►
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
Tutorial: Building & Running Kubernetes from Source and Submitting Your First PR - Ian Chakeres, Silvia Xu, Michael Taufen, & Federico Bongiovanni, Google
(Limited Seating Available - See Description for Details)
A
All
right
we're
gonna,
go
ahead
and
get
started
with
the
tutorial
today.
Thank
you
very
much
for
joining
us.
It's
a
packed
house.
There
are
some
more
space
kind
of
around
the
front
here,
as
well
as
you're
looking
for
a
spot,
and
today
we're
going
to
do
a
hands-on
tutorial
about
building
and
running
kubernetes
from
sources
and
submitting
your
first
of
the
art,
but
before
we
do
that,
I
want
to
we're
going
to
introduce
ourselves
talk
a
little
bit
about
what
we're
doing
what
we've
been
doing
with
kubernetes
and
how.
A
A
B
Everyone
I'm
Steven
and
I
walking.
We
go
specifically
in
the
experience
of
Google
managed
to
bananas
and
the
mother
press
experience
Oprah's
on
the
couch
to
see
how
it
improve
the
best
for
an
86
bears
can
go,
and
maybe,
like
many
of
you
like
in
relatively
relatively
new
to
the
always
alternative
communities
community
for
experience
for
me
so
again,
going
to
try
to
help
you,
sir
all
experience
and
adventure
just
like
how
I
did
it
myself,
people
it's
fun.
C
Good
morning
and
I
need
Federico
Ventura
Vani
my
journey
with
parentheses,
interesting
I
was
a
user
of
covering
this
with
my
previous
company.
So
I
learned
a
lot
about.
You
know
good
sandbox
and
the
pain
points,
and
you
know
the
opportunities
and
when
I
had
the
opportunity
of
turning
evening
de
twice
I'm
working
as
an
engineering
manager
on
the
HPI
machinery
team,
which
is
the
team
that
handles
most
of
the
components
that
run
on.
E
Hi
good
morning,
everyone
you'll
have
your
fun
and
so
I
might
talkin
I
worked
at
on
Cooper
days
at
Google
for
about
three
years
now,
mostly
in
cig
node,
an
open
source
so
working
on
the
qubit
and
those
images
and
I've
done
a
bunch
of
these
things.
Around
release
of
my
structure
on
the
ground
and
a
few
other
areas
so
excited
to
be
helping.
You
guys
all
learn
how
to
build
the
Canadian
Centre
for
today.
E
A
Right
well,
I
introduce
myself
I
want
to
please
if
you
have
not
already
follow
the
instructions
at
this
link
or
the
page
that
you
received
start
a
yeah
at
the
console
and
I'm
gonna
tell
you
about
a
little
bit
about
myself
and
why
we
created
this
tutorial.
So
my
name
is
Ian
checkers
I'm,
a
manager,
people
working
on
kubernetes,
but
I
was
not
always
actually.
My
first
touch
with
kubernetes
was
a
tutorial
with
Kelsey
and
2015.
A
Looking
at
kubernetes
when
it
was
first
released
and
I
was
in
another
company
and
we
actually
had
a
lot
of
success
using
kubernetes
for
our
stateless
workloads
and
we're
looking
at.
How
can
we
provide
a
similar
experience
for
our
state
for
work
loads?
And
this
was
in
the
beginning
of
2017
and
kubernetes
for
stateless
workloads
was
not
nearly
as
mature
and
so
I
knew.
I
was
gonna,
have
to
start
contributing,
so
I
started
attending
sick
meetings.
That
I
wrote
some
unit
tests,
I
refactored,
so
cover
the
road
and
to
end
tests.
A
A
Okay,
Janine
mentioned
that
people
are
copying
from
the
PDF
and
the
dash
dash's
for
launching
the
VM
are
not
copying
and
pasting
Pro
properly.
If
you
actually
go
to
the
third
file
on
the
website,
it
contains
the
command
and
what
don't
Penina
good
like
the
copy
and
paste
from
there
and
the
dashes
will
not
disappear.
A
So
the
outline
for
today's
talk,
we
will
get
you
get
at
the
end
ready
and
then
we
will
build.
Kubernetes
resource
will
actually
clone
the
repository
and
then
we'll
start
up
that
and
actually
Sylvia's
gonna
cover
this
part
of
the
session
and
then
the
second
half
Michael
is
going
to
talk
about
forking
the
code,
making
changes
running
the
unit
tests
and
submitting
the
PRA.
A
Now
there
are
a
few
prerequisites,
so
everyone
it
looks
like
has
a
computer
so
that
they
can
log
into
the
cloud
console
to
launch
a
VM
as
well
as
to
open
a
so
sage
session.
To
that
the
other
two
are
primarily
for
the
second
part
which
is
around
submitting
of
the
art.
So
you
need
a
github
account
so
that
you
can
fork
the
code.
You
may
not.
You
have
time
to
do
that,
or
we
have
not
built
in
time
to
do
that.
That's
part
of
the
second!
A
B
A
These
these
also,
they
prevent
you
from
some
mating
of
the
art
but
I
think
you'll
still
get
process,
so
the
full
instructions.
Many
of
you
received
a
sheet
when
you
walked
in
those
are
linked
off
of
this
website
as
well.
It
has
instructions
for
starting
at
the
end
and
all
the
instructions
for
the
other
parts
of
the
talk.
A
C
A
Session
we
are
going
to
be
showing
you
on
the
screen,
all
the
different
actions
that
were
expected
to
do
to
take
it.
We're
gonna
start
from
the
point
of
the
VM
being
launched
a
new
having
SSH
that,
but
we're
gonna
go
step
by
step
through
with
the
audience,
and
every
step
will
kind
of
pause
and
get
a
gauge
of
how
everyone
is
doing.
We
really
want
to
try
and
keep
people
in
sync,
so
periodically
I'll
be
asking
people
or
you
know
if
they
are
with
us
and
have
completely
step.
Why
not?
A
So,
please
do
raise
your
hand
when
I
request
that
so
that
we
can
set
the
appropriate
pace
for
some
of
you.
This
may
appear
to
go
extremely
slowly
and
those
of
you
who
maybe
run
into
issues.
This
is
going
to
feel
like
we're
going
really
quickly,
but
we're
gonna
try
and
keep
the
pace
appropriate
so
that
we
give
the
audience
through
the
session
and
the
key
is
we
want
you
to
walk
out
of
here,
having
built
and
run
your
own,
the
kubernetes
and
then
I'm
also
knowing
how
to
submit
your
first
PR.
A
D
D
D
B
Of
how
to
create
a
VM,
so
I'm
going
to
walk
through
the
UI
or
flow
to
how
to
create
a
VM,
and
you
can
follow
mr.
net.
It's
basically
a
set
of
outcome
upwards
of
the
classroom,
but
if
you
know
probably
smarter
for
you
to
do
it
in
your
eye
power,
digital
recession,
you
couldn't
do
it
in
culture.
Right
now,.
D
D
B
C
B
D
D
B
So
for
the
UI
for
the
operation,
here
you
want
to
create
an
instance
in
the
disease,
so
we
recommend
it
just
use
the
u.s.
press,
one
zone
which
is
horrible,
pretty
close
to
us
now.
So
your
have
added
Network
magazines
for
SSH
to
be
a
messin
with
tons
of
available
here
and
for
Shippuden
recording
is
we
do
require
certain
requirements.
The
minimum
is
the
standard,
808
CPUs
and
certificate
by
the
memory.
Welcome.
A
B
B
D
B
D
D
E
A
D
D
B
D
D
B
D
B
A
D
A
D
A
D
A
All
right
just
a
reminder
for
some
people
who
may
be
having
problems
finding
the
SSH
to
enter
the
VM.
That
they've
created
make
sure
you've
selected
your
project
on
that
top
line
project
in
your
account
and
if
you're
having
a
problem
with
your
account.
Please
raise
your
hand.
We
have
other
accounts,
we
can
hand
you
a
different
one
and
I'm
gonna
get
started
on
the.
A
D
B
So
one
is
added
a
new
pathway,
is
it
go
exactly
pass
Java
local
provider
and
we
are
good
to
get.
This
is
subtraction
the
go
past
ka
is
the
seduction
welcome
to
clone
the
kubernetes
software
hope
so
are
going
should
angel
by
the
way
helped
organize.
The
source
code
is
usually
that
into
home
go
SRC,
and
this
is
organization-
and
this
is
a
big
repository.
We're
going
to
have
favored.
So
a
boomsticks
for
the
pastor,
neutral
tree
shows
the
best
provides
value
and.
B
B
D
B
A
B
So
what
I
would
recommend
doing
to
close
your
posture
now
and
we're
longing,
but
think
so
because
to
do,
though,
is
we
added
in
research
to
the
new
super?
So
these
is
really
to
take
exactly
strangers.
Talk
about
it,
okay,
so
just
close
across
Europe.
Are
you
partying
and
click
on
exercise
again
so
intersession.
D
D
B
A
How
many
people
have
are
with
us
here?
They've
started
their
main
quick
release.
Please
raise
your
hands.
Alright,
anyone
who
has
not
started
make
quick
release
who
needs
help.
Please
raise
your
hand.
This
is
going
to
be
this
process.
Gonna
take
about
ten
minutes,
so
we
can
get
everybody
back
up
to
date.
Anyone
with
a
problem
raise
your
hand.
Please.
D
D
A
D
D
D
D
B
A
D
A
A
D
E
So
we
mention
this
briefly
at
the
beginning
of
the
session
before
so.
If
you
send
a
PR
before
it
can
be
merged
into
the
upstream
repo,
you
need
to
have
done
these
two
steps.
Maybe
there
Brendon
agreed
to
the
code
of
conduct
and
you
need
to
assign
the
CLA
if
you
haven't
done
that,
yet
that's
okay,
you
can
still
follow
along
with
all
of
these
instructions
and
the
robots
on
the
upstream
repo
will
just
tell
you.
E
D
D
E
A
D
A
A
E
So
now
we're
going
to
go
into
the
basics
of
making
a
change,
setting
it
up
strings.
So
the
first
thing
I'm
going
to
do
is
I'm
going
and
check
out
a
master
branch
from
the
upstream
of
my
boat.
That
is
we're
all
about
these
different
entities
in
development
and
it's
a
good
idea
to
transfer
them.
There.
E
E
A
Gonna
mention
something
my
branch
or
patatas,
so
your
mo
best
practices
that
branch
we
have
some
meaning
for
what
you're
going
to
accomplish
within
that
little
branch
and
Francisquito
branch.
As
an
example,
you'll
don't
describe
to
feature
there
so
that
people
can
understand
and
you
yourself
can
figure
out
later.
What
is
this
particular
branch.
D
E
A
D
E
E
D
E
E
A
A
E
E
A
It
just
Nancy
had
to
do
that.
L
mentioned
something
about
these
commands.
The
user
commands
here
so
the
way
that
the
robots
accounts
work
for
kubernetes.
They
leverage
this
metadata
to
know
that
you
have
signed
the
CLA
and
agreed
to
this
ence
F
code
of
conduct.
So
it's
very
important
to
can
put
these
before
you
do
your
commit
otherwise
I
won't
be
recognized
on
github.
E
D
D
A
E
Okay,
so
this
is
the
long
way
to
create
a
pull
request,
sometimes
github.
What
once
you've
recently
push
the
branch
will
give
you
a
quicker
way
to
do
it.
We're
gonna
show
you
the
long
way,
because
that
doesn't
always
show
up.
If
that
does
show
up,
you
get
a
shortcut
button
and
it's
really
obvious
that
it's
just
this
long,
yellow
line
across
the
top
of
the
page
with
the
green
button.
E
E
This
view
is
scoped
just
to
compare
two
branches
in
the
same
repository,
but
since
your
changes
are
over
in
your
fork,
you've
got
to
click
this
link
that
says
compare
across
forks
once
you
mean
that
it'll
give
you
this
fancier
view
that
compares
the
once
you're
here.
There's
these
two
drop-down
boxes
on
the
right
hand,
side
of
this
menu
here
and
in
the
first
one
you
want
to
make
sure
you
select
in
New
York
for
that
Connecticut.
So
in
this
example,
I'm
selecting
a
McLaughlin,
/
kubernetes
and
then
in
the
second
box.
E
You
want
to
select
the
branch
that
you
make
it
so
Ian
approve
my
desire
to
call
my
rapeseed
networth
change,
and
so
it
says,
see
the
important
change
here
and
that's
what
I'm
selecting
for
this
example
and
once
you've
done,
that
it
will
give
you
a
preview.
If
you
scroll
down,
you
can
kind
of
see
your
commits
and
all
your
changes.
E
E
That
will
contain
all
the
important
information
about
your
pull
requests
that
are
trying
to
communicate
to
the
community
when
they're
going
to
come
up
with.
So
it's
really
important
to
follow
the
instructions
in
this
template
on
this
menu
and
I'm
gonna
switch
over
to
the
readme,
which
has
just
a
copypasta.
E
So
this
is
an
example
of
a
template.
There
are
few
recommendations
and
the
template
it
just
to
read
through
and
consumers
time.
So
a
lot
of
the
stuff.
The
news
recommendation
they've
also
talked
about
an
accession,
but
if
you're
really
interested
in
diving
in,
we
can
sorry
through
that.
Also
there
are
links
to
the
contributor
guidelines
in
there
there's
links
to
a
developing
guide
which
can
help
you
in
some
additional
cooling
and
environment
setup.
E
There
is
some
information
about
how
to
which
labels
should
be
applied
to
the
pull
request.
I'll
get
to
that
in
a
second
there,
yeah
there's
some
information
on
testing
some
information
on
best
practices.
How
to
write
a
release.
Note,
if
you
have
the
user
facing
a
change
and
also,
if
you're,
just
sending
a
PR,
that's
not
condition
yet,
and
you
just
want
some
people
to
look
at
it.
Some
additional
things
you
can
do
to
label
a
difference
that
so
back
to
the
labels.
We
have
a
categorized.
E
And
so
we
don't
just
give
everybody
permission
to
have
another
bangles.
They
want
all.
The
labels
are
added
and
removed
and
managed
through
what
we
call
problems.
There's
a
system
called
prowl
that
runs
all
of
our
test
infrastructure.
It's
actually
a
kubernetes
cluster,
with
this
process
and
layered
on
top
and
the
way
you
interact
with
it
is
via
the
prowl
box,
which
is
a
github.
E
E
E
E
E
Finally,
this
is
important
to
get
right.
Is
there
a
release,
notes,
section
and
I'm
just
writing
tests
and
don't
you
know,
release
note,
but
you
do
need
to
type
of
words.
An
onion
in
between
be
roughly
1/3
inventx
here
or
the
box
will
say
that
you
just
make
sure
that
everybody
is
completely
explicit
about
whether
or
not
they.
A
A
We're
going
to
just
in
the
last
ten
minutes
we
can
help
those
people
who
are
who
who
are
still
struggling.
We
did
want
to
kind
of
all
they're
here.
You
know
become
a
part
of
this,
exploit
that
and
maybe
they'll
be
like
Saturday,
oh
okay,
when
we're
successful
in
building
your
own,
your
own
building
community
running
it
and
then
understanding
solubility
are
like
I
started,
telling
the
city's
right
easing
test.
Revent
Isaac's
become
a
part
of
this
community
and
there
is
a
bunch
of
great
documentation
on
the
official
website.
A
So
you
want
to
thank
everybody
for
coming
today
and
I.
Think
anyone
who
is
interested.
We
have
some
$20
credits
that
you
can
apply
to
your
own
Google
cloud
account
just
stop
by
on
your
way
out
or
the
people
in
the
university
upon
shirts,
and
they
can
give
to
a
cloud
press
to
really
do
need
help
in
this
racing
in
basketball.
So
can
you
give
a
round
of
applause
for
the
volunteer.