►
Description
Join members from the ASP.NET teams for our community standup covering great community contributions for ASP.NET, ASP.NET Core, and more.
Community Links: https://www.theurlist.com/aspnet-community-standup-2020-03-24
A
A
A
B
This
reminds
me
of
with
the
four
boxes
and
people
with
headphones
and
stuff
is
like
like,
like
Power
Rangers
or
like
one
of
those
shows
where
you
got
the
people
and
like
form
X,
and
then
they
show
them
all
like
having
a
conversation.
This
is
just
that
like
this
is
just
that
the
remote
work
is
just
Power
Rangers,
it's
wonderful,
yep,
okay,.
A
A
A
Well,
so
they
hear
us
I'm
gonna
rip
through
the
e
community
stand-up
links,
because
we've
got
a
bunch
and
I
want
to
get
through
them
so
that
we
can
let
you
do
your
thing.
So
let
me
check
I'm
singing
things
all
over
the
place
here.
I
have
they
say,
good
audio,
hooray,
that's
exciting,
to
hear
alright
I'm
switching
monitors
I'm,
seeing
if
it
works
I
mean?
Oh,
no,
sorry,
that's
complicated,
I'm
not
used
to
both
hosting
the
call
and
sharing
the
links,
and
it's
just
too
much
for
my
little
brain
okay.
A
B
A
Now,
if
you
want
to,
you
can
go
to
live
dotnet
and
that'll,
that's
the
YouTube
thing
or
you
can
go
to
twitch.tv,
slash
visual
studio.
Those
will
show
to
you
it's
it's
complicated
if
I
share
it
on
the
teams,
call
it's
not
ideal
for
sharing
on
OBS
and
stuff.
So
so
then
he
also
goes
into
Razer
components
and
compares
Razer
components
and
blazer,
and
he
was
looking
at
a
use
case.
A
It
sounds
like
kind
of
integrating
Razer
components
in
with
blazer
inside
a
server-side
web
app
like
a
razor
pages
app
and
ran
into
some
some
I
guess
some
places
where
it
wasn't
ideal
for
exactly
what
he
was
trying
to
do.
So
that's
not
an
indictment
of
Razer
components
at
all.
It
was
just
the
fit
of
what
he
was
trying
to
do
so,
for
instance,
things
like
partials
and
tag
helpers.
Not
not.
You
know
functioning
inside,
like
kind
of
roaring
them
all
together
in
one
big
thing.
So.
C
A
Yeah
so
picking
the
right
model,
I'm
not
trying
to
throw
it
all
together.
So
there's
a
lot
more
here,
but
he
basically
that's
kind
of
what
what
he
kind
of
settles
on
his
follow-up
posts,
though,
is
is
kind
of
a
thing
of
here's.
Here
is
a
good
fit'
example
and
he
shows
a
very
common
Ajax
sample,
which
is
cascading
dropdowns,
which
we've
all
done
a
thousand
times.
But
this
is
an
example
where
there's
you
know
multiple
things:
cascading
dropdowns,
where
you
select
a
country
and
then
that's
going
to
select
a
province
or
state.
A
So
then
he
kind
of
goes
through
and
walks
through
that
sample
and
reminder.
You
know
you're
in
one
universe
or
the
other
you're,
not
you're,
not
gonna,
be
happy
if
you're
kinda,
just
throwing
it
all
together
right
so
and
Damian
were
ripping
through
these
this
week,
I've
got
a
bunch
of
links,
but
I
want
to
give
the
team
time
to
time
to
show
their
stuff
off.
So
it's
right
through
okay,
so
on
security,
headers
one
from
Scott
Hanselman
he
had
his
site
all
set
up
with
the
security
headers
had
an
a
beautifully.
C
D
A
E
So
what
happened
was
I
had
done
years
ago,
another
blog
post
about
getting
an
a
on
security
headers
and
by
by
adding
all
that
stuff
to
web
config.
When
you
are
behind
I
is
I
is
goes
and
does
all
of
that
because
it
lists
since
it
lives
in
system
dot
web
server,
then
I
moved
to
Linux
to
save
some
money
and
forgot
never
bother.
I
have
no
tests
that
tell
me
that
I
went
from
an
a
to
an
F
I
moved
everything
into
a
container
and
everything
broke.
C
A
B
E
Yes,
okay,
so
this
is
my
main
camera.
We've
talked
about
this
before
I'm
prepping
for
build
we're
trying
to
make
sure
that
everyone
looks
good
for
build.
So
here's
one
camera,
here's
camera
two
on
another
webcam,
here's
picture
and
picture.
Oh
and
here
is
the
ever-important.
Please
work,
please
work!
E
E
E
C
E
B
B
A
C
E
E
A
Okay,
cool
rippin
through
okay,
so
so
yeah,
so
good,
post
from
Scott
up
and
coming
down
that
blogger
Winkies
at
a
promising
future
in
blogging
industry,
okay,
also
on
the
security
headers.
Just
by
coincidence,
here's
a
blog
post
from
the
ELMO
folks
by
Thomas
writing
up
security
headers
like
what
those
headers
actually
mean
so
I
felt
like
that
was
kind
of
cool
that
they're.
Both
we
get
them
both
together,
one
one
on
how
to
set
them
up
and
the
other
on
what
the
heck
you
actually
did
set
up.
I
have.
A
That's
cool
all
right.
This
is
interesting,
the
J
project.
This
is
a
it's
an
SSO
kind
of
administrator,
so
you
get
like
an
admin
dashboard
working
with
asp,
net
identity
and
also
with
so
it's
you
know
an
admin
UI
for
your
identity,
I'm.
So
like
user
management
and
all
that
kind
of
stuff,
so
running
on
top
of
asp
net
identity,
I
thought
that
was
kind
of
I
mean
cool,
all
right
event,
aggregator
for
blazer,
so
McHale
posting
about
the
2o
release
and
the
idea
here
with
event.
Aggregator
is
prostate
component
to
component
communication.
A
So
it's
an
abstract
like
it's
in
there
in
direct
communication,
I've
seen
people
do
different
things
when
you've
got
complicated
things
when
you've
got
multiple
components
connecting,
but
this
seems
to
be
a
popular
pattern.
So
it's
that
Alright,
post,
comparing
open
to
API,
slash,
swagger
graph
queue
and
GRP,
see
and
I
was
happy
to
see.
This
isn't
like
this
one
sucks
this
one
rules,
it's
more
just
kind
of
a
like
here's,
the
fit
here's
where
they
fit
together.
A
Here's
the
different
kind
of
no
good
things
and
bad
things
of
you
about
each
one
speed
speaking
a
graph
QL,
so
Rob
was
on
the
visual
studio
toolbox,
show
giving
an
introduction,
so
you
know
high-level,
introduction
to
gr
PC.
This
is
a
nice.
You
know
25
minute
kind
of
review,
open
silver
I,
don't
know
what
to
say
about
this,
but
it
is
interesting
to
at
least
mention
it.
So
the
idea
here
is
a
running
on
top
of
blazer
and
trying
to
implement
the
Silverlight
API.
So
this
is
I.
A
A
Okay,
cool!
This
is
cool.
This
was
passed
over
to
me
by
Shane
this.
So
this
is
a
free
workshop
on
dotnet,
core
graph
QL
and
server
list
from
Chris.
So
this
is
a
whole
kind
of
nice.
You
know
walkthrough
of
all
the
stuff
leading
a
cool.
You
know
like
website
where
you
can
kind
of
click
through,
and
you
know
beautiful,
Docs
and
everything
there.
It's
nice
ahead,
continuing
a
serious
this
one
on
logging,
with
asp
net
core.
So
just
kind
of
you
know
she
had
doing
his
thing.
A
This
is
the
grab
a
few
cups
of
coffee
post
of
the
week,
as
you
can
see
from
the
very
long
slider
bar
but
nice,
showing
how
both
up
to
a
few
things
showing
the
both
the
Azure
configuration
running
against
the
azure
hosted
Redis
back-end,
but
also
he
shows
some
things.
Several
things
running
through
the
Redis
desktop
managed.
So
you
know
doing
local
development.
A
I've
did
one
project
using
Redis
and
I
really
liked
it.
It
was
the
dotnet
detonate
conf.
We
did
some
back-end
caching
using
Redis
and
it
was
beautiful
and
using
that
library
also
and
then
I
just
figured.
You
know
as
to
hop
over
here's.
Here's
I
was
playing
around
I
was
poking
at
your
docks
and
I
was
quite
impressed.
B
A
B
F
B
B
We
thought
it
was
a
memorable
name
for
a
command-line
tool,
it's
nice
and
short,
but
it
turns
out
to
be
really
good
because
Tai
is
our
tool
for
sort
of
tying
a
bunch
of
microservices
together,
putting
a
bunch
of
services
together
and
that's
really
like
the
role
of
the
tie
on
the
ship.
It's
a
big
rope
that
ties
stuff
to
the
mast
and
raises
and
lowers
the
sail.
So
we
we
inadvertently
chose
a
really
good
name
so
Amy.
Why
don't
you
go
ahead
and
share
your
screen
and
I'll?
G
Okay,
so
before
I
kind
of
jump
into
the
demo,
I
wanted
to
go
over
the
different
parts
of
the
demo
app.
So
we
have
a
voting
sample
application
and
it
contains
five
different
services
and
the
way
that
it
works
is
that
you
will
cast
your
vote
and
that
vote
will
enter
a
Redis
as
an
entry
and
that
entry
will
then
be
taken
in
by
the
worker.
G
Who
will
then
take
that
and
insert
or
either
add
a
new
record
into
Postgres,
as
well
as
write
the
data
via
signal
our
to
the
results
page
and
then
the
results
page
will
then
go
ahead
and
display
those
results.
So
now
I
want
to
see,
go
ahead
and
just
kind
of
show
the
code,
so
this
is
kind
of
the
code
for
the
sample
application.
So
we
have
our
different
projects:
local
projects,
that
we
have
the
results,
page,
the
vote
page
and
as
well
as
the
worker.
G
G
G
So
that's
going
to
execute
and
we're
going
to
see
that
it
built
our
three
projects,
so
you'll
see
the
voting
project,
the
worker
and
the
results
project.
You'll
also
see
that
it's
running
two
containers,
someone
for
Redis
and
one
for
Postgres,
which
are
our
external
dependencies,
and
it
also
creates
this
dashboard,
and
this
dashboard
allows
you
to
be
able
to
see
the
current
status
of
your
services
and
application.
So
we're
gonna
go
over
to
the
dashboard.
So
I
can
show
you
guys
what
that
looks
like.
G
Okay,
so
here's
the
dashboard,
the
dashboard
shows
you
like
I,
said
all
of
the
different
services
that
you're
running
along
with
the
different
bindings,
which
are
just
supports
where
all
of
your
services
are
running
at
as
well
as
logs
for
each
one.
So,
for
example,
we
can
view
the
voting
log
here
so
now,
let's
actually
go
and
cast
our
vote.
G
Gonna
vote
for
cats
awesome,
so
I'm
gonna
cast
my
vote,
so
here
in
the
results
page,
we
should
see
one
big
pie
chart
showing
that
we
have
one
vote
for
our
guests.
So
going
back
to
the
dashboard.
Now
we
can
view
the
worker
log
and
we'll
see
that
the
worker
got
a
new
vote
from
and
be
representative
vote
for
cats,
and
we
can
also
see
that
the
worker
has
written
the
data
to
Postgres,
as
well
as
using
7rr,
to
call
the
other
clients
with
updated
results
which
then
just
displays
here
on
our
results
page.
G
So
this
is
the
pie.
Chart
that
you
guys
see
so
I
also
wanted
to
talk
a
little
bit
about
like
how
a
lot
of
this
stuff
is
working
on
the
backend.
So
we
have
like
this
tie,
tie
to
a
MoU
file
and
here
on
this,
just
gets
generated,
and
it
contains
like
all
of
your
different
projects
on
the
increase
in
my
services.
It
also
contains,
like
your
docker
images
as
well,
and
the
binding
ports
which
we
automatically
choose
for
you.
G
So
that
way,
you
can
avoid
things
like
port
conflicts,
so
this
is
essentially
the
yo
mofos
that
you
can
configure
as
well
to
add
any
additional
projects
or
dependencies
that
you
want
for
your
application.
So
this
is
how
tyrunt
works
and
how
you're
able
to
kind
of
spin
up
and
run
all
of
your
applications
at
once.
So
a
quick.
G
A
B
It's
not
a
dumb
question
so
with
we
have
another
command
tie
in
it.
That
will
generate
one
of
these
files
for
you
and
it's
based
on.
What's
in
your
solution
or
what's
in
your
like
current
directory,
so
we
don't,
we
don't
have
the
ability
to
know
that
you
need
postgrads
or
what
image
Postgres
is
or
how
to
configure
it.
I
kind
of
like
I
kind
of
like
the
idea
that
we'd
have
some
recipes
for
common
stuff
like
that,
but
we
don't
have
anything
that
does
that
today.
G
F
D
F
F
Great,
it
shows
how
tight
I
can
be
run
locally,
but
a
lot
of
times.
You
also
want
to
be
able
to
publish
this
production,
so
this
is
a
fairly
complex
application.
There's
five
services
are
going
on,
so
how
can't
I
actually
use
cirrage
and
do
that
so
I'm
gonna
preface
this
by
a
couple
things
I've
already
done
so
one
thing
that
I've
already
done
is
I've
already
supplied
a
couple
things
necessary
in
order
to
configure
Redis
Louis
Chris
by
default.
When
you
do
at
I
deploy.
F
We
don't
really
know
like
what
the
connection
string
you
should
use
for
for
Postgres
and
write
us
would
be,
and
things
like
that.
So,
if
I
do
get
pods
right
now,
I
should
be
able
to
see
all
the
pods
that
are
currently
running
and
two
of
them
are
gonna.
One
of
them
is
going
to
be
a
post.
Chris
ryan
commute.
Sorry,
one
of
them
is
going
to
be
a
post
cursed
instance,
and
also
I'm
going
to
have
a
secret
for
Redis
as
well
as
post,
poops.
F
Okay,
good
I'm
gonna
have
one
for
Postgres
here
as
well
as
Redis,
and
this
is
just
connections
to
any
information.
That's
that
I
pre-configured
to
make
this
demo
pretty
much
work,
so
I'm
gonna
do
a
command
called
high
deploy
and
I'm
gonna
do
interactive,
because
I
need
to
supply
my
container
registry
that
I
would
like
to
use
what
this
is
going
to
do.
Is
it's
going
to
take
all
of
my
projects
that
are
currently
available?
F
It's
going
to
build
them
is
going
to
create
containers
for
each
of
them
and
it's
going
to
put
art
and
then
it's
going
to
put
them
in
Krita
manifest
kubernetes
manifest
that
will
be
used
to
apply
to
kubernetes.
So
right
here
you
can
see
that
it's
reading
a
bunch
of
project
information,
it's
compiling
and
creating
different
outputs
reach
the
applications,
it's
building,
docker
images
and
it's
pushing
the
docker
images.
So
this
would
be
the
voting
app
that
we
saw
earlier.
It
says:
I've
found
some
detail
or
secrets
for
Redis
and
Postgres
already
available.
F
So
I
have
two
pages
that
are
already
the
URLs
that
are
going
to
be
there.
So
I
should,
as
soon
as
this
is
all
done.
I
should
be
able
to
refresh
the
page
and
it
should
be
available
already.
Another
thing:
I've
already
done
as
well
as
I've
set
up
an
ingress
resource
to
make
it
so
this
application
is
actually
available
to
the
internet.
So
you
just
used
an
engine,
X
ingress
the
mo
file
to
do
that,
and
it
should
be
wrapping
up
right
about
going
manifests
and
it
should
be
good
to
go
right
about
now.
F
Let's
here
there
we
go
cool,
so
it's
already
deployed
and
it's
finishing
up
right
now.
So
if
you
actually
go
to
this
URL
right
now,
I
believe
it
should
be
a
veil
with
everybody,
it's
just
my
IP
and
then,
if
the
I/o
and
then
Sascha
vote,
you
can
vote
whether
or
not
you
want
dogs
or
cats,
then
I.
Also,
okay,
I'm
gonna
vote
for
dogs
is
the
right
answer.
F
C
A
A
F
So,
as
you
can
see,
the
results
are
actually
getting
updated
right
now
with
people
voting.
Some
of
those
dogs
feel
urban
for
cats,
their
answers
dog,
so
you
should
definitely
vote
for
dogs
and
four
cats
are
incorrect,
but
yeah
that's
the
application
running.
So
if
I
actually
do
Q
controller
we'll
get
pods,
you
can
see
the
three
pods
they're
running
right
now,
the
worker,
the
vote
and
the
results
page
if
I
get
logs
for
that
for
a
let's.
Do
the
worker
I'm
just
talking
this
because.
F
F
Bunch
of
vlogs
here
saying
that
we
got
a
boat
from
Redis
because
everything's
getting
into
Redis,
and
then
we
are
putting
the
results
in
Postgres
and
then
using
single
are
to
broadcast
it
as
well.
And
then
we
are
that's
everything
yeah.
That
is
pretty
much
everything
with
deploy.
So,
as
you
can
see,
we
have
an
application
running
though
we
deployed
successfully
to
Cuba
nineties
cool.
B
F
So
the
beauty
of
ty
deploy
with
interactive
mode.
It
allows
you
to
configure
secrets
on
the
fly
if
you
didn't
have
them
there
before
so,
let's
say
we
were
doing
the
ty
deploy
and
the
Redis
secret.
Wasn't
there,
but
I
needed
it
to
make
the
application
actually
work.
I
would
notice
this
when
you're
actually
deploying
and
you
would
be
able
to
fill
in
the
connection
string
when
you're
doing
it
interactive,
deploy.
F
If
you
weren't
doing
an
interactive,
deploy,
I
believe
it
would
just
fail
to
deploy
saying
you
need
the
secret
to
be
configured
and
that's
very
useful
because,
like
these
secrets
are
usually
kind
of
like
a
one-time
thing,
you
usually
don't
need
to
do
it
multiple
times
like
you,
don't
need
to
deploy
secrets
on
every
deployment.
It's
just
like
a
one-time
configuration.
So
that's
a
nice
option
of
trying
to
ploy
that
allows
you
to.
A
F
A
F
F
So,
for
example,
when
we
deploy
we
effectively
just
deploy
a
course
when
we
create
a
kubernetes
animal
file,
we
effectively
just
have
a
deployment
and
a
service
for
each
of
your
projects,
and
then
we
do
some
necessary
things
to
make
it.
So
these
these
services
can
communicate
with
each
other
by
setting
up
environment
variables
appropriately
Ryan
I,
don't
know
if
you
want
to
talk
about
service
discovery
at
all,
but
I
think
it's
yeah.
A
B
I
think
the
primary
way
that
I
would
say
that
this
is
different
is
right-click
right
to
right
click.
Deploy
is,
first
of
all,
as
a
per
project
is
a
per
project
gesture.
The
second
thing
that
I,
the
second
thing
that
I'd
say
about
it
is
ty,
is
meant
to
work
for
local
development
for
like
cheap
experiments
like
I've
got
this
code
and
I
just
want
to
publish
it
somewhere,
and
it's
also
meant
to
be
used
as
part
of
a
CI
CD
pipeline.
B
Today
we
use
the
versioning
rules
that
are
set
up
in
your
Donette
projects
for
out
a
version
stuff,
we'll
probably
do
something
slightly
better
in
the
future,
but,
like
you,
actually
get
things
deployed
to
kubernetes
with
proper
versioning
and
stuff
like
that,
so
it's
like
the
tool
is
meant
to
take
you
from
I
have
an
idea
and
I
want
to
hack
it
to
I,
want
to
deploy
this
somewhere
and
see.
If
it
actually
runs,
I
now
have
an
actual
CI
pipeline.
That's
working
right!
B
F
B
Yeah,
so
some
of
the
things
that
tyre
is
doing
here-
that's
pretty
pretty
interesting,
is
ty,
has
intimate
knowledge
of
net
there's
a
lot
of
dotnet
going
on
inside
the
tool,
and
so
like
what
you
don't
see
here.
There
are
no
docker
files.
There
are
no
helm
charts,
there
are
no
kubernetes
manifests
like
we
haven't
done
any
of
these
things.
There's
another
demo
that
we
like
to
do
for
ty,
where
we
have
a
front
end
and
a
back
end.
B
B
The
other
thing
that
I
would
mention
I
think
it
was
Giovanni's
question-
is
that
we
haven't
gotten
to
yet
is
that
ty
is
a
very
experimental
like
incubation
sort
of
project
that
the
team
is
working
on
right
now,
so
is
is
the
destiny
of
the
things
that
you're
seeing
here,
that
they
show
up
in
Visual
Studio,
and
you
can
right-click
to
do
the
same
things
that
we're
showing
that's
a
possible.
That's
a
possible
endpoint
for
what
we're
doing
so.
B
This
is
a
little
bit
more
of
an
incubation
and
we're
we're
sort
of
feeding
ideas
that
are
were
we're
growing
ideas
that
are
going
to
drive
the
state
of
the
art
forward
in
the
future.
So
I
would
expect
to
see
a
lot
of
more
innovative
kind
of
radical,
takes
on
stuff
right
now
and
then
see
those
kinds
of
things
either
incorporated
into
other
projects
in
the
future
or
ty
continues
on
as
a
project
who
knows.
Okay,.
B
B
We
haven't
so
this
is
one
of
the
things
that
we're
really
curious
about
and
we're
so
curious
about
it
that
it's
part
of
our
it's
part
of
our
like
feedback
issue,
template
I,
would
say
right
now.
We
are
optimized
for
the
case
where
all
your
code
builds
together.
One
of
the
other
problems
that
we're
interested
in
taking
on
and
I'm
I'm
asking
because
I'm
answering
it
now,
because
it's
a
very
good
question.
B
You
know
applications
so
the
idea
being
that
you're
using
Thai
some
other
team
in
your
company
is
using
Thai
and
you
can
run
their
services
locally
by
bringing
in
their
manifest
or
something
like
that,
so
that
idea
hasn't
really
been
fleshed
out,
but
we
know
that
the
need
is
there.
I
think
we're
interested
in
any
feedback
from
anybody
doing
micro
services
development
on
dotnet,
about
what
are
your
biggest
pain
points,
and
what
can
we
do
to
resolve
them?
Well,
yeah.
B
So
let
me
let
me
take
control
then,
and
I'm
gonna
jump
to
our
slides
and
hopefully,
if
there's
hopefully
we'll
probably
be
answering
some
of
these
questions
as
we
continue
through
the
slide,
we're
yeah.
So
we
did
that.
So
talking
a
little
bit
about
local
development,
which
Amy
showed
us,
we
automatically
choose
ports
to
avoid
port
conflicts
when
you're
running
multiple
things.
B
If
you
try,
if
you
today
on
on
Mac
I
think,
is
avoid
cleverly
avoids
this
problem,
but
if
you're
on
Mac
or
you're
you're
running
at
the
command
line
today-
and
you
just
create
two
new
dotnet
projects
and
try
to
run
them
that
will
not
work
because
of
a
port
conflict.
You
need
some
kind
of
Orchestrator
to
make
it
so
that
your
applications
actually
get
assigned
unique
ports
for
local
development
to
work,
and
we
also
have
the
ability
to
run
containers
and
other
processes
for
your
data
dependency.
B
So
whatever
external
thing
is
that
you
want
to
have
as
dependencies
to
make
your
app
work,
we
have
the
ability
to
run
those
locally
and
we're
looking
at
expanding
this
set
beyond
containers
and
just
shelling
out
to
lots
of
other
things
that
people
might
want
like
they
might
want
to
run
the
Azure
functions
emulator,
for
instance,
how
we
do
service
discovery.
This
will
get
better
in
the
future
in
the
near
future.
B
Ty
today,
basically
takes
the
descriptions
of
what
are
all
your
services.
What
ports
are
they
listening
on
bakes
that
into
a
big
list
and
then
constructs
a
bunch
of
environment
variables
and
config
options
from
those
things
and
feeds
them
to
each
other?
So,
within
the
pantheon
of
service
discovery,
there's
a
couple
different
levels
right,
one
level
is
you
could
plug
in
at
the
network
level.
Another
level
is,
you
could
plug
in
at
a
config
level
and
then
the
third
level
is,
you
could
have
like
a
discovery
service
and
things
like
that.
B
So
we're
kind
of
in
the
middle
of
those
two
things
and
on
kubernetes
we
do
a
little
bit
of
both,
but
for
local
development
we
use
config.
So
we'll
probably
package
all
this
all
this
boilerplate
that
you're
seeing
here
into
a
library
by
the
next
time
that
you
hear
from
us
we've
been
copying
pasting
a
bunch
of
code
in
between
samples,
and
we
know
it's
not
a
good
solution.
B
Coincidentally,
we're
also
a
libraries
teams,
we
can
build
libraries
so
expect
more
on
that
from
now,
but
this
is
probably
similar
to
a
lot
of
the
conventions
that
people
are
doing
today.
You
know
when
I
talk
to
a
lot
of
users,
they
would
say
things
to
me
like
I'm,
hard
coding,
I'm,
hard
coding,
the
URLs
and
the
ports
of
things
in
production
and
then
I'm
using
environment
variables
or
launch
settings,
or
something
like
that
to
override
them
or
docker
compose
to
override
them
for
local.
B
So
if
you
can
think
about
how
that's
kind
of
annoying
to
maintain
it
or
it's
kind
of
annoying
to
script,
all
that
out,
so
that
you
can
just
get
these
five
things
to
run
and
communicate
we're
trying
to
make
that
as
simple
as
possible
and
the
way
that
we
do.
That
is
basically
trying
to
take
the
problem
of
figuring
out
exactly
what
these
values
are
out
of
your
hands.
How
we
deal
with
deployments
and
how
we
make
deployments
repeatable
and
useful.
B
You
just
want
it
to
go.
So
the
experience
with
this
today
is
you
don't
even
have
to
know
these
things
exists.
So
if
you're
doing
done-that
apps-
and
you
don't
have
a
lot
of
weird
requirements
like
you-
don't
even
have
to
know
that
docker
images
exist
or
the
kubernetes
exists
in
those
things
will
those
things
that
will
will
work
for
you
and
how
we
do.
This
is
a
sort
of
lens
on
how
we
do
configuration
and
secrets
over.
On
the
right
hand,
side
you've
got
a
view
of
what
things
look
like
in
kubernetes
locally.
B
We
do
a
lot
of
things
with
environment
variables
in
kubernetes
we
actually
use
kubernetes
secrets
and
map
them
to
disk,
as
well
as
environment
variables.
So
all
of
your
all
of
your
service
discovery
concerns
like
what
URL
do
I
talk
to
to
talk
to
the
order,
service
or
just
environment
variables.
All
of
your
actual
secrets,
like
what's
the
password
to
my
Rhetta,
get
put
on
disk,
which
is
slightly
more
secure.
The
fact
that
we're
using
secrets
and
kubernetes
by
the
way
is
not
just
meant
to
be
an
implementation
detail.
It's
meant
to
be.
B
It's
meant
to
signal
a
little
bit
of
a
handoff
there.
So,
like
we
have
a
nice
interactive
experience
for
configuring
secrets
for
things
when
you're
deploying
interactively,
but
we've
we've
kind
of
thought
through
like
how
we
want
that
to
work
for
CI
right
so
like
if
you're
trying
to
deploy
something
in
your
CI
and
you
haven't
configured
all
the
necessary
things
to
configure,
it
will
refuse
to
deploy
and
say
no
I
can't
do
this.
B
Likewise,
the
fact
that
we
bake
things
down
into
kubernetes
secrets
could
be
a
point
of
like
handoff
for
an
ops
team
or
people
who
are
gun.
Ops
in
your
organization
to
say:
hey,
so
we're
put
we're
gonna
put
this
app
in
production
and
all
we
need
from
the
ops
team
is
provisional,
reticence,
tenue
and
then
set
up
a
secret
with
a
URL
that
I
can
use
to
connect
to
it
and
we'll
take
it
from
there.
B
So
that's
kind
of
the
that's
kind
of
the
gist
of
the
main
thrust
of
what
ty
is
about.
Do
we
get
to
people's
questions?
Did
we
answer
the
important
questions
that
people
have
about?
What
have
you
been?
What
have
you
seen
so
far,
and
how
does
it
work
kind
of
things
because
I'm
going
to
deviate
it
into
something
else?
Briefly,
if
we're,
if
we're
ready
to
move
on.
B
B
What
you
think
is
what
else
you
think
is
necessary,
or
what
else
do
you
think
is
needed?
You
can
also
ignore
it.
So
if
you
don't
want
to
use
our
config,
if
you
don't
want
to
use
the
values
that
we
said
in
config,
to
understand
how
to
talk
to
the
service,
if
you
have
a
solution
that
you
like
nothing's,
forcing
you
to
do,
nothing's
forcing
you
to
call
that
method.
B
So
we
don't.
You
know
we're
worthy
is
being
a
team,
and
hopefully
you
have
a
little
bit
of
trust
for
how
we,
how
we
do
things
at
this
point
as
far
as
baking,
in
things
we
try
to
avoid,
we
try
to
avoid
baking
in
opinions
at
at
the
default
layer
unless
they're
really
useful
opinions
that
we
all
agree
on,
and
we
try
to
always
give
you
an
escape
hatch.
B
All
right,
let's
let
me
let
me
show
some
more
stuff,
then
so
one
of
the
things
that
I
heard
pretty
loud
and
clear
from
people
when
we
started
talking
more
openly
about
ty
is
people
had
heard
about
dapper
and
they
wanted
to
know
our
tie
and
dapper
in
competition
with
each
other.
Can
you
use
these
things
together?
Do
you
choose
one
or
the
other
and
what's
the
point
of
tie
if
dapper
exists?
B
So
the
idea,
the
idea
of
something
like
dapper
is
we're
going
to
raise
the
level
of
abstraction
of
all
the
things
that
you're
doing
so
you're,
not
so
coupled
to
the
underlying
platform,
so
dapper
provides
a
set
of
building
blocks.
I,
don't
give
me
credit
for
these
slides
by
the
way
I
didn't
create
them,
I,
just
copied
them
and
David
copied
them
from
somewhere
before
that.
Dapper
provides
a
bunch
of
sort
of
primitives
for
building
micro
services
services,
service
invocation,
which
is
like
method.
B
You
know,
RPC
method
calls
state
management
using
a
like
schema
list,
state
store,
publish
and
subscribe.
So
so
so
not
not
message
queue
like
functionality,
but
pub/sub
like
functionality
and
then
resource
binding
and
triggers,
and
then
there's
a
bunch
more
stuff.
But
the
way
to
think
about
these
things
that
I
think
that's
valuable
is
service
to
service
invocation.
We
know
what
that
is
like
that's
calling
our
calling
an
RPC
endpoint
on
another
service
like
we've
written
HTTP,
endpoints
and
we've
written
G,
RPC
apps
before
we
know
what
that
is.
B
State
management
is
an
interesting
one,
because
it's
it's
just
a
key
value,
store
and
they're,
trying
to
basically
abstract
away
what
your
data
storage
is,
so
that
your
data
storage
is
replaceable,
so
you
might
be
using
read
us
for
local
development
and
you
might
be
using
cosmos,
DB
in
production
and
so
on,
and
then
publish
and
subscribe.
Similar.
Similar
idea
right,
like
instead
of
taking
a
coupling
to
your
cloud
provider,
is
like
messaging
framework.
B
Instead,
you
would
take
a
dependency
on
dapper
and
sort
of
bake
in
the
knowledge
that
you're
using
dapper
and
then
be
independent
of
what
the
actual
like
technology
is
like.
Why,
similar
to
things
like
Azure
functions,
there's
a
whole
bunch
of
triggers
and
bindings
that
you
can
use
to
say
I
want.
You
know
my
app
to
be
called
when
this
thing
happens
in
the
cloud.
B
So
the
really
neat
thing
about
dapper
is
the
dapper
is
architected
in
such
a
way
that
all
these
primitives
are
surfaced
to
you
as
HTTP
or
G
RPC,
because
the
idea
is
that
every
technology
stack
every
language
and
every
framework
knows
how
to
do
HTTP.
Most
of
the
good
ones
know
how
to
do
G
RPC,
and
so
can
we
just
abstract
away
all
these
things
by
turning
a
message.
Queue
message
into
a
CH,
TP
or
writing
to
the
database
is
turned
into
HTTP
or
somebody
uploading.
B
A
file
is
turned
into
HTTP
and
then
basically
get
any
kind
of
stack
or
any
kind
of
web
framework
working
as
a
dapper
application,
with
a
small
amount
of
stuff,
with
a
small
amount
of
build
up.
So
how
this
actually
works
in
production
is
when
you're
in
kubernetes
here
you're
using
a
kubernetes
feature
that
we've
talked
about
I.
Think
the
last
two
times
I've
been
on
here,
which
is
that
pods
can
be
made
up
of
multiple
containers.
So
you
deploy
a
dapper
application.
B
You
put
an
annotation
on
your
deployment
and
then
dapper
will
create
an
instance
of
its
surface
running
in
your
containers
like
address
space,
and
then
you
just
effectively
communicate
with
dapper
and
dapper.
Does
all
the
heavy
lifting
for
you
to
network
you
with
all
these
other
kinds
of
things.
So
it's
a
little
bit
similar
to
using
a
service
mesh
if
you're
familiar
with
that.
But
it's
not
for
all
network
traffic.
B
It's
for
specific
application
level
features
so
I
want
to
show
a
demo,
because
after
all,
we're
here
to
talk
about,
tie
I
wanted
to
introduce
dapper,
because
we've
gotten
a
lot
of
questions
about
dapper
and
about
how
these
things
go
together.
So
I've
got
a
helpful
diagram
here.
I've
built
a
store
like
e-commerce
kind
of
application,
and
it's
a
dapper
application
and
I've
got
I've
got
my
store
front
end
and
I've
got
two
other
applications.
B
I've
got
a
product
service
which
maintains
the
list
of
products
and
I've
got
an
order
service
which
is
responsible
for
my
sort
of
like
ordering
and
fulfillment
and
I'm
gonna.
Do
all
the
communication
between
these
services
through
dapper.
So
what
it's
going
to
look
like
to?
You
is
I'm.
Just
calling
some
net
API
in
my
code
from
the
dapper
SDK.
B
But
what's
actually
going
to
happen
is
the
dapper
is
going
to
send
a
bunch
of
G
RPC
messages
back
and
forth
and
then
depending
on
which
interaction
it
is,
it
might
result
in
data
being
stored
in
Redis
as
key
values,
or
it
might
result
in
Redis
being
used
as
pub/sub,
because
I'm
using
Redis
for
that.
But
if
I
were
to
deploy
this
somewhere,
real
I
could
use
a
bunch
of
Azure
features
or
a
bunch
of
AWS
features
for
this.
So
let
me
jump
to
my
demo
and
I
have
already
got
this
running
here.
B
I
did
a
tyrunt,
let
me
zoom
in
a
little
bit,
I
did
a
Thai
run
and
you
can
see
that
I
built
my
three
projects
and
I
launched
my
three
sir
and
then
I
launched
a
bunch
more
stuff
here,
and
the
reason
for
this
bunch
more
stuff
is
I
have
turned
on
dapper.
Now
here
are
my
three
projects
and
what
I've
done
I've
got
orders
products
in
store,
I've
got
a
tie,
dat,
yeah
Mille,
and
the
only
thing
that
I've
added
to
this
as
I've
said
extensions
dapper,
so
I've
basically
said
like
dapper
equals.
B
Yes
is
kind
of
how
you
can
think
about
that,
and
so
these
are.
These
are
very
hot
bits
by
the
way
this
is
not
checked
in
this
is
super
duper
prototype
mode,
so
I'm
not
going
to
show
you
deployment,
because
I
haven't
implemented
that
yet,
but
I
will
show
you
this
running
locally
and
I've
already
started
it.
So
I
can
go
to
my
local
host
8080.
B
Neurs
into
your
pod,
so
that
you
have
a
service
to
talk
to
well
I've,
basically
got
tyrunt
to
do
the
same
trick.
That
happens
when
you
use
dapper
and
kubernetes.
So
I've
got
my
three
services
here.
I've
got
three
dapper
services
running
here.
These
are
just
these
little
like
go
processes
and
you
know
I
can
see
their
logs
and
everything
here.
So
I
can
see
all
the
all
the
stuff
that's
going
on
inside
these
dapper
services.
So
let's
do
one
of
these
communications.
So
I've
gone
to
my
store.
B
It's
loading,
my
list
of
products,
so
my
list
of
products
actually
comes
from
my
order
service
or
my
my
inventory
service.
My
product
service
I
called
it.
So
this
is
going
over.
Dapper
invoke
and
I'll
show
you
what
that
code
looks
like
so
inside
my
store
inside
this
product
display.
You
can
ignore
some
of
these
errors.
I
will
get
tailored
to
fix
them,
they're
just
bugs
with
the
razor
editor
I'm,
getting
this
I'm
getting
this
list
of
products.
So
it's
not
this
one.
B
It's
in
the
index
I'm
getting
the
list
of
products
and
I'm
getting
this
dapper
client
from
di
and
I'm
using
the
dapper
client
I'm
saying
invoke
method
async,
which
means
invoke
a
method
on
the
product
service
and
the
method
is
called
list
I'm
just
going
to
list
the
products
so
I'm
going
to
get
the
products
and
I'm
going
to
loop
through
them
and
I'm
going
to
do
a
display
of
each
product.
Then
the
other
thing
that
I'm
here
is
when
I
place
an
order
I'm
using
pub/sub.
B
So
when
I
place
an
order,
I'm
gonna
publish
an
event
on
the
order,
placed
topic
and
then
I'm,
basically
going
to
just
I'll
wait
for
the
confirmation
and
the
pattern
that
I'm
using
here
I've
written
an
event
broker
that
runs
in
the
process,
and
so
what
I'm
doing
is
I'm
subscribing
to
pub/sub
I'm,
saying
I'm,
waiting
for
a
I'm
waiting
for
a
response
for
an
order
for
this
order.
Id
then
I
place
my
order
and
I
wait
for
the
response
to
come
in,
which
is
how
that
works.
B
So
if
I
go
back
to
my
store,
I
can
do-it-yourself
haircut
kit
I'm
a
book
to
read
new
video
game
lots
of
things
that
I
might
want
to
buy
I,
really
kind
of
need.
The
do-it-yourself
haircut
kit,
so
I
think
I'll
buy
one
of
those
and
oh
they're
sold
out.
So
there
are
three
people
ahead
of
me
in
line
so
one
of
the
things.
How
can
we
understand
a
little
bit
more
about
how
this
works?
Well,
I
mentioned
that
this
is
going
over
Redis
I'm
doing
pub/sub
with
Redis
and
how
do
I
actually
receive
that.
B
So
let
me
look
at
my.
Let
me
look
at
my
order
service
and
I've
written
a
normal
asp
net
core
controller.
So
the
only
things
about
this
that
are
dapper
actually
is
I've
got
this
attribute
here
that
binds
it
to
a
topic
and
then
I'm
just
getting
the
dapper
client
from
di
like
we
saw
in
our
blazer
component.
So
when
I
get
the
order
in
I'm
going
to
use
the
dapper
state
store
and
I'm
gonna
check
the
basically
the
status.
B
How
many
do
I
have
remaining
I'm
gonna
decrement
it,
because
somebody
wanted
one
I'm,
gonna,
save
the
state
and
then
I'm
gonna
send
an
order,
confirmation
and
I'm
either
gonna
say
you
know
your
order.
Your
order
has
been
sent,
it'll
be
delivered
a
year
from
now
or
I'm,
gonna,
say:
yeah.
Sorry,
it's
on
backorder,
you're
you're
in
line
for
this
long
and
I'm,
publishing
that
back
to
another
pub
subtopic
and
then
I'm,
basically
just
doing
the
same
thing
in
my
in
my
UI
application.
B
B
It's
going
to
send
me
a
little
JSON
blob
and
then
I'm
gonna
tell
my
event:
broker
hey
I
got
a
completion
for
this
thing,
so
kind
of
that's
kind
of
the
flow
of
this
is:
how
does
that
all
work
like
it
I
get
I
go
to
this
page,
it
loads
the
list
of
products
from
the
product
service
I
place.
An
order
sends
a
pub/sub
message
to
the
order
service.
The
order
service
responds
via
pub/sub.
We
wait
for
that
confirmation
and
we
can
do
that
thing.
B
So
we
can
try
buying
something
else
and
we
could
see
that
we're
actually
able
to
preorder
the
videogame,
but
it's
going
to
come
next
year.
So
how
does?
How
does
this
all
work
and
how
does
like?
How
can
we
understand
some
of
the
moving
parts
that
are
here
right?
So
there's
there's
a
couple
moving
parts,
there's
the
three
services
and
how
I
got
this
is
I.
Basically
just
did
a
tie-in
it
on
my
dapper
SLN,
which
gives
me
these
three
services
listed
here
and
then
the
only
thing
I
had
to
add
to
do.
B
This
was
basically
say:
let's
turn
on
dapper.
What
dapper
is
doing
here
is
dapper
is
actually
wired
up
a
pub/sub
message,
bus
and
a
state
store,
and
one
of
the
things
that's
that's
interesting
about
this-
is
that
dapper
is
actually
managing
this
like
Redis
or
I.
Think
this
is
like
just
like
running
on
my
machine.
It's
not
part
of
my
tie
that
yeah
Mel
today
and
I
have
to
manage
these
configs,
so
it
would
be
kind
of
nice
if
I
could
put
these
in
tie-dye.
B
B
So,
even
though
there's
no
like
service
discovery
inside
of
there's
dapper
is
a
framework
or
an
application
sort
of
programming
model
that
brings
its
own
like
service
discovery
to
the
table,
you
can
still
use
that
with
tie
like
there's.
No
reason
why
you
have
to
sort
of
be
exclusive
to
the
same
set
of
primitives.
That
tie
provides
it's
a
little
bit
more
of
a
menu
that
makes
sense.
A
B
So
so
I
think
I
think
the
thing
is
how
I
would
think
about.
Dapper
is
I,
would
position
I
would
position
dapper
as
a
like
a
programming
model
and
I'm
gonna
go
back
many
slides,
because
there's
a
reason
why
I
put
this
slide
in
the
stack?
Is
the
dapper
dapper
basically
operates
like
at
these
levels,
they've
added
secrets,
but
they
basically
operate
at
the
communication
protocols,
service
discovery,
Diagnostics
and
observability
layers.
Dapper
doesn't
have
any
opinions
about
how
you
deploy
your
application.
B
They
have
a
few
opinions
about
how
you
can
run
it
locally,
but
I
can
do
more
and
they
don't
have
a
ton
of
opinions
about
configuration,
but
they're,
adding
some
so
like
dapper
is
actually
solving
a
bunch
of
problems
that
ty
doesn't
solve
and
it's
doing
it
in
a
pretty
opinionated
and
like
platform-agnostic
kind
of
way.
So
it's
really
it's
really
sort
of
like
people
have
asked
the
question:
can
you
use
these
two
things
together
and
I?
Think
about
it
as
like?
B
B
So
with
that,
I
want
to
say
a
few
more
things.
That's
that's
all
the
demos
that
I've
got
for
today.
I
wanted
to
say
a
few
more
things
about
what
we're
sort
of
doing
so
so
ty
is
an
experimental
project
from
Adana
team.
We're
going
to
be
working
on
this
throughout
da
no
5.
The
repo
is
public.
We
have
bits
that
you
can
try
out
some
of
the
things
most
of
the
things
that
you've
seen
here,
except
for
my
demos,
you'll
be
able
to
do
with
the
bits
that
are
linked
to
there.
B
B
What's
in
it,
maybe
once
a
month
and
have
some
updated
samples
and
demos,
and
things
like
that-
we're
in
kind
of
an
early
adopter
sort
of
phase
where
we're
looking
for
people
to
try
and
engage
with
it
and
give
feedback
and
collaborate
with
us
who
have
got
a
little
bit
of
experience
in
this
space
or
maybe
no
experience
in
the
space
and
want
to
help,
try
and
resolve
what
some
of
your
pain
points
are
either
for
development
or
for
running
things
in
production.
So
I
think
yeah.
A
A
It
is
much
easier
than
I
thought
it
was
gonna
be
like
I
was
like
okay
I,
should
you
know
not
be
completely
stupid
on
this?
All
and
I
should
take
a
look
and
I
poked
around
and
I
had
it
running
in
sec,
or
you
know
a
few
minutes
and
I
was
like
wow.
This
is
super
easy,
like
I
mean
you
know,
and
it
was,
and
it
was
explaining
stuff
pretty
well.
So
anyhow,
yeah.
H
A
You're
showing
that
you
can
solve
these
big
complicated
problems
and
that
you've,
you
know
that
you've
solved
a
big
problem
space
and
for
some
people
on
the
on
the
call
like
from
the
chat,
it's
obvious
that
you're
solving
some
solutions
and
they're
like
wow.
This
is
perfect,
you
know,
but
it
was
nice
for
me
as
kind
of
a
newbie
at
some
of
this
to
see
how
smoothly
it
work
to
like
I
could
get
you
on
pretty
quickly.
F
B
F
B
A
F
A
I
F
So
maybe
maybe
I'll
get
done
with
this
demo
fairly
quickly
in
the
middle
of
QA
time.
Just
to
do
that
all
right,
so
I
wanted
to
very
briefly
go
over
a
demo
about
pretty
much
what
I
guess
the
idea
is
like.
Let's
say
you
have
a
kubernetes
cluster,
that's
deployed
to
production
or
deployed
in
like
a
dev
environment,
or
something
like
that,
and
you
have
this
app.
That
has
many
many
services.
F
Let's
say
you
have
like
15
services
but
for
the
sake
of
demo
I'm,
eliminating
if
you're
running
this,
like
this
15
service
thing
locally
like
if
you
have
a
front-end
application
and
you
wanted
to
get
the
backend
application,
you
can
do
like
again
request
for
HTTP
local
hosts
with
some
URL
right,
and
that
would
all
work.
But
let's
say
I
wanted
to
develop
this
front-end
locally
and
I
wanted
to
reference
something
inside
the
kubernetes
cluster
so
like.
If
I
wanted
the
front-end
to
talk
to
the
backend,
how
would
I
do
that?
F
Well,
the
problem
is,
is
that
if
you
have
the
backend
inside
kubernetes
and
it's
totally
not
being
like
it's
not
exposed
to
the
Internet
at
all,
the
front-end
can
never
actually
get
the
the
back-end
at
all.
It
just
won't
work
at
all,
because
the
backend
is
totally
isolated
inside
of
the
kubernetes
cluster.
F
So
if
we
were
to
have
the
front-end
inside
of
the
grenades
cluster
effectively,
you
can
be
able
to
get
the
backend
by
doing
a
get
request
for
the
backend,
just
with
the
DNS
name
called
back-end,
and
when
both
is
its
are
running
inside
kubernetes
like
this
all
works
totally
fine.
So
the
question
is:
could
we
enable
a
scenario
like
this?
Actually
work
and
I
have
a
prototype
of
this
at
least
trying
to
do
this
in
a
way
that's
somewhat
effective.
Can.
F
I
hate
it.
If
you
thoughts
to
do
that.
Okay,
so
I
have
a
front
and
back
in
application,
this
app
is
fairly
straightforward.
All
it
has
is
a
front
end
which
calls
into
a
back-end
what
this
does
is.
It
gets
the
information
about
where
the
front
end
is
listening.
Zoom
in
a
bit
get
some
information
about
where
the
front
end
is
as
well
as
the
back
end.
I
guess
my
IP
information
pretty
basic
stuff
and
the
back
in
itself
is
just
responding
with
the
IP
and
os
name
of
where
it's
at
very
straightforward
example.
F
So
today,
I
have
this
currently
published.
Chicharones
I
have
a
front
end
and
a
back
end
there.
Currently
there
right
here.
So
if
I
were
to
do
a
if
I
were
do
a
port
forward
in
order
to
get
the
front
end,
I
could
hit
the
front
in
there
in
order
to
get
information
from
it.
But
what
if
I
wanted
to
try
using
a
front
end?
That's
local
as
well
as
a
back
end.
That's
remote,
so
I'm
gonna
run
a
command
called.
F
D
F
May
have
broken
something
d:
ty,
it's
ty,
because
I'm
using
a
experimental
thing,
I
may
have
broken
something
anyways.
The
the
demo
is
going
to
be
me
connecting
to
a
having
a
front-end
that
connects
to
a
back-end
and
actually
get
connects
the
your
local
computer
to
the
kubernetes
cluster.
To
make
it
so
things
like
calling
the
DNS
name,
back-end,
actually
works
idem,
Alyssa
Fowler
about
like
20
minutes
ago
and
I.
Guess
it's
not.
F
So
we're
not
we're
trying
to
play
around
with
a
lot
of
things
in
this
area.
We're
not
necessarily
sure
like
what
what
the
end
story
of
that
is,
but
connected
development
seems
to
vary
a
very
common
theme
of
like
having
something
you
run
locally,
while
the
rest
of
your
app
runs
in
production
or
in
a
dev
test
test
lab
or
something
like
that,
and
we
were
trying
to
prototype
around
with
that
and
figure
out
some
solutions
for
it.
Unfortunately,
my
demo
failed,
which
sucks
might
happen
so
maybe.
A
A
F
H
A
A
F
Me
like
three
seconds
to
finish
the
demo
so
anyways.
Normally
you
do.
This
normally
I
wouldn't
be
able
to
query
the
backend
from
my
like
browser,
but
I
am
able
to
like
refresh
with
the
DNS
name
back-end
and
actually
get
the
backend.
The
other
thing
I
can
do
is
I.
Can
you
tyrunt
de
time?
Sorry
the
time
run
then
remotes
as
this
and
local
as
front-end,
and
this
is
gonna
run
the
front-end
locally
and
remote
run
the
rest
of
the
remote
services
remotely
so
which.
F
Great
sure
anyways
this
is
gonna,
show
that
the
front
end
is
running
locally,
while
the
back
end
is
run
externally
and
if
I
actually
go
to
the
URL
I
should
be
able
to
see
the
front
end
information
locally
and
the
back
end
information
for
my
kubernetes
cluster.
So
this
is
showing
effectively
this
diagram
of
I'ma
frontin.
That's
running
locally
I'm
able
to
call
get
on
the
back
end
and
actually
get
the
backend
information
from
kubernetes.
F
There's
a
lot
of
magic.
That's
going
on
I,
don't
necessarily
know
how
much
you
can
reveal
yet,
but
there's
a
lot
of
magic
in
order
to
make
it.
So
when
you
do
a
DNS
lookup
on
that
actually
looks
inside
the
Cooper's
cluster
first,
before
doing
other
things
got
a
lot
of
stuff
like
we're
still
trying
to
figure
out
how
to
get
it
all
working.
But
this
is
a
little
of
demo
magic,
but
a
little
bit
of
some
craziness.