►
From YouTube: Kubernetes SIG Cluster Lifecycle - AWS API 20190107
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
B
So
we
started
talking
about
this
in
a
separate
conversation,
but
I
think
it's
community
driven,
it
should
be
I
think
it
should
be
a
German.
There
are
a
set
of
tests
that
we
would
like
to
have
available.
The
question
is:
what
would
we
like
to
get
done
before
the
MVP
there's
a
couple
of
different
issues?
I
think
they're
all
assigned
to
Liz,
Curley
and
just
sort
of
enumerated.
What
we
would
like
to
do
for
this.
You
know
for
the
v1
alpha
one
release,
so
we
can
get
those
things
in
place.
B
C
B
B
B
B
And
we
wrote
down
some
notes
here,
but
the
question
is:
what
type
of
verification
do
we
want
to
do
on
a
per
PR
basis,
and
what's
the
cut
line
that
we
want
to
set
folks
truck
against
to
get
that
done?
So
the
one
thing
we
had
the
internal
conversation
on
was
to
get
very
basic
verification
up
like
a
single
smoke
test,
a
single
end-to-end
test
that
basically,
is
the
equivalent
of
suitably
quick,
which
basically
verifies
that
you
have
all
the
work
basic
working
pieces
of
a
cluster.
B
B
C
B
The
question
is
whether
or
not
you'll
you'll
have
breaking
things
on
a
per
PR
basis,
which
you
probably
will
like.
It
really
needs
people
to
look
at
it
and
dive
deep
on
it
and
that
are
going
to
devote
themselves
to
making
sure
that
when
a
Peter
gets
put
up
that
you
know.
If
we
get
to
the
point
where
on
v1
alpha
one,
we
feel
comfortable
actually
releasing,
because
we've
gone
through
enough
test
coverage
to
make
us
feel
happy
that
the
automation
is
working.
B
D
I
think
I
want
I.
Think
the
the
thing
that
we
need
right
now
is
more
information
on
how
long
these
texts
are
gonna.
Take
like
I
think
that
falls
on
a
boy
conformance
run
is
gonna.
Take
too
long
pretty
much,
regardless
of
how
we
slice
it
for
a
PR,
but
I,
don't
know
how
long
it's
going
to
take
to
stand
up.
Eh
like
how
long
does
it
take
to
stand
up
a
standard
awf
share,
what
kind
of
flake
rates
are
we
getting
on
that
and
I
think
we
just
need
more
data
there?
Okay,.
B
Well,
I
think
this
is
enough
to
get
us
rolling
and
we
can
always
come
back
to
it
later
on
publishing
of
artifacts.
Obviously
we're
gonna
have
to
get
to
that,
but
I
think
that's
a
later
date.
I
think
these
two
are
the
highest
priority
ones
right
now:
blocking
PR
jobs
and
periodic
full
conformance
suite
tests,
I
think
are
of
the
highest
in
the
p0
range
they're.
The
highest
of
the
pz
rose.
Do.
A
D
Made
some
incremental
progress
towards
prowl
being
able
to
handle
tag
pushes,
but
if
we
want
to
use
that
mechanism
we're
going
to
meet
some
of
these
a
champion,
it
I
think
that
would
be
the
ideal
because
it
you
know
for
a
brand
new
product
I,
don't
think
we
want
to
be
standing
up
more.
This
has
to
run
on
some
specific
person's
laptop,
but
that
is
not
something
we're
going
to
get
for
free
with
power
right
now,.
B
D
D
B
D
A
Yeah,
the
biggest
thing
for
me
is
that
I
think
we
really
need
to
have
the
repository
that
were
publishing
images
into
owned
by
kubernetes
or
the
CNC
F,
rather
than
an
individual
company,
because
we're
already
damages
we're
publishing,
right
now
or
in
a
hefty,
owe
controlled
account
and
there's
a
little
question
about.
How
do
we
is
it
possible
to
transfer
that
somewhere
or
not
so
the
quicker
we
can
get
over
to
a
CN
CF
owned
account,
I?
Think
the
better.
B
A
E
Cool
so
I,
don't
know
what
was
going
on
anyway,
so
I've
been
making
some
progress,
but
not
as
much
as
I
thought.
I
should
have
finished
I'm.
Currently,
at
this
stage,
where
the
cube
Adm
joined
for
the
control
plane
is
the
documentation
for
that
is
a
little
unfamiliar
or
like
not
insufficient.
I
would
say,
especially
around
the
control
frame
stuff,
so
trying
to
figure
that
out
to
make
progress.
I
have
a
list
of
things
that
are
currently
not
working
in
the
PR,
like
as
a
comment,
so
I'm
kind
of
working.
E
My
way
down
that
list,
I
I,
also
started
a
I
just
posted
in
the
cube,
a
DM
channel,
as
well
as
asking
for
help
so
once
I.
Get
that
help
that
I
needed
or
I
need.
I
should
be
make
a
bulimic
more
progress
for
folks
who
are
interested
or
want
to
kind
of
give
me
a
quick
review,
more
kind
of
see
where
what
I'm
talking
about
you
can
head
over
to
the
PRM
link
to
PR
and
also
the
issue
in
the
agenda,
doc
that
we
have
that's,
that's
pretty
much.
A
E
E
A
E
I,
while
I
was
working
my
way
on
the
cube,
ATM
stuff
I
realized
that
we
don't
the
cube
idiom
token
create
to
create
tokens
for
joining
nodes
in
the
cluster
I
wanted
to
see
if
what
the
folks
think
about
using
switching
over
to
cube
a
DM
to
generate
tokens
instead
of
the
cluster
bootstrap
token.
Let
me
use
I,
see
the
GCP
provider
uses
the
cube.
Atm
token
create
stuff
so
on.
Maybe
that's
a
good
idea.
I,
don't
know
what
you
guys
think
so.
A
C
C
So
on
the
link
that
I
pasted
it
in
the
document
is
just
a
couple
of
various
wait:
approaches
to
actually
exposing
cuvee
diem
config,
and
if
people
wanted
to
put
their
thoughts
on
that
issue,
it
would
be
great
to
get
some
other
some
other
opinions
in
there,
but
yeah
that's
pretty
much
it.
So
if
you
have
some,
if
you
have
some
time,
take
a
look
at
that
comment.
A
And
one
of
the
things
that
I'm,
hoping
that
we
can
eventually
do
because
kind
of
veteran
get
a
particular
version
of
cube.
Atm
is
a
little
worrisome
to
me
and
that
it
could
potentially
cause
a
tighter
dependency
than
then
we
really
want
to
accept
so
I
hope.
That's
at
some
point
in
the
future
will
be
able
to
expose
those
types
for
cube
ATM
in
a
way
that
you
don't
have
to
actually
pull
in
all
of
cube,
ATM
and
definitely
not
all
of
the
cake.
Every
boat
I
think.