,
we
can
okay,
fine,
then
it'll
just
return
it's
up
and
we're
happy,
and
if
not,
then
we
do
it's
down
and
down
the
build
and
it
will
kind
of
say:
well,
it's
not
really
up.
Yeah.
A
C
A
C
A
C
A
C
We
might
be
interesting,
metrics,
here's
just
a
small
snippet
from
like
those
kind
of
metrics
you
get
out
of
the
box
just
out
of
the
box
from
micro
profile.
You
get
metrics
on
things
like
how
many
threads
on
use.
How
many
CPU
cores
are
used
memory
footprint,
all
those
things
like
infrastructure
general
generate
things
from
a
computer
and.
B
I
mean
I,
remember
when
it,
when
I
was
working
a
few
years
back
on
it,
like
very
quite
big
enterprise
piece
of
software,
with
lots
of
lines
of
code
and
I
was
like
closer
to
half
a
million
lines
of
Java
code.
We
ended
up
spending
quite
some
time
and
quite
a
bit
of
code
implementing
those
metrics
and
just
putting
them
in
in
the
console
of
the
application,
server
running
and
stuff
like
that,
exposing
it
through
whatever-
and
here
you
just
do
at
something-
and
it
just
magically
appears
there.
It's
pretty
cool
and
there
will
be.
C
B
What
happens?
Is
that
probably
worth
mentioning
just
in
case
you
wonder,
I
mean
what
is
it
happen.
Is
that
whenever
you
do
all
those
things
Michael
Provo
would
Adam
automatically
put
it
on
to
it
will
generate
the
URL
fear,
so
anything
whatever
you're,
using
as
your
kind
of
starter
URL.
So
if
you're
running
locally,
it
will
be
like
localhost
8080
or
something
like
that,
but
then
slash
metrics.
This
is
where
it's
gonna
expose
all
of
those
things
in
in
a
ways
human
readable,
but
it's
actually
also
much
you
readable
anyway,
as
well.
Yeah.
C
You
never
saw
so
worth
mentioning
cuz
if
some
of
you
listening
out,
there
are
used
to
Prometheus
yeah
you'd
recognize
the
format
I
guess,
because
this
is
the
Prometheus
format,
so
you
can
punch
this
connect,
these
things
on
to
Griffin
or
whatever
tools
you
typically
use
yeah
they're
not
doing
anything
more
now.
That's.
B
Pretty
cool
yeah
yeah
and
then
it
will
also
create
actually
are
under
metrics.
It
will
also
create
different,
like
slash
application,
slash
different
things
that
will
be
defined
by
there
are
a
few
good,
pretty
good
posts
around
about
that
at
out
there
written
by
told
me
people
I,
think,
buy
buy,
buy,
buy
yeah.
Yet
so
it's
it's
quite
a
bit
quite
a
few
sory.
B
If
you
google,
for
that,
you
will
find
quite
a
well
explanation
of
how
metrics
are
done
and
stuff
like
that,
but
I'll
I'll
show
you
also
maybe
at
the
end,
if
we
have
time
I'll
show
you
also
another
example,
but
it's
it's
pretty
easy
and
who
just
kind
of
looks
like
a
magic
at
some
point.
In
a
beginning,
okay,.
C
We're
heading
on
to
for
torrents
and
on
this
slide
you
can
see
quite
simple
programs
at
once.
So
here
we
have
a
rest,
endpoint
get
weather,
so
this
is
a
Vista
from
The
Weather
Service
and
they
will
say
that
okay
retry
max
retry
stews,
like
it,
saves
the
first
time.
Try
again
or
you
can
you
know
you
can
skip
the
parameters
which
it
would
default
to
time.
I
think
the
default
value,
but
get
specify
yourself.
How
many
times
are
you
interested
in
retrying.
A
C
B
So
you
might
want
to
add
a
circuit
breaker
that
will
kind
of
at
some
point
whenever
it's
not
getting
hold
of
that
service
will
just
kind
of
stop
nagging
and
just
do
something
else
and
we'll
get
back
to
do
something
else.
Part
in
it.
Just
a
few
slides
so
just
bear
with
us
for
a
few
more
seconds.
Yeah.
C
B
B
So
then,
if
that
one
fails
or
if
it's
too
busy
or
I
don't
know
anything
happens,
you
can
actually
add
a
fallback
method
that,
if
that
one
fails,
go
and
use
the
other
call.
The
other
method
and
and
the
other
method
might
do
things
like
defaulting
to
some
kind
of
specific
value
or
trying
another
Redis
service.
That
is
less
preferred,
but
it's
still
there
you
can
do
whatever
you
want
really
yeah.
C
And
I
think
you
can
chain
them
as
well
yeah,
but
one
thing
tomorrow
to
mention
here
is
a
thing:
we've
spent
a
couple
of
hours.
Debugging.
Is
that
this
method,
the
fallback
method,
must
have
the
same
signature
as
the
as
the
method
you're
imitating,
for
you
get
strange
exceptions,
it
kind
of
makes
sense,
but
it's.
B
C
Alright
yeah
yeah,
so
it
up
a
little
bit
a
rest
client,
so
you're
typically
doing
rest
calls.
Nowadays,
most
people
are,
and
in
micro
profile
you
can
say
that
okay,
so
this
is
in
the
Gateway.
This
is
where
the
Gateway
is
trying
to
talk
with
the
facade
like
the
light
control
service.
Sorry
in
the
Gateway,
we
create
the
interface
with
the
same
signature
as
I
saw
on
the
slides
earlier
for
methods
that
are
exposed
on
the
under
lights
like
control
service.
So
this
is
like
saying
yeah,
and
this
is
going.
C
This
is
actually
going
over
over
rest.
It
is
because
so
you
could
say
that
these
annotations
are
registered
rest
client
and
dependent,
dependent,
I,
think
that's
not
necessary
anymore,
but
the
register
rest
client
thing
to
tell
America
profile
out.
Okay.
This
is
actually
arrest,
arrest,
endpoint.
So,
even
though
you
call
this
method
as
it
as,
if
it
were
on
your
same
application,
it
really
is
a
rest
call
under
the
hood.
A
C
This
is
you
define
the
interface
and
then
you
that
in
it
in
the
term
code,
we
are
going
to
call
the
endpoint,
you
inject
it
an
annotated
with
address
client
and
then
you
do
some
black
magic
in
a
config
file.
So
we
mean
like
the
entire
path
to
the
class
or
the
class
for
this
to
this
interface
and
the
nude
stretch.
Mp
rest
rest,
URL,
and
then
you
add
the
action
URL
to
the
rest.
You
are
going
to
walk.
C
Yeah
this
is
pretty.
This
is
pretty
neat
in
your
code,
but
yeah
the
Dragons,
and
by
that
we
mean
we
spent
quite
some
time
during
the
initial
development
figuring
out.
Why?
Why
isn't
it
just
working?
Why
do
I
get
a
strange
error
and
so
on?
And
then
we
well,
we
figure
out
that
okay,
this
is
so
hard
to
do
in
I.
Think
what's
one
day
and
we
only
west
of
fontelle
and
the
well
we're
having
so
much
trouble
with
this.
B
B
That's
where
you
can
match
where
you
can
actually
just
switch
an
implementation
and
see
if
that
one
helps
and
if
it
works
and
stuff
like
that,
it's
supposed
to
be
portable
between
those
I
mean
some
implementations
might
be
a
little
bit
behind,
and
some
of
them
will
be
on
like
totally
bleeding
edge
and
implementing
all
the
new
news
to
the
coolest
features,
but
in
most
of
the
cases,
will
be
pretty
well
in
in
pretty
well
sync,
I,
say
I.
Think.
C
Yeah-
and
it's
also
worth
mentioning
that
it's
yeah
it's
getting
more
mature
now,
so
the
bug
way
like
talking
about
it's
probably
fixed
like
long
time
ago,
but
yeah.
We
have
some
handsome
issues
and
you
also
yeah,
like
these
things,
aren't
necessarily
obvious.
We
need
to
look
it
out.
Look
it
up
or
know
about
it.
Yeah.
B
And
it's
pretty
much
I
would
say
it's
quite
mature.
Now
I
would
say
it's
pretty
stable
and
I
would
say
it
actually
works
pretty
nice
nicely,
and
then
you
have
also
the
thing
that
we
don't
really
mention
in
the
slides.
But
then
you
have
also
this
a
starter
page
which
will
help
you
to
get
started
with
all
the
features
and
stuff
you
just
go
and
pick
once
that's
your
once.
We.
C
So
yeah
microphone
in
timer
weather
services,
so
we're
so
far
been
talking
a
lot
about
Gateway
service
and
electrical
control
service.
But
what
about
the
time
service
in
the
weather
service
yeah?
Well
much
of
the
same
like
yeah.
It's
the
time
service
gets
you
the
current
time
and
leaders
about
that.
So
it's
not
that
complex,
you
guess
and
you're
right,
but
what
we
do
have
is
using
open
API,
which
is
documentation
of
your
on
the
API
by
adding
annotations-
and
this
is
yeah.
Swagger
is
the
most
popular,
well-known,
implementational
open
api.
So.
B
B
Actually
again,
it
would
like
Rock
Island,
developing
some
stuff,
small
micro
services
and
stuff
and
I
was
like
how
was
this
wits,
my
what's
the
URL
to
that
thing,
I
created
and
stuff,
and
then
you
just
go
to
open
API
and
it
will
list
you
all
the
things
in
you
know
oh
yeah.
This
is
how
I
was
thinking,
and
this
is
what
I
was
doing.
So
it's
pretty
nice
way
of
doing
stuff,
and
then
you
can
also
well
since
you're,
using
it
for
way
of
documenting
things.
B
C
And
it's
quite
easy
to
do
yourself
as
well,
because
here's
what
we
do
if
we
produced
it,
the
text
you
saw
the
previous
slide
to
achievement
and
annotation
like
at
operation-
and
you
had
the
summary
and
then
you
can
add
the
description.
The
format
is
reptiles
like
yeah,
so
that
the
summer
in
description
that
those
texts
we
wrote,
you
can
see
here
like
if
there's
some
additional
information
with
yeah
yeah
I
think
they
are
optional.
At
least
this
friction
is
yeah
yeah.
It
is
great
forward
all
right
demo.
So
it's
demo
time.
C
B
B
C
C
A
C
B
C
A
A
C
C
A
B
We
did
not
sacrifice
enough
rubber
ducks
for
the
demo.
Yes,
that's
the
that's
what
they
did.
They
sacrifice
currency
of
that
or
whatever.
No,
we
got
a
question
in
the
meantime
how
Java,
EE
8
is
different
from
micro
profile.
Yes,.
B
C
But
like
you
can
also
say
that
well
most
application
service
combined
Java
EE
8
with
micro
profiles,
so
you
can
use
features
from
Java,
EE
8
or
you
can
use
just
from
micro
profile
depending
what
you
need
in
your
in
your
context.
So
yeah
it
depends
what
your
depends
alright.
So
here
we
are
with
that
demo,
no
I
think
it's
working.
B
A
C
Yeah,
ok,
then
you
can
remember
those
lights.
Remember
we
have
some
blacks
here
and
you
can
see
that
the
lights
are
pretty
dim
down
and
on
this
simple
view,
mu,
J
as
front
end
so
attack,
it's
just
a
simple
front,
end
doing
rest
calls
to
the
back
games
and
the
back
ends.
Oh
sorry,
I
forgot
that
I
don't
share
my
screen
anymore,
but
you
can.
A
B
My
hand
I,
guess
yeah,
but
then
it's
some
kind
of
issue
with
the
camera
being
and
the
lights
being
too
bright
and
the
difference
of
the
rice
and
stuff.
The
other
thing
is
that
those
two
are
green
and
this
is
white,
but
the
reason
for
that
is
that
this
one
is
like
only
shades
of
white
and
those
are
coloured
lamps.
So.
C
B
There
is
a
lot
of
my
calls
and
Jason
and
rest
APRs
and
everything
happening
in
the
background,
but
then,
in
reality,
what
you
actually
see
is
the
lights,
changing
slightly
color
and
changing
a
bit
of
so
what
we
did
for
the
demo.
So
you
can
actually
see
things
happening.
We
added
a
bit
of
randomness
there,
so
the
the
brightness
and
stuff
like
that
will
change
for
each
time.
We
refresh
things
just
because
we
have
might
have
there
is
there.
Are
people
saying
that
they
can
see
video
and
up
but
I'm,
not
sure?
B
C
C
So,
yes,
it
looks
like
the
break,
so
no,
no,
it
just
dimmed
them
down
because
they
are
a
bit
disturbing
to
us
and
bright
lights
right
out
of
us
yeah.
But
perhaps
we
should
so
gonna
show
some
code
as
well.
I
think,
like
the
time
is
running
fast.
This
I
think
this
is
so
fun
that
we,
like
speaking
for
way
too
long.
Sorry
about
that.
C
B
Implementation
is
most
suited
for
T
domain
and
in
general,
which
are
most
reliable
for
production
usage,
I
would
say,
they're
quite
stable.
Most
of
them
I
mean
yeah.
A
C
A
C
Yeah,
so
it
like
there
are
different
strengths
and
weaknesses,
so
but
I'd
say
yeah
I
tend
to
stick
to
either
Liberty
or
quark,
as
if
like
I,
had
to
choose
one,
but
there's
no
right,
sorry
to
say
it
depends,
but
it's
Syria,
looser,
but
back
to
workers.
You
can
see
that
if
here
I
started
my
services
again
after
stopping
tapping
them,
I
guess
a
techie
compose
up
and
you
can
see
time
service.
This
one
is
the
one
making
workers
so
time
service.
Well
here
it
starts
down
here
at
this
line.
C
C
B
Should
we
have
a
few
minutes
a.
C
Few
minutes
yeah,
okay,
so
yeah
we're
showing
a
screen
good
all
right.
So
what
have
we
learned?
Probably
I
had
lots
of
fun,
but
we
also
learned
some
things
that
we
hope
that
we
can
share
with
you
so
I
quondam.
He
said
yeah
mutual
things
keep
on
coming,
such
as
yeah.
The
things
were
struggling
will
struggle
with
initially
have
been
sorted
out
since
then,
and
court
case
has
been
current
has
has
come
since
we
started,
but
also
things
like
the
micro
profile
started
like
you
written
talked
about.
A
C
You
could
say
that
oh
I
want
to
use
microphone
file
version
so
and
so
and
yeah
what
works
and
so
I
want
examples
of
these
thanks
and
I
want
to
use
PRM
a
micro
organ
blows
or
whatever.
And
then,
when
you
push
the
down
the
button,
you
get
home
fire
than
the
project
set
up
on
all
this.
So
it's
easy
to
get
started
this
one
came
not
that
long
ago.
C
Yeah
another
thing
we
learned
is
that
yogini,
you
income
need
it,
and
so
we
started
out
by
using
Lombok
a
lot
across
our
application.
Lombok
is
a
tool
that
generates
much
of
the
boilerplate
code
for
you
like
by
adding
the
annotation
a
scanner
you
you
don't
have
to
write
your
own
getters
and
it's
a
murder
would
require
arguments
constructor.
There
are
lots
of
these
and
we
found
out.
We
figured
out
once
that
right
before
doing
a
presentation,
yeah.
B
C
Was
a
boy
it
was
a
problem.
Maven
yeah,
it's
not
really
related
to
the
book,
but
we
thought
it
was
real
but
because
well
we
didn't
understand
the
error,
so
we
just
try
to
figure
one
in
to
start
and
before
you
figured
out
that
okay,
so
this
introduces
a
lot
circle.
That's
the
complexity!
Here
we
don't
really
need
it's
like
not
that
much
boilerplate
anyway,
so
we
figure
that
we
we
didn't
need
it
yeah
and
your
mileage
may
of
course,
vary
mm-hmm.
A
C
And
the
same,
we
did
some
quite
fancy:
advanced
logging
advanced
in
quotation
marks,
but
yeah
logging,
so
we
using
block
per
day
and
we
injecting
it
with
the
slider.
You
can
see
like
a
hacked
to
inject
the
logger
in
your
code,
how
we
had
life
lots
of
lucky
things
spread
or
in
their
code.
Let
me
figure
it
out.
Oh
well,
we
ain't
gonna!
Need
it
because
we
aren't.
We
are
anything's
in
containers
we're
on
the
printing
things
to
to
the
console.
So
why
don't
we
use
system.out.print
line
and
of
course
we
are
enterprise
developers.
C
So
we
we
written
an
application,
scope,
printer
printer
print
line
so
for
your
tests.
You
can
don't
do
it
and
take
like
that
every
rather,
this
was
more
than
good
enough
yeah
and
it
also
fits
well
with
a
12
factor
thing
and
more
the
cloud
way
of
thinking.
So
you
can
do
this
uh-huh
yeah
absolutely
far.
Well,.
B
A
A
C
A
C
It's
an
original
Red
Hat
initiative,
so
it's
been
heavily
used
in
Quercus,
infantile
and
in
world
fly,
which
are
off
application
service
from
a
net
and
also
it's
telling
it
starting
to
get
used
in
open
Liberty
by
IBM
yeah.
So
who
knows
what
happens,
but
it
yeah
yeah,
and
this
one
came
just
yeah
a
year
ago
or
so.
Yeah.
A
B
Like
a
few
honorable
mentions
before
we
get,
we
got,
we
will
finish
it
off,
so
really
quick.
This
is
not
really
a.
The
next
slide
would
be
like
nothing
really
Michael
Provo
related,
but
just
worth
mentioning
is
that
whenever
you
have
jars
lying
around
that
you
don't
really
have
control
over
and
you
cannot
just
put
on
a
public
repository
and
stuff,
you
can
always
create
a
local
maven
repository.
So
in
our
case
that
would
have
been
those
jars
that
are
construed
to
control
the
delights.
B
C
B
C
A
B
Will
be
get
they
will
get
duplicated
in
us
sometime
in
the
future,
so
do
not
use
the
jars
that
they're
that
are
provided,
and
probably
you
want
really.
It
won't
be
that
easy
to
get
hold
of
them
anyway.
Use
the
REST
API
instead
yeah
yeah
freakin,
an
API
is
also
things
that
we
can
have
done,
and
this
is
kind
of
de
faking,
an
endpoint
that
what
I
would
usually
end
up
doing
so
it's
it's.
C
B
C
B
B
C
So
it
just
a
quick
reminder:
these
are
standards
from
micro,
profile,
3.0
and
you've.
Seen
a
few
thing
most
lambie
haven't
seen
us
using
me,
the
JV
j
JV
t
just
I
I
guess
it's
just
for
propagation.
No,
you
haven't
seen
it's
been
using
upper
Tracey,
that's
obvious!
The
future
works,
the
keys,
open,
tracing
and
use
just
gives
me
propagation.
So.
B
C
B
We
kind
of
try
to
to
keep
it
up
and
update
it
to
new
versions
and
stuff,
like
that.
The
other
thing
that's
really
wanted
to
do
is
to,
if
you
add,
a
bit
more
of
fun
stuff
with
kubernetes
and
open
shift
and
stuff
like
that,
and
also
maybe
even
an
e
still
to
see
if
we
can
do
like
the
routing
and
all
this
kind
of
api's
and
stuff
like
that.
For
that
can
do
for
us.
C
Yeah-
and
we
can
also
say
that
quickly,
micah
profiles,
if
it's
we're
a
little
bit
open,
if
they
give
me,
doesn't
he
steal
when
all
of
it?
So
it's
a
perfect,
it's
an
easy
fit,
and
there
are
people
out
there
doing
talks
on
this
that
we
do
like
Emily
Jiang,
for
example,
has
specific
talks
on
micro
profile
on
misty,
oh
yeah.
So
if
you're
interested
in
that
code,
yeah
just
search.
C
All
right
everything
we're
talking
about
we've
been
talking
about
at
least
the
code
and
the
documentation
we
created
as
well.
Isn't
it
up?
We
accept
all
requests,
happily
so
yeah
or
if
you
have
questions
or
issues
or
whatever,
and
we
were
happy
to
get
feedback,
yeah,
yeah,
so
where's.
The
link
well
I
think
we
also
publish
the
slides
afterwards.
I
guess
yeah.
B
B
So
the
first
one
is
they
put
a
link
from
this
previous
slide
so
and
then
the
second
one
is
this:
first
part
of
a
kind
of
serious
of
posts
that
I've
been
I
could
have
started
on
writing
to
explain
the
basics
and
getting
started
of
marker
profile
and
stuff
like
that,
so
that
one
is
more
about
defining
end
points
and
stuff
like
that
and
how
you
can
define
different
mind
types
and
return,
different
things
and
generate
files,
and
things
like
that.
I'm.
C
Like
the
lower
one,
this
well
I've
gotta
get
getting
questions
from
here
and
they're
like
what
is
Michael
profile,
what
I
use
it
for?
No,
why
and
looks
like
a
brief
overview
and
it's
gonna
be
kind
of
a
summary
I.
What
we'll
be
talking
about
for
an
hour
now
summarizing
a
short
blog
posts,
but
of
course
you
can
also
go
to
micro
profile,
dot,
IO,
the
effects
/
trinket
in
more
information.
B
So
this
is
us
if
I
tried
to
follow
up
with
the
questions
and
stuff
in
the
chat,
but
I
might
have
oversea
like
overlooked
some,
if
I
did
I'm
really
sorry
about
that
and
just
feel
free
to
type
them
over
again
or
copy/paste
them
or
ping
us
on
Twitter.
So
this
is
us
two
of
us
on
Twitter
or
well
in
any
other
kind
of
media.
To
get
hold
of
us
is
fine,
yeah,
that's
I!
Guess
that
would
be
it
in
case
you
have
more
questions
or
anything
either.