►
From YouTube: Kubernetes Cluster API Azure 2019-11-07
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
we
are
recording
hi
everyone.
This
is
the
November
7th
edition
of
the
cluster
API,
a
sure
provider
meeting.
This
is
a
meeting
that
is
recorded
and
available
on
the
internet,
so
please
be
our
will
be
available
on
the
internet.
So
please
be
mindful
of
what
you
say.
What
you
do
please
be
sure
to
adhere
to
the
kubernetes
code
of
conduct
and
just
in
general,
be
awesome
people.
So
we've
got
light
agenda,
light
agenda,
we're
going
to
go
through
the
project
board.
A
Yeah,
let's
well,
let's
do
let's
do
one
announcement
first,
so
we
have
Cecile
has
been
working
incredibly
hard
over
the
last
month,
plus
versatile
and
the
v1
alpha
to
work
working
on
image,
building
and
a
whole
bunch
of
stuff
to
make
the
project
better.
So
we
are
going
to
be
graduating
seal
to
an
approver
for
the
repo
I
just
approved
the
PR,
and
there
are
a
bunch
of
other
pr's
that
we
have
to
to
do
but
Cecile.
Thank
you.
Thank
you.
Thank
you.
Thank
you
for
your
work.
B
Okay,
so
David
and
I
have
been
working
really
hard
fighting
with
the
marketplace
team.
Fighting,
sorry,
collaborating
or
different,
collaborating,
get
our
image
into
first
party
mode
and
we
finally
succeeded,
so
it
got
merged.
So
now,
users
don't
need
to
accept
legal
terms
anymore
when
they
use
the
image
for
the
first
time,
which
will
make
the
user
experience
I
want
friend
render
and
then
so
I
added
my
source
tags,
so
that
was
kind
of
a
prerequisite
or
bring
your
own
v-necks
and
bring
your
own
resource
group.
B
B
Sorry,
we
ran
a
bunch
of
people
are
in
the
same
room,
so
we
have
a
quick
start
duck
in
the
cluster
API
book
or
a
journal.
So
please
go
try
that
the
PR
just
burst
this
morning.
Let
me
know
if
you
have
any
feedback
and
I
think
that's
it's
unless
I'm
forgetting
anything
but
yeah.
I
would
like
to
talk
about
the
release,
because
I
think
we're
pretty
much
there.
Yes,.
B
So
that
I
want
to
work
on
automating
it
because
right
now
it's
not
to
be
honest,
but
the
plan
is
that
we
have
pipeline
that
we
can
just
press
the
button
and
it
does
it
end
to
end,
and
actually
david
has
been
working
on
a
tool.
I
don't
you
want
to
talk
about
that,
but
to
automate
publishing,
do
you
want
to
talk
about
on
my
micro
site?
I'll,
just
look
it.
B
Okay,
yeah
sounds
good,
so
yeah,
so
we'll
be
we'll
be
using
a
pipeline
so
that
we
can
just
like
do
all
the
building
and
all
the
publishing
automate
automated
and
then
the
time
like
how
long
it
will
take
will
depend
on
how
long
it
takes
to
replicate
the
image
legally
which
isn't
under
our
control,
but
at
least
from
our
side
we'll
try
to
optimize
it
as
much
as
possible.
Okay,.
A
Cool,
so
with
regards
to
permissions,
who
will
be
able
to
do
this.
B
So
right
now
I
think.
Well,
we
know
the
pipeline
does
exist,
but
right
now,
I
think
we
need
to
have
access
subscription,
which
is
sponsored
by
Microsoft's
the
sense
yep
subscription.
So
I
think
you
you
have
access
to
even
and
yeah,
but
not
anyone
will
be
able
to
you
have
to
be.
You
have
to
have
like
credentials
for
the
subscription
to
be
able
to
build
the
images,
and
then
you
also
need
to
be
well.
We
didn't
talk
about
that
later.
I
think.
C
A
A
Yeah
so
I
mean
as
much
as
possible,
I
to
start
just
having
community
members
able
to
do
this
as
well.
So,
like
me
as
a
community
right
like
that's
enough
for
now,
as
we
move
forward
with
the
image
building,
we
should
be
able
to
do.
Do
it
very
similarly
to
the
way
that
we
do
the
container
image
building
right.
So
something
merges
in
the
you
know
something
merges
in
the
repo
a
pre
cement
kicks
off
a
job
right
so
that
that
job
runs
on
GCB.
D
A
Great
cool
cool
cool,
so
release
yes,
asked
about
the
release.
I
Oh
docks
at
this
point
and
I
think
that's
the
last
thing
that
we
need
to
do
before
kicking
out
the
release.
I
have
been
on
some
cube.
Con
driven
development
for
release
engineering
in
cig
release,
so
I
haven't
had
as
much
time
to
work
on
the
cap
C
stuff,
but
now
that
you've
merged
the
QuickStart
and
all
that
stuff
it
should
be.
A
A
E
E
E
A
Yeah
so
branching
wise,
we're
we're
assuming
that
master
right
now
is
for
you
an
alpha
2
right
all
right.
So
once
it
gets
to
the
point
that
we
actually
cut
the
real
V
1
alpha
2,
so
we've
cut
like
an
alpha
0
or
alpha
1.
At
this
point,
once
we
once
we
cut
an
actual
V
1
alpha
2
or
0
3
0
release
well
weaking.
We
can
drop
a
branch
for
release,
0
3
right,
that'll
become
the
V
1
alpha
2
and
then
we'll
move
master
into
V
1
alpha
3
state.
Ok,
it.
A
A
A
No,
it's
it's
all
related
to
release
engineering.
So
what
we've
been
doing
over
the
past
two
cycles
or
so
really
is
so
like
some
of
the
stuff
around
the
container
image
building
some
of
the
stuff
around.
We
just
started
refactoring
the
bash
scripts
that
set
run
all
of
kubernetes
really
into
indigo
tools.
A
Stuff,
like
that,
you
know,
documentation,
building
out
the
team's,
so
a
lot
of
it
has
been
focused
on
documenting
the
the
the
technical
debt
that
we've
been
cleaning
up.
So
I
have
two
talks
that
I'm
doing
around
that.
Well,
it's
like
a
sake,
release
deep
dive
and
then
Tim
and
I
are
talking
about
like
kubernetes
the
kernel
and
which
is
kind
of
like
a
an
answer
to
Tim
Hawken
and
Michael
ribbon
stock
from
like
2017,
where
we
see
how
far
we've
gotten
over
the
last
two
years.
B
A
A
E
E
Are
pretty
minimal,
so
I
could
justify
the
changes
there
and
we
can
just
go,
buy
them
separately.
Exactly.
Is
this
really
like
to
be
honest?
Just
like
cluster
name
is
the
biggest
breaking
change?
Okay,
you
know
I
think
I
can
just
do
that
and
then
yeah
we
just
as
soon
as
that
branch
is
is
available,
I'll
just
rebase,
and
then
we'll
figure
out
what
to
do.
Yeah.
E
A
E
E
A
A
F
B
Yeah,
we
don't
even
have
an
to
end
test
right
now,
sir
touch
like
that's
an
open
issue,.
B
F
B
B
A
A
A
E
A
E
A
A
A
B
One,
but
not
in
this
repo,
because
it
was
multi
repo
but
yeah
I-
can
create
one
and
add
the
details
of
what
I'm
doing.
Okay.
A
A
Alright,
the
release
engineering,
tooling
I
just
need
to
do
it
right
up
on
that
stuff.
Honestly,
that's
you
know
that's
between
the
image
folder
and
the
GC,
the
GC
builder,
whatever
we're
calling
that
thing
now,
there
are
two
image:
folders,
there's
one
that
actually
does
the
you
know
the
a.m.
eyes
or
whatever
image
type
and
then
there's
also
the
container
image
folder
that
lives
in
intestine
4.
So
just
documenting
that
stuff,
this
kind
of
existed
before
the
time
that
we
had
the
repo
wired
up
correctly,
like
what
you
know,
reporting
status
into
prowl
and
what-have-you.
A
A
A
D
A
A
100
percent
yeah
thanks
to
yeah,
so
it's
so
it's
really
a
question
of
like
do.
We
want
to
keep
doing
that.
I.
Think
I.
Think
what's
happened,
you
know
thus
far.
Is
there
have
been
more
important
fish
to
fry?
Really
right.
So
yes,
I
mean
this
is
an
an
eventually
thing.
Maybe
we
can
maybe
we
fix
it
at
the
tail
end
of
fee
1,
alpha
2
or
you
know
it's.
It
feels
like
a.
It
feels
like
a
next
thing
to
me.
Yeah.
E
A
E
E
They've
been
pretty
severe,
and
so
you
know
really
we're
really
looking
at
you
know
for
ideas
on
how
we
can
you
know
deal
with
these
kinds
of
problems,
and
one
interesting
proposal
is
to,
instead
of
all
the
things
you
know
and
kind
of
related
to
kubernetes
calling
the
address
you
keep
or
go
directly.
You
know
we
have
a
CRD
for
that
resource
and
all
of
those
things
use.
E
You
know
that
clients
and
call
you
know
that
CRT,
then
we
can
use
that
as
a
you
know
as
a
point
to
do
all
of
the
event,
well,
that
we
call
it
rate
limiting
mm-hm
and
even
do
things
like
you
know,
merging.
You
know
multiple
requests
into
a
single
request
and
things
of
that
sort.
So
just
kind
of
thinking
out
loud
here,
so
anyways
yeah.
A
E
Imagine
a
situation
where
even
the
some
of
the
cloud
provider
resources
instant.
You
know
that
the
cloud
provider
itself
calls
this
C
or
D.
Instead
of
calling
the
you
know
the
resource
directly,
and
so
you
know
where
you
know
where
the
cloud
provider
has
to
you
know,
go
and
muck
with
an
assess
or
a
little
Beltzer,
where
these
those
various
clients
will
compete.
You
know
for
overriding
the
other
I
guess
yeah.
A
A
A
We
cut
that
you
will
ye
well,
yes,
once
we
will
kind
of
release
branch
first
and
then
once
we
cut
the
release,
that'll
be
that'll,
be
true:
oh
cool,
yeah,
I.
Don't
think
it
I,
don't
think,
there's
much
value
enough.
Staying
on
on
B
1,
alpha
2,
since
since
everyone
else
is
moving
towards
v1l
for
3
all
right.
A
A
A
E
A
D
B
A
B
A
F
A
Let's,
let's
punt
it
even
further,
honestly,
let's
let's
try
it
if
it's
possible,
let's
see
if
we
can
have
something
kind
of
informal
at
cube
con.
If
people
are
there
and
then
target
to
try
to
schedule
something
after
cube
con,
because,
honestly,
it's
not
just
the
117
release.
It's
everyone
is
everyone's
kind
of
in
cube
con
driven
development
mode
right,
so
it
makes
sense
yeah.
It
makes
sense
to
pick
up
later.
A
A
F
A
B
B
B
A
A
B
A
A
You
no
thank
you.
I've
got
a
bunch
of
PRS
to
do
for
your
access,
because
it's
like
the
image
feel
of
the
image
access
on
I
know
the
GCP
permissions,
the
some
of
the
teams
on
kubernetes
SIG's
to
make
sure
that
you
have
like
admin
access
to
the
repo
and
stuff
like
that.
So
I'll
work
through
that
by
early
next
week,
okay,
sweet,
sweet!
All
right!
Thank
you!
Everyone
have
a
great
day.