►
Description
Join members from the .NET teams for our community standup covering great community contributions for Azure, AI, ML, Containers, and more.
Suggest an idea for a standup: http://aka.ms/dotnet-standup-suggestion
A
B
Sure
no
spaces
is
an
add-on
for
Azure
kubernetes
service,
which
is
Microsoft's,
managed
cloud
kubernetes
offering
and
really
deaf
spaces
is
focused
on
developer
scenarios.
So
how
can
we
help
developers
be
more
productive,
developing
as
they
can
and
getting
into
a
lot
of
details,
but
essentially
allows
you
to
develop
really
effectively
large
micro
services
based
applications
as
part
of
a
team?
You
can
debug
your
applications
live
in
communities.
You
can
create
what
we
call
dev
spaces
that
isolate
your
code
from
anything
else.
B
A
That
makes
that
really
easy
and
I
think
solves
a
lot
of
problems
that
people
who've
never
tried
to
have
more
than
maybe
two
or
three
things
talk
to
each
other.
At
the
same
time,
hopefully,
we'll
be
able
to
relate
to
and
understand
before
that,
however,
we're
gonna
switch
over
and
we're
gonna
look
at
a
couple
blog
posts
from
community
members,
which
is
our
tradition
here.
A
So
this
one
is
something
popped
up
in
my
feed
yesterday
actually
was
just
author,
March
26,
so
its
blog
post
by
Andrew
Locke,
great
name
by
the
way,
and
he
talked
about
exploring
dotnet
core
MCR
docker
file.
So,
let's
start
off
by
talking
what
is
microsoft
container
registry
Lisa?
You
explain
that
to
me.
B
A
B
So
a
container
registry
is
a
place
where
you
can
host
container
images
that
that
you've,
created
or
obviously
companies
like
Microsoft,
can
host
their
own
container
images.
You
might,
if
you've
been
in
the
container
world,
you
might
have
heard
of
docker
hub
is
sort
of
the
first
or
the
the
most
well
known
container
registry,
but
we
also
have
Azure
container
registry,
which
is
an
azure
based
service
or
where
you
can
create.
You
can
create
your
own.
B
As
your
container
registries,
you
can
put
whatever
images
that
you
want
up
there,
they're
close
to
your
wherever
your
containers
are
running
so
pulling
those
container
images
and
running
them.
It's
very
fast
and
then
MCR
microsoft.
Container
industry
is
just
for
official
microsoft,
images,
okay,
so.
A
What
I
think
what
I
heard
you
say
is
that
a
container
registry
is
basically
generally
hosted
somewhere
in
the
cloud
could
be
on-premises.
But
it's
a
storage
mechanism
where
we
can
write,
make
containers
available,
and
then
anybody
who
has
access
to
that
registry
in
the
case
of
micro,
container
registry
or
docker
hub
they're,
public
or
you
could
have
private
ones
as
your
container
registry
or
set
up
your
own.
If
you're
so
inclined.
That
then
lets
people
who
have
accessed
polling
to
use
those
images
that
become
exactly.
A
And
so
one
of
the
things
that
Andrew
talks
about
here
in
this
blog
post
at
the
beginning-
and
he
talks
about
the
idea
of
installing
base
images,
which
is
really
what
most
of
our
image
is
available
in
Mexico
container
registry,
are,
is
a
base.
Image
I
think
is
one
of
the
things
that
makes
containers
really
cool,
but
you
could
start
by
explaining
to
me
what
is
the
base
image
yeah.
B
Base
image,
it's
an
image
that
you
can
use
as
a
starting
point
for
your
own
images
and
typically
a
base
image
will
package
up
some
things
that
are
useful.
Certainly,
an
operating
system
would
be
the
the
first
or
the
most
obvious
part
of
the
base
image,
so
Linux
or
Windows
different
Linux
distributions,
different
on
Windows
server,
core
versus
Nano
and
so
on,
but
you
can
package
additional
things
into
a
container
image
so
or
into
a
base
image.
B
Excuse
me
so
for
dotnet,
for
example,
we
have
some
that
are
available
that
are
really
optimized
for
running
a
production
image,
they're
a
lean
and
mean,
and
fast
there
are
others
that
are
that
include
SDKs
and
other
components
that
that
you
would
use
to
actually
build
another
image
or
to
build
a
container
out
of
a
container.
These
images
could
also
we
have
base
images
for
node
and
for
other
programming
languages
that
again
just
make
sure
that
you've
got
everything
that
you
need
to
build.
Your
application
run
your
application,
whatever
you're
trying
to
do
so.
A
If
I
was
gonna
kind
of
compare
this
to
the
virtual
machine
world
right
and
I
think
one
of
the
things
that
docker
tries
to
solve
for
me
as
it
tries
to
solve
a
lot
of
the
problems,
a
virtual
machine
solved,
but
while
being
more
efficient
about
some
of
those
things
right
so
isolation
of
application.
So
anything
in
a
container
should
know
unless
I
give
permissions
to
it
to
know
about
anything
else
in
another
container.
But
I
don't
need
an
entire
copy
of
that
operating
right.
So
the
host.
B
B
A
I
think
one
of
the
things
that
base
images
do
that's
really
cool
and
tell
me
if
I'm
right
on
this
is
you
mentioned
that
they're
pre-installed
with
a
bunch
of
stuff?
So
it
would
almost
be
if
I
go
back
to
the
idea
of
like
a
virtual
machine
world,
the
ability
to
get
a
much
smaller
but
VHD
or
virtual
hard
disk,
that's
pre-installed,
with
the
copy
of
every
anything
that
I
want,
and
these
are
things
that
are
being
made
freely
available
by
companies
like
Microsoft
and
other
people.
A
So
if
I
want
to
run
an
asp.net
core,
2.1
application,
I
can
just
go
get
microsoft
say
it's
been
a
core
to
it
and
it
already
comes
pre-installed
with
a
fully
up-to-date
version
of
asp.net
course.
Then
all
I
have
to
do
is
say
use
this
some
a
base
image,
here's
my
code
and
problem
solved.
I
don't
actually
have
to
go
through
the
effort
of
like
figuring
out
the
install
script,
to
get
it
set
up
and
install
dotnet
core
onto
that
image.
Is
that
exactly.
B
A
If
you
weren't
aware
this
is
that,
unlike
a
VHD
where
you're
actually
bringing
your
own
copy
of
the
operating
system,
and
so
I
could
run
a
Linux
VM
on
a
Windows
machine
or
a
Windows
VM
on
a
Linux
machine
and
docker,
because
at
the
end
of
the
day,
to
make
them
a
much
smaller,
more
efficient
image,
we're
actually
sharing
the
underlying
operating
system.
Kernel
I
couldn't
run,
for
example,
a
Windows
image
on
a
Linux
machine
or
a
Linux
arm
machine
on
a
Linux
arm.
Image
on
a
Linux
AMD
64
processor
is
that
yeah.
B
A
So
one
of
the
things
he
talks
about
a
little
bit
here,
which
I
think
may
be
the
last
thing
that's
worth
talking
about
is
the
idea
of
how
does
docker
deal
with
with
version.
So
one
of
the
things
Andrew
talks
about
here
is
he
says
you
should
only
be
specific
about
which
version
of
docker
image
you
want
to
build
on
your
docker
files,
I,
don't
use
latest.
For
that
reason,
all
the
images
I
mentioned
in
the
post
use
the
current
latest
version,
but
he
actually
provides
specific
numbers.
Can
you
talk
to
me
a
little
bit?
B
So,
typically,
with
an
with
a
given
image,
you
will
have
a
tag
called
latest
which
refers
to
the
very
latest
version
of
whatever
you're
dealing
with.
So
maybe
it's
the
latest
version
of
Windows
server,
including
all
the
latest
security
patches
and
so
on,
as
well
as
the
latest
version
of
the
dotnet
framework,
or
you
can
specify,
depending
on
the
tags
that
are
provided
or
available
for
a
given
image.
You
can
say.
A
B
Want
this
specific
version
of
of
this
particular
image
and
the
difference
there
comes
when
you
specify
that
in
your
docker
file
and
when
that
docker
file
is
read
as
part
of
your
container
build
process,
it
will
pull
that
image
sort
of
real
time.
So
if
you
have
latest
specified
and
you
you've
tested
against
the
current
latest,
but
tomorrow
we
release
a
new
security
patch
for
the.net
framework
or
for
for.net
core
and
that
gets
incorporated
into
the
image.
That's
tagged
latest.
B
If
you
were
to
build
your
container
image
tomorrow,
it
would
actually
be
a
different
container
image.
It
would
be
base
that
the
base
image
would
be
different
because
when
you
built
your
image,
it
would
pull
the
very
latest
version
of
the
the
image
that
you
specified,
and
so
it
could
auto
update.
That
can
be
good.
That
can
be
bad
depending
on.
Ideally,
of
course,
there's
no
breaking
changes
there,
but
we
know
that
a
lot
of
times
people
like
to
have
more
control
over
what
exactly
they
are
specifying
as
their
base
image.
B
A
So,
by
not
using
latest
I'm,
pinning
myself
to
a
to
one
that
I
know
that
every
time
I
trigger
my
bill,
I'm
going
to
get
the
same
thing
I
tested,
whereas
latest
it
just
means
hey,
wait.
Microsoft
pushes
an
update
the
next
time,
I
rebuild
I'm,
getting
something
I
may
or
may
not
have
tested,
and
despite
our
best
efforts
software
it
turns
out,
especially
when
it
comes
to
operating
systems.
A
Runtimes
frameworks,
things
like
that,
it's
very
hard
to
change
anything,
even
security
bugs
that
may
not
have
some
negative
impact
on
some
application
in
the
world
needs
to
be
reacted
to
exactly
yeah
cool.
Thank
you.
That's
been
excellent
introduction
to
kind
of
containers
and
container
registries,
but
I
think
it's
a
great
foundational
transition
to
some
X
blog
post.
We
have
out
here
talking
about
as
your
dev
spaces
a
free
tool
developing
an
app
on
top
of
a
KS.
A
B
So
kubernetes
is
a
container
Orchestrator.
There
are
a
lot
of
container
orchestrators
out
there,
but
kubernetes
is
really
it's
really
taking
the
lead.
As
far
as
the
container
orchestration
role
goes
now,
what
is
container
orchestration?
We've
talked
quite
a
bit
about
containers
and
talked
about
building
a
container
image,
but
in
the
real
world
you
usually
as
soon
as
you
get
to
any
level
of
complexity
with
the
micro
services,
application
you're
going
to
have
multiple
containers
running,
and
so
then
you
start
running
into
problems
like
well.
B
How
do
they
talk
to
each
other
if
this
container
needs
to
talk
to
that
container?
How
does
it
refer
to
itself,
and
how
can
we
make
it
easy,
as
we
move
through
from
dev
to
test
production
and
so
on,
to
refer
to
these
containers
consistently,
you
run
into
problems
like
what
happens
if
a
container
crushes
or
fails
for
some
reason
how
do
in
production?
How
do
we
make
sure
that
that
container
gets
spun
up
again?
B
How
do
we
scale
so,
if
it's
black
friday
and
we
need
to
scale
out
our
website
to
handle
increased
order
of
traffic?
How
can
we
do
that
really
easily
and
not
sit
there,
and
you
know
type
docker
run
a
million
times
to
spin
up
additional
container
images,
and
those
are
the
sorts
of
problems
that
a
container
Orchestrator
like
kubernetes
addresses.
Yes,.
A
I
think
one
thing
they
BD
packing
that
a
little
bit
that
I
heard
I
think
there's
some
concern
a
lot
of
the
concerns
you
talked
about
that
kubernetes
or
any
orchestra
communities,
obviously
being
probably
the
industry
leader
right
now.
A
lot
of
those
feels
like
concerns
for
an
Operations
time,
not
so
much
development
time.
For
me
as
a
developer,
I
think
there's
one
thing
you
seven.
That
is
actually
very
interesting
for
me
as
a
developer,
which
is
how
to
services
talk
to
each
other.
A
B
Within
within
kubernetes,
each
service
has
a
specific
name
associated
with
it.
So
if
I
have
service
a
running
and
serve
be
running,
I
don't
have
to
change
connection
strings
or
anything
like
that.
It's
just
a
for
the
two
of
them
to
talk
to
each
other.
It's
just
a
very
simple,
it
just
say
HTTP
service
a,
and
that
is
the
way
that
service
B
can
refer
to
service
a
and
vice
versa,
and
that
stays
consistent
as
long
as
you're
within
the
scope
of
that
kubernetes
cluster.
A
Saying
would
you
said
there
is
a
give
me
relative
DNS
or
domain
name
services
like
domain
name
lookup.
So
if
I
have
a
some
sort
of
Orchestrator
running
on
my
local
machine,
whether
that
be
docker,
swarm
or
mini
coup,
which
is
a
local
version
of
kubernetes
that
if
I
write
in
my
code
or
have
in
my
config
file,
oh
I'm,
looking
to
talk
to
service,
say
he's
doctor
service,
B
I
can't
refer
to
a
service.
A
B
and
that'll
always
provide
DNS
for
me
and
then,
when
that
moves
into
maybe
a
production
environment
or
a
staging
environment.
As
long
as
they're
named
in
that
environment
service
a
and
service
B,
when
the
environment
set
up
like
nothing
about
my
code
or
my
config
has
to
change,
because
it's
all
relative
within
the
scope
of
that
cluster.
Exactly.
B
Yep
and
I
would
say
that
from
a
developer
perspective,
you're
right,
a
lot
of
kubernetes
is
about
operation
operations.
In
fact,
in
talking
to
a
lot
of
developers
out
there,
they
say
you
know:
I,
don't
I,
don't
use
kubernetes
at
development
time,
but
almost
everybody
has
to
solve
that
problem
of
how
to
services
talk
to
each
other
at
development
time
and
that's
certainly
running
in
kubernetes.
This
is
one
way
to
do
that.
Great.
A
A
There's
some
certain
challenges
and
having
things
talk
to
each
other
is
kind
of
a
tip
of
that
iceberg
that
come
try
to
develop
micro
service
applications
that
will
ultimately
be
hosted
in
a
orchestrated
container
environment
Cooper
don't
make
some
blankly,
but
maybe
something
else
coming
in
the
cloud.
So
you
talked
about
what
some
of
those
challenges
are
and
why
dev
spaces
or
how
does
that
phases
helps
me
address
those
problems
and
makes
developing
this
type
of
application
easier.
Yeah.
B
I
think
a
lot
of
what
we
see
so
I
mean
developers
have
generally
have
a
great
experience
for
getting
started,
coding
and
debugging
and
so
on.
That's
that's
not
really
a
problem.
We've
been
doing
that
for
decades,
where
the
challenges
really
arise
in
development
or
where
some
of
the
challenges
that
we're
trying
really
trying
to
address
with
dev
spaces
occur.
B
When
you
get
your
code
to
the
point
where
you're
you're
pretty
confident
in
it
developing
locally
and
now
it's
time
to
integrate
as
part
of
the
larger
application
and
typically
up
to
that
point
developers
that
they
are
running
locally,
we
see
some
developers
who
are
running
literally
just
using
I
as
Express
or
whatever
native
processes.
They
were
using
before
others
who
are
using,
for
example,
our
container
tools
and
Visual
Studio
or
NBS
code
to
develop
within
containers
locally,
but
typically
that's
just
the
scope
of
one
developer.
B
So
if
I'm
working
on
a
feature
I'm,
not
integrating
your
code
at
that
point,
if
we're
part
of
a
development
team,
I
might
be
using
mocks
or
stubs
to
simulate
your
feature
or
maybe
I
am
pulling
your
code
and
compiling
that
locally.
But
then
that
introduces
challenges
like
what,
if
your
code
relies
on
a
database
and
how
do
I
figure
that
out?
How
do
I
know
what
database
to
spin
up
and
what
to
put
in
there.
B
So
what
we
see
a
lot
is
that
once
developers
finish
their
code,
they
they
check
it
in
or
there's
some
sort
of
process
to
get
to
an
integration,
testing
environment
and
that's
the
first
time
where
I
would
run
my
code
against
your
code
and
against
all
of
our
the
other
members
of
our
development
team,
so
that
again
most
most
people
have
that
already.
But
there
are
some
challenges
with
an
integration
testing
environment.
B
It
can
be
slow
to
deploy
to
that
or
difficult
a
lot
of
times,
you're
right,
relying
on
a
CI
CD
pipeline
or
something
like
that
to
get
your
code
into
into
an
integration
testing
environment
which
you
know
can
take
five
or
ten
minutes
and
then
any
sort
of.
If
something
goes
wrong,
I
mean
not
that
that
ever
happens
right
when
you're,
when
you're
integrating
different
four
components
that
haven't
been
tested
together.
B
Yet,
but
if
something
happened
to
go
wrong
troubleshooting
that
can
be
very
difficult
because
you're
it's
hard
to
hook
up
a
debugger
in
a
remote,
especially
for
running
in
kubernetes
environment,
or
something
like
that
and
so
you're
relying
on
logs
and
that
can
be
challenging.
So
that's
really
where
we
think
that
dev
spaces
can
help
and
what
it
brings
is
a
very
easy
way
to
deploy
your
code
into
an
azure
kubernetes
service
cluster
and
run
it
in
the
context
of
that
full
application.
But
you
also
have
logs
that
are
streamed
back
to
your
development
console.
B
We
can
even
hook
up
a
debugger
for
you
with
just
one
click,
and
so
you
get
a
lot
of
the
benefits
of
local
development,
but
you're
not
something
locally
or
developing.
Again.
You're
running
against
your
code,
developing
against
your
code,
that's
running
in
the
cloud
and
really
in
a
more
production,
realistic,
environment,
great.
A
B
So
what
I've
brought
here
is
a
sample
application
that
we
felt
on
the
dev
spaces
team.
This
is
out
on
github
and
we're
working
to
package
this
up
with
some
quick
starts
and
other
documentation,
but
I'll
give
you
a
little
bit
of
a
sneak
peek
here
and
it's
a
bike
sharing
application,
so
I'm
gonna
log
in
as
a
customer
here
and
I
am
going
to
I,
can
get
a
list
of
bikes.
I
can
rent
a
bike
here,
so
great
I've,
rented
this
lovely
bike
and
now
I
can
return.
B
It
ask
me
to
confirm
the
return
and
I
can
write
a
little
review
here
and
now.
I'm
back
to
my
list
of
bikes,
so
I
could
go
and
run
another
bike.
Now.
I'm
gonna
show
what's
happening
behind
the
scenes
here
and
I'm
just
gonna
to
get
my
mic
kubernetes
pods
here.
So
this
is
running
a
command
against
my
kubernetes
cluster
and
once
it
connects
in
list
you'll
see
that
there's
a
number
of
different
services
running
as
part
of
this
application
and
there
are
database
dependencies.
I
mean
this
is.
B
B
Yes,
exactly
right
and
yes,
we've
got
a
number
of
different
languages
running
there.
We
go
ok,
so
you
can
see
these
the
services
that
are
running
in
this
namespace
called
dev.
These
are
what
make
up
my
application,
so
I
expect
sharing
web
and
so
on.
Now-
and
this
is
so-
you
can
think
of
this
as
my
integration
testing
environment,
where
I've
deployed
my
entire
application
from
end
to
end-
and
this
is
really
this
candy
outside
of
dev
spaces.
In
fact,
I
use
just
a
standard
helm
install
command
to
run
all
these
services.
B
Helm
is
a
package
manager
for
kubernetes.
That
makes
it
really
easy
to
to
run
a
big
application
like
this
and
make
sure
that
everything's
in
place
that
needs
to
be
there
for
it
to
run
successfully.
So
that's
great,
so
I've
got
my
application
running,
but
now
I've
been
making
some
changes
to
this
reservation
engine
surface
here
and
maybe
I
want
to
want
to
test
them
out
here.
B
So
got
the
code
running
or
I've
got
the
code
pulled
up
here
in
NPS
code
and
we
also
have
a
plugin
by
the
way
for
visual
studio
I'm
just
using
code.
Today,
I'm
gonna
go
and
I'm
going
to
debug
just
this
service
and
a
couple
of
things
to
note
here:
I
only
have
the
code
for
reservation
engine
pulled
up.
B
For
me,
it's
actually
doing
that
up
in
Azure
I
happen
to
have
run
docker
running
locally
at
our
desktop
on
on
my
laptop
here,
but
it's
not
required,
and
it
is
then
spinning
up
that
container
in
my
aks
cluster
and
it's
running
in
what
we
a
deaf
space,
so
I'm
gonna
go
back.
No,
this
is
running
and
I'm
gonna
run
my
coop
control
command
again.
B
I
think
won't
take
quite
as
long
this
time,
but
what
you'll
see
here
is
that
I
still
have
all
of
my
services
running
in
that
dev
namespace,
but
you'll
see
another
instance
fun
up
of
just
of
the
reservation
engine
service
that
I'm
debugging
here
and
it's
running
in
a
certain
a
dev
space
called
or
a
namespace
called.
A
new
feature
is
what
I
called
it
I
don't
know,
don't
know
why
my
cluster
is
responding
slowly
today,
but
that
so
what
happens
is
there?
B
We
go
yes,
so
right
down
here,
I've
got
this
new
instance
of
reservation
engine.
That's
running
a
new
feature,
and
what
happens
here
is
that
dev
spaces
automatically
routes
requests
between
all
these
services
when
I
use
a
special
URL
to
access
the
site.
So
you
can
see
up
here.
I
was
I
was
just
on
Deb
bike-sharing
web
blah,
blah
blah
I'm
gonna.
Add
a
prefix
here
to
this
URL.
It
is
that
space
name
that
I
mentioned
of
a
new
feature
and
then,
of
course
the
rest
of
the
URL
is
the
same.
B
B
You'll
see
that
it's
gonna,
my
browser's
gonna
stop
here
for
just
a
minute
and
I've
actually
hit
the
breakpoint
here
that
I
had
set
previously.
So
this
is
a
live
debugging
session
here.
I
can
step
through
my
code.
I
can
inspect
the
values
of
my
variables
here.
I
can
do
anything
that
I
would
do
if
I
were
debugging
locally,
but
I'm
actually
debugging
live
against
aks.
Now
you
might
be
saying.
B
B
Engine
service
to
the
one
that's
running
in
the
dev,
the
dev
namespace
right
here,
so
I
still
have
my
debugger
hooked
up,
but
you're
gonna
see
that
my
rent
bike
flow
is
going
to
work
just
fine
here,
because
I
am
running
against
completely.
This
is
this
is
actually
accessing
a
completely
different
version
version
of
reservation
engine,
so
I
could
break
reservation.
Engine
I
can
debug
all
day
long
and
I
don't
affect
your
work,
the
work
or
the
rest
of
the
team.
A
B
Yeah
and
I
mean
it's:
it's
a
huge,
it's
really
a
lot
more
efficient
too.
We
talked
to
a
lot
of
customers
who
set
up
something
similar
to
this.
They
might
have
isolated
environments
for
each
developer.
So
you
have
your
kubernetes
cluster
I
have
my
kubernetes
cluster,
obviously
that
that
can
get
to
be
pretty
pricey
and
hard
to
me
or
they
have.
You
know
you
might
have
your
namespace
within
the
cluster.
I
have
my
namespace,
but
we
then
we
have
to
run.
We
each
have
to
run
all
of
these
services.
B
A
My
local
machine,
and
so
typically
the
read
either
it
is
feasible
when
we
write
a
script
and
we
do
it
or
it's
not
feasible
and
we
have
to
invest
a
bunch
of
effort
into
getting
mocks
or
stubbing
things
out
or
you
know,
have
different
logic
like
if
I
see
the
you
know,
a
local
development
environment
variable
set
like
just
you
know,
read
this
file
in
and
pretend
it
was.
The
results
from
your
service
remain
that
which
obviously
is
time
and
it's
time
that
may
be
necessary
to
make
me
productive
as
a
developer.
A
But
if
we
just
start
a
project
or
migrate
a
project
to
dev
spaces,
it
feels
like,
if
it
makes
sense,
we
may
be
able
to
eliminate
a
lot
of
that
overhead
and
set
up.
You
know
you
joined
the
team
tomorrow
and
hey,
you
know
just
start
working
on
this
service
and
it
all
just
works
as
long
as
you
remember
to
you
know,
throw
in
your
own
namespace,
like
everything's,
great
you're,
using
all
the
rest
of
our
services.
No,
you
know
power,
shell
or
bash
script
to
run
to
install
a
bunch
of
stuff
I'm,
not
wasting.
A
B
Yeah
I
think
you
have
a
great
point
about
the
fact
that
you
may
not
be
able
to
run
everything
locally.
This
is
this
is
a
relatively
small
application
with
just
you
and
it's
like
six
or
seven
services,
but
we
to
customers
who
have
hundreds
of
services
running
and
that
at
some
point
you
just
run
out
of
computing
power
on
your
local
development
machine.
But
you
never
run
out
of
computing
power
in
the
cloud,
so
you
can
always
run
everything
up
there
and
then,
yes,
from
a
new
developer
perspective.
B
B
That
is
so
there's
somebody
who
is
somebody
for
the
whole
team
who
is
responsible
for
setting
up
that
team's
baseline,
and
then
it
is
literally
just
as
easy
as
the
new
dev
starts
tomorrow,
they're
working
on
reservation
engine,
they
sync
the
code
just
for
reservation,
engine
and
they're
off
and
running,
and
they
can
do
whatever
they
need.
To
do.
A
Angle,
it's
actually
an
interesting
point.
You
talk
about
the
NGO,
which
obviously
is
its
own
language,
but
in
going
back
to
why
are
containers
an
interesting
piece
of
the
puzzle
for
solving
some
of
these
problems,
one
of
the
things
even
in
dotnet
land
that
containers
solved.
For
me,
if
I
wanted
to
have
a
pure
dotnet
solution
is
because
of
the
way
we
talk
about
containers,
work,
I
could
have
different
versions,
of.net
running
in
different
container
images.
Absolutely.
B
A
You
know
I
talked
to
a
lot
of
people
over
the
years
that
suffer
from
I
call
it
the
tyranny
of
the
lowest
common
denominator.
So
I
was
visiting
a
customer
a
couple
years
ago.
They
were
having
an
issue
in
one
of
their
servers
and
they
say
I
maybe
can
help
us
debug
it.
That's
great.
You
know
we
got
dumped
back.
We
should
start
trying
to
run
some
tools
against
some
of
the.
A
B
A
A
Out
how
to
get
it
to
work,
because
it's
critical
and
like
it
just
works
out,
so
we
keep
dessert
and
so
the
beauty
of
containers
from
a
technology
perspective,
whether
it's
mixing
go
or
but
since
forgotten
community
standard
well
focus
on
that
is,
if
I
want
to
mix
dotnet
core
to
you
know
or
dotnet
framework.
Something
that's
on
for,
oh,
that
these
things
have
to
still
talk
to,
but
I'd
like
to
take
advantage
of
some
of
the
more
modern
features
and
security
benefits
and
other
things.
A
I
could
have
a
container
that's
running
the
dotnet
for
application
and
I
have
another
container.
It's
running
to.net
for
72
application
and
I
can
use
a
lot
of
the
great
new
functionality
and
so
by
you
know,
and
because
of
that
isolation,
I'm
no
longer
tied
to
the
tyranny
of
the
same
server,
either
right.
B
Yeah
I
mean
going
back
to
what
we
were
talking
about
at
the
start
of
this
about
base
images
and
all
of
that
there's
a
docker
file
that
literally
there's
a
docker
file
for
each
of
these
services,
and
it's
specify
each
one
specifies
a
base.
Image
and
Mike's
does
not
have
to
be
running
the
same
base
image
as
billing.
For
example.
Let
me
take
a
minute.
B
Over
here
and
I'll
pull
one
up
with
with
dev
spaces
if
you're
starting
from
scratch,
if
you
know
nothing
about
docker
files,
we
can
actually
scaffold
a
basic
docker
file
for
you,
and
so
you
don't
have
to
know
the
syntax.
But
it's
it's
not
too
hard
to
understand.
This
is
what
we
call
a
multi-stage
docker
file,
so
you'll
see.
That
means
that
there's
multiple
from
statements
here
and
just
looking
at
this
first
one.
This
is
the
base
image
tag
or
the
the
image
tag.
That's
used
again,
as
we
were
talking
about
earlier.
B
This
one
specifies
has
a
version
number
in
it.
So
it's
a
ASP
nut
core
2.1
runs
image.
That
means
that
well
with
2.2
out
there
you're
not
going
to
get
2.2,
it's
not
going
to
magically
update
to
a
later
version
of
the
runtime
for
you,
and
so
this
says
that
the
application
is
going
to
be
in
this
app
directory,
we're
going
to
expose
this
application
on
port
80,
which
is
very
standard.
Your
standard
HTTP
port,
but
you
can
certainly
we've
been
talking
a
lot
about
HTTP.
B
It's
probably
worth
mentioning
that
that
containers
do
not
just
have
to
be
web
applications.
You
can
run
all
sorts
of
different
applications
and
containers.
You
know
with
the
sample
application,
we
run
a
database
and
also
a
sequel
database
in
containers,
not
usually
a
pattern
you
want
for
production,
but
it's
fantastic
for
development,
because
you
can
just
spin
up
this
container.
You
have
everything
that
you
need
right
in
that
container
sequel's
already
installed,
Monaco's
already
installed,
whatever
you
need
and
then
you're
up
and
running.
B
A
I
think
what
you're?
What
you're
telling
me
is
that
a
docker
file
is
the
installation
/
configuration
file
that
builds
my
final
container
is
that
so
so
we
start
with
a
base
image
which
is
gonna.
Your
case
have
asp
net
core
one
in
and
I
think
it's
re
screams
a
little
bit
small
over
there
and
then
the
right
basically
series
of
steps.
Then
what
I
have
the
ability
is
I
have
the
ability
to
customize
it
in
any
way
that
I
want.
A
A
B
Right
so
obviously
you
know
in
this
this
particular
taco
truck
trailer
sort
of
the
one
thing
that
you're
almost
always
going
to
be
doing
unless
you're,
using
something
like
a
sequel
base.
Image
is
copying
your
application
in
and
probably
building
in
this
case,
building
your
application
within
that
container.
So
that's
a
very
common
thing
to
do,
but,
yes,
there's
there's
a
wide
variety
of
customizations
that
you
can
make
here
too.
So.
A
A
Don't
want
you
mutating
data,
while
you're
testing
your
application,
but
I
could
just
create
a
container
flavor
of
that
or
maybe
what
I
do
is
I
put
mock
data
in
that
it's
gonna
return,
and
that
would
be
as
part
of
that
installation
script.
Maybe
I
just
want
to
rep
rate
representatives
like
JSON
or
something
I
make
it
then
sort
of
okay.
Let
me
take
have
a
different
docker
file
to
build
the
handoff
to
friends
and
family
for
testing
purposes
that
have
valid
scenario.
Absolutely.
B
A
B
A
I'll
just
quick
thing
to
people
who
are
watching
the
stream
or
any
questions
that
you
may
have
release
about
containers,
no
spaces
or
anything
like
that.
While
we
continue
chatting
so,
please
feel
free
to
ask
we'd
love
to
answer
them
while
we're
talking
so
going
back
to
the
dockerfile
a
little
bit.
So
that's
a
standard
right.
It's
not
there's
nothing
specific
or
proprietary
about
doing
this
in
vs
code
or
with
Deb
spaces.
It's
just
an
open,
declarative
standard
and
defined
by
docker.
A
B
Definitely
not
editors
specific
I
mean
just
a
text
file.
If
VI
is
your
floats,
your
boat
for
editing,
you
know,
go
for
it,
there's
nothing
Microsoft,
specific
here
and
and
yeah.
If
you
have
any
questions
about
how
do
I
do
this
or
that
then
there's
definitely
information
on
the
internet
and
and
dev
spaces
to
utilizes
these
these
standard
files,
so
you're
not
you're,
not
doing
something
that
is
not
going
to
translate.
B
If
you
know
you,
you
decide
to
move
to
two
on-prem
kubernetes
or
something
like
that
or
you
know
you're
running
a
mini
coop,
maybe
maybe,
if
your
local
development
and
then
you
want
to
use
to
have
spaces
at
a
future
step,
it's
not
like
you're,
coming
up
with
a
whole
lot
of
totally
different
files.
There's
one
file
that
dev
spaces
creates
that's
its
configuration
file.
It's
this
ACDs
demo
file,
but
everything
else
is
really
just
standard.
B
A
B
We're
very
sad,
that's
a
great
question
James,
because
a
lot
of
people
are
in
that
phase,
where
they're
just
getting
their
feet,
wet
and
figuring
things
out
and
what
I
would
say
is
we
really
we've
been
focused
on
dev
spaces
and
this
particular
talk,
but
there's
a
wide
variety
of
containers
out
there
and
jumping
with
deaf
spaces.
If
you're,
just
getting
started
to
be
honest,
is,
is
probably
not
or
may
not
be.
The
the
best
thing
to
do.
B
I
think
the
one
of
the
big
one
of
the
benefits
of
dev
spaces,
if
you're
just
getting
started,
is
that
it
does
make
it
ridiculously
easy
to
deploy
to
a
kubernetes
cluster
I
mean
literally,
you
saw
with
one
click
of
a
button.
My
code
went
up,
it
was
built,
it
was
running.
Kubernetes
I
did
not
have
to
know
any
of
those
concepts.
So
if
you
have
that
need,
dev
spaces
can
be
great,
but
if
you're
really
just
getting
started
with
containers,
I
would
suggest
just
going
local
we've
got
great,
there's
great
extensions
for
PS
code.
B
We've
got
in
Visual
Studio
big
Visual
Studio.
We've
got
our
container
tooling.
That
will
scaffold
the
basic
docker
file
for
you.
You
can
build
that
docker
image
locally.
You
can
run
it
locally
and
so
on,
and
sometimes
that's
a
little
easier
than
throwing
together,
docker
plus
kubernetes
plus
ashore,
plus
all
of
these
concepts
I'm.
A
B
B
In
fact,
that's
literally
all
that
I
have
installed
and
the
the
docker
tools
come
with
that
the
up
spaces
tooling
comes
with
that
as
well,
so
I'm
going
to
go
in
I'm,
going
to
create
a
new
asp.net
core
web
application
and
oh
call
it
stand-up,
and
all
of
this
is
just
standard
standard,
asp.net
core
stuff
and
when
I
get
to
this
screen,
I
have
this
lovable
checkbox
that
says
enable
docker
support.
I'm.
Actually,
I
can't
check
this,
but
I'm
not
gonna.
B
Do
that
at
this
point,
I'm
gonna
go
ahead
and
create
my
my
project
just
to
show
you
another
way
to
to
add
docker
support
project.
But
if
you
know
from
the
start,
you
know
you're
creating
a
brand
new
project.
You
know
from
the
start
that
you
you
want
to
docker
eyes
it.
That
is
a
great
way
to
go,
so
it's
gonna
create
the
project
here.
B
Yeah,
thank
you.
I
do
have
that
running
done
here.
You
can
see
docker
desktop.
That
is
a
requirement
for
playing
around
with
containers
locally.
You
can
use
the
the
community
edition
of
docker
desktop
I
think
it's
still
called
community
edition.
It's
completely
free
and
but
you
do
need
to
be
running
back
yes,
so
I'm
gonna
go
to
add
and
then
I
have
the
option
to
add
docker
support
here.
B
It
asks
me
what
OS
I
want
to
target
again.
This
needs
to
talking
about.
You
really
do
need
to
match
your
host
OS
with
your
with
your
guest
OS
for
containers.
In
the
case
of
doctor
desktop
it
does
have
basically
a
Linux
VM.
It
runs
behind
the
scenes
to
enable
running
Linux
containers,
even
though
obviously
I'm
on
Windows
here
so
I'm,
going
to
go
ahead
and
do
Linux
and
you'll
see
that
it's
gonna
scaffold
a
docker
file
for
me,
and
it's
also
going
to
change
my
debug
launch
target
here.
B
We
know
exactly
yes,
not
even
not
even
the
public
premium
on
the
internal
preview,
so
things
can
can
definitely
go
a
little
little
sideways,
but
it
did
indeed
scaffold
my
docker
file
here.
So
this
is
a
great
way
to
get
started
and
play
around
with
it.
If
I
click
yeah,
that's
clear,
then
what
this
will
do
is
it
will
actually
build
container
image
and
run
that
locally
hook
up
my
devourer
just
like
what
we
saw
with
Deb
spaces.
This
is
completely
contained
on
my
machine,
no
pun
intended
there
so
that
I
can
get
started.
B
A
B
Yeah
I'd
container
Orchestrator
support
and
we
have
support
surprised
the
docker
compose
doesn't
show
up
in
there.
It
should
so.
If
I
go
through
this
flow,
then
I'm
able
to
add
not
just
a
docker
file.
If
it
doesn't
already
exist
but
also
add
helm,
chart
for
this
particular
project,
then
I
can
use
to
easily
deploy
that
to
a
kubernetes
and
any
kubernetes
cluster,
because
helm
is
again
a
kubernetes
standard.
It's
not
a
Microsoft
or
an
azure
thing.
I
know
it
also
adds
files
so
that
I
can
run
this
very
easily
in
dev
spaces.
A
A
Is
I
think
actually
docker
swarm?
Is
the
easiest
place
to
get
started
with
locally
and
I?
Think
if
you're
really
trying
to
get
a
feel
for
learning
about
docker,
the
best
way
to
get
started
is
to
actually
add
the
orchestration
support.
Create
may
be
too
little
services
that
you
want
to
talk
to
each
other
and
then
use
the
extra
file
that
comes
in
called
docker,
compose
the
kind
of
structure
and
met,
and
then
you
start
to
learn.
A
Those
like
hey,
like
I,
can
change
the
ports
that
this
is
exposed
on
the
docker
file
and
I
can
name
this
in
my
compose
file,
which
creates
my
little
mini
cluster
and
now
I
can
see
the
relative
like.
Oh
I,
can
just
you
know,
call
call
Web
API
a
from
our
Web
API
on
project
B.
For
my
project,
a-and
I
could
just
use
HTTP
service
B
or
whatever
and
I.
A
So,
even
even
though
it
word
bugs
you're,
an
internal
review
of
2019
I
think
it's
worth
pointing
out
all
this
stuff
works
on
bills
to
do
2017
as
well,
which
is,
of
course
the
one
in
the
market.
That's
stable,
but
I'm.
Assuming
most
people
have
yeah
so
I
think
that's
how
I
really
learned
I'm
starting
to
feel
get
a
feel
for
pain
or
appreciation,
and
things
like
that
and
that's.
B
Not
wrong
I
think
it's
a
dirty
little
secret
that
most
of
the
people
I
talked
to
who
are
developing
for
kubernetes
are
still
using
docker
compose
locally
if
they're
they're
doing
some
sort
of
orchestration
locally.
For
many
of
the
reasons
that
you
mentioned,
that
it's
very
simple,
it
gives
you
really
it's
that
inner
service
communication
that
that
people
need
for.
For
the
most
part.
B
The
the
experience
that
we
scaffold
initially
is
very,
very
simple:
where
it's
just
a
docker
file,
you
don't
even
have
the
extra
layer
of
compose
once
you're
comfortable
with
that
composes
a
fine
choice
or
when
you
start
wanting
to
to
see
the
magic
of
weight,
I
just
write,
HTTP
my
service
name,
and
it
works
like
what.
What
is?
What
is
that
about?
Compose
really
gives
you
a
good
experience
there.
So.
A
I'm
gonna
make
you
do
one
more
little
plug,
because
this
is
obviously
cloud
tool
stand
up,
so
I
have,
let's
say,
I've
done
what
you've
done
there,
so
we
can
back
out
of
the
add
orchestration
support
and
I
really
would
like
to
test
a
ever
of
this
container.
It
seems
weird
locally
one
of
the
promises
of
containers
that
we
didn't
talk
about
a
lot
in
the
beginning,
but
as
in
theory
whatever
I
remember,
the
COO
they
should
be
exactly
the
same.
Like
should
be
an
immutable
image,
just
like
an
executable.
A
B
B
There
they
can
run
Tanner's,
obviously,
as
the
dab
spaces
p.m.
I'm.
A
huge
fan
of
Azure
community
service
and
I
think
it's
the
best
way
to
run
kubernetes.
But
if
you're
just
getting
started
again,
it
can
be
really
overwhelming.
Thankfully
there
is,
you
can
run
containers
in
app
surface
as
well,
and
that's
a
much
simpler
experience.
I
think
is.
B
To
right
click
and
publish-
and
there
are
all
sorts
of
different
options
here-
I
can
publish-
we
were
talking
about
container
registries
earlier
and
I
can
publish
directly
to
a
container
registry.
That
would
be
probably
a
little
bit
further
down
the
line
when
I've
got
my
code
running
well
and
I'm,
ready
to
really
package
that
up
on
the
way
to
production.
But
a
really
great
way
to
just
test
and
get
started
is
to
do
the
app
service,
linic
option,
Linux
option
and
I
can
publish
their
it
offers
I
can
write
in
Visual,
Studio
I.
B
A
And
then
I
think,
if
we
just
back
out
of
this
real
quick
I,
think
that
kind
of
the
last
aspect
of
that
then
would
be
going
back
to
this.
Is
you
mentioned
if
I
wanted
the
trike,
whatever
I
wanted
to
try
as
your
computer
instances
a
CI
the
reason
we
have
the
container
registry?
There
is
once
I
published
into
that
as
your
container
register.
A
B
A
A
The
only
difference
is,
I
created
my
app,
I
published
it
to
container
registry,
and
then
you
just
need
the
registry
URL,
which
we
actually
put
on
the
summary
page
after
you're
done,
and
then
you
just
paste
that
into
the
CLI
and
then
you
it
was
actually
able
to
go.
Try
aks
from
visual
studio
with
just
visual
studio
and
the
azure
CLI
exactly.
B
Yeah,
I
think
of
container
registries
is
like
the
universal
donor.
So
what
stuff
is
up
there?
It
can
go
to
just
about
anywhere.
You
want
that.
That's
really
just
a
real
standard
and
docker
hub
after
container
registry.
They
all
have
ways
to
or
any
any
service
that
utilizes
containers
it's
going
to
have
ways
to
hold
those
container
images
from
from
one
of
those
registries
that
so
from
Visual
Studio.
You
can
get
your
code
up
to
that
registry
and
then
it's
really
easy
to
pull
it
in
wherever
you
want.
Alright.
A
B
About
PS
2019,
as
this
new
views
2019
and
yes,
it
is
but
I
everything
we
have
showed
is
available
in
the
new
vs
2019,
which
is
out.
You
know
you
can
get
the
release
candidate
download
that
if
you
want
to
play
around
with
it,
but
there
it's
also
all
available
in
Visual,
Studio
2017.
So
as
long
as
you're
on
a
reasonably
recent
builder,
be
as
2017.
You've
got
access
to
all
this
as
well
and.
A
I
think
yeah
and
I'll
just
kind
of
do
the
last
little
plug.
So
the
question
the
flow
for,
because
we
have
a
nice
thing
up
here.
That
shows
us
streams
comments
from
different
sources
if
you're
not
seeing
it.
Somebody
said,
is
this:
the
new
27th
2019
James
said
yeah.
It
is
and
then
you
black
TV,
you
said
sweet.
When
is
it
leaving
preview?
We
have
publicly
announces,
so
you
can
already
go
sign
up
for
to
watch
our
keynote
and
live
stream
events
that
correspond
with
be
released,
2019
and
that
will
be
on
Tuesday.
A
So
this
coming
Tuesday,
April,
2nd
it'll
show
your
2019
will
go.
Rtw
it'll
be
GA
and
they're
an
entire
series
of
events
that
day,
starting
with
a
great
keynote
by
Scott
Hanselman.
Where
he's
gonna
give
you
a
great
behind-the-scenes,
look
at
the
team
that
built
it
that
plugged
and
then
we
will
there
be
a
series
of
live
streams
all
day
talking
about
various
new
things
in
2019.
A
So
if
you
want
a
deeper
dive
on
things
like
the
debugger
or
the
cloud
tools
or
live,
share
or
Intelli
code
or
how
to
be
more
productive,
developing
applications,
there
will
be
half
an
hour
sessions
that
you
can
tune
into
all
day.
Last
thing
just
a
reminder:
we
do
these
stand-ups,
every
Tuesday
and
Thursday,
so
Tuesdays
are
I.
Believe
asp.net.
Stand-Ups
Thursday's
are
a
rotating
schedule
of
dotnet
things.
We
do
cloud
that
was
this
week.
We
do
mobile,
we
do
dotnet
to
lean
and
we
do
net
run
time
so
be
sure
to
subscribe.