►
Description
Lost the first 10 minutes or so due to a zoom crash :/
A
A
B
B
So
the
things
that
we
need
would
be
like
a
bootstrap
provider
to
run
hcdm,
then
an
actual,
then
a
controller
within
clustered
api
to
manage
the
fcd
cluster
life
cycle
and
also,
I
think,
a
lot
of
changes
within
machine
controller,
and
I
think
february
suggested
that
we
can
like
collaboratively
work
on
the
proposal
and
hackmd
was
the
like.
They
suggested
using
hack,
md
proposal
so
that
everyone
can
access
and
edit
it.
B
So
I
just
wanted
to
make
a
note
here
that
this
is
the
link
to
that
proposal
that
I
have
so
far
and
right
now,
it's
still
not
complete,
like
it's,
not
detailed,
so
everyone's
free
to
access
it
and
comment
on
it
and
because
I
myself
don't
have
a
lot
of
understanding
of
plus
how
clustered
api
works.
Just
for
my
own
sake
to
know
what
all
changes
will
be
needed.
B
I
thought
it
would
be
good
if
I
just
started
working
locally
on,
like
my
own
separate
repo
for
the
hcdm
bootstrap
provider,
you
just
bring
up
a
single
node
cluster,
so
I've
just
linked
that
repository
here
and
again,
it's
just
for
my
own
understanding,
so
based
on
that
I'll
keep
on
adding
things
to
the
proposal
as
well
yeah.
I
just
wanted
to
make
a
note
here.
A
A
C
Yeah,
so
if
you
scroll
up
a
little
bit
there
is
I
updated
the
open
proposals,
section
404
somewhere
in
there
yeah,
so
we
have
a
lot
going
on
and
I'm
so
like.
We
are
past
phrase
right
now
so,
like
I
would
say,
we're
probably
not
gonna
accept
new
proposals
at
this
point,
but
like
one
thing
to
like
popped
up
here,
it's
like
there's
a
lot
of
work
to
do
so.
I've
chatted
with.
C
C
C
This
is
blocking
for
zero
four,
because
all
influential
providers
will
need
to
implement
this.
I
don't
know
how
we
actually
gonna
check
for
conformance
here
because,
like
the
annotation
has
to
be
respected
in
the
infrastructure
providers.
So
that's
something
like
we
might
want
to
discuss
a
little
bit.
C
C
This
needs
like
a
whole
proposal,
though,
and
then
there
is
the
cluster
managed
topologies,
which
is
a
simple
way
to
kind
of
spin
up
a
cluster
using
cluster
classes
to
create
control
planes
and
not
tool
management
using
machine
deployments.
But
these
two
they
need
to
be
thought
out
like
and
given
their
are
additional
their
auditions
and
there
shouldn't
be
any
breaking
changes
to
support
these.
They
might
come.
C
Also
later
I
don't
know
the
status
on
the
term
initial
design
handler
I've
asked
for
a
status
and
the
person
that
was
working
on
it
like
said
that
it
was
all
good
to
go,
although
there
were
a
lot
of
comments
still
open,
so
like
I'm
kind
of
confused
on
the
status
of
this.
So
if
anybody
has
any
idea
like
please
like
update
on
the
issue,
then
we
have
opt-in
for
auto
scaling
from
zero.
I
know
this
is
in
review
and
mike
is
working
on
it.
I
don't
like.
D
C
C
Thank
you,
so
all
of
this
to
say,
like
we're,
probably
not
gonna
hit
the
the
april
14th
or
20th.
I
don't
remember
like
right
now,
like
what
you
did.
We
set
that
line
and
that's
okay,
like
you
know,
like
we're,
trying
to
figure
like
our
processes
out,
but
at
the
same
time
like
it
would
be
great
to
actually
set
a
date
and
also
like
freeze
any
further
proposals.
A
E
Yeah,
so
I
agree
that
we
can't
like
push
dates
forever.
I
think
that
june
seems
reasonable,
given
the
amount
of
things
we
need
to
do.
It
might
be
interesting
to
also
set
a
date
for
enhancement,
freeze,
meaning
that
we
need
the
caps
merged.
It's
like
at
some
specific
date
this.
This
would
allow
maintainers
more
time
and
implementers
more
time
to
work
on
the
code
and
review
the
prs.
C
C
E
Yeah,
what
I
mean
specifically
is
we
need
like
we
need.
We
probably
need
a
specific
date
to
to
use
as
a
working
function
to
ensure
that
proposals
are
merged
early
and
not
just
languishing
through
the
review.
So
if
we
can
set
a
proper
date
where
we
say
okay,
this
is
the
deadline
for
the
proposals
to
be
merged.
Otherwise
you
can,
you
can
ask
for
an
exception
or
or
you
could
just
get
kicked
out
of
the
mushroom.
F
Cecil
yeah-
I
just
want
to
pile
on
to
that.
I
think
when
we
said
the
proposal
freeze
before
it
was
just
like,
we
need
to
know
about
the
proposal
like
the
proposal
has
to
have
an
issue
or
a
pr
open,
but
I
I
think
at
this
point
we
also
need
to
have
a
date
like
yesterday
said
where
the
proposals
actually
merged
and
agreed
upon,
just
because
it's
hard
to
say
we're
going
to
block
the
release
on
a
proposal.
That's
still
at
the
proposal
stage
and
hasn't
been
reviewed
and
approved.
A
A
C
Yes,
so,
like
there's
also
a
commentary
that
would
like
the
icos
like
for
proposals
that
like
they,
if
we
need
changes
on
all
providers,
those
are
those
are
big
and
like
also
like,
they
raise
the
bar
a
little
bit
for
infrastructure
providers.
That,
like
don't
super,
follow
like
the
the
cappy
release,
release
change
log.
So
we
need
to
help
those
providers
like
to
make
sure
that
they're
up
to
speed
in
terms
of
deadline.
C
E
E
E
D
I
just
wanted
to
point
out
what
jason
was
saying
here
in
in
chat
that
the
kubecon
recording
deadline
is
on
the
fifth
which
might
also
impact
people
who
are
working
on
these.
So
we
might,
we
might
want
to
go
to
like
april
21st
or
something
as
the
drop
dead
date
just
to
give
everyone
like
three
full
weeks
to
work
on
it
or
something.
E
A
H
Yeah
sure
so
in
the
list,
I
don't
see
any
proposal
link
marked
on
the
open
proposal
list
as
well.
As
I
looked
at
slack
to
see.
If
there
is
any
information
around
this
couldn't
see
much
and
as
it
was
marked,
this
needs
an
issue
and
a
proposal.
So
I
don't
know
it's
just
curious
to
know
like.
Is
there
any
write-up
or
if
is
there
any
discussion
that
had
happened
on
any
of
the
slack
challenge?
I
would
like
to
know
more
about
this.
C
E
C
Yeah,
so
that
that
this
is
me
like,
I
have
not
had
time
to
to
write
up
the
proposals
like
at
all,
so
let
me
do
that
in
the
next
like
a
week
or
so,
but
like
the
tldr
on
these
things
is
to
improve
the
date
zero
operations
so
like.
How
do
we
get
to
a
cluster
way
faster
than
we
do
today,
and
one
idea
was
to
first
have
a
way
to
stem
clusters
that
then
infrastructure
providers
could
include
in
their
own
releases
if
they
want
to.
C
So
this
is
a
totally
opt-in,
opt-in
and
the
other
thing
is,
if
you
do
have
that
stamping.
We
can
also
like
have
a
way
for
the
cluster
object,
to
be
a
little
bit
more
useful
and
have
like
the
topology
of
that
cluster
inside
the
cluster
object.
So,
for
example,
you
say
I
want
a
control
plane
which
is
highly
available
or
you
know
like.
C
H
A
D
F
And
are
we
all
okay
with
april
21st,
I
think
like
let's
give
people
a
chance
to
dissent
in
the
next
24
hours,
maybe
on
slack
or
asynchronously.
If
you
weren't
able
to
make
this
meeting
and
we'll
we'll
post
about
this
on
cycle
as
well,
so
that
it's
there
yeah,
let's
do
a
lazy
consensus.
What's
what's
a
good
end
of
the
week?
Is
that
consensus
about
the
consensus?
E
C
A
Yeah,
I
can
do
that.
I
can
prune
this
for
the
ones
that
are
interesting,
so
I
obviously
merged
this.
One
is
not
this
one's
already
merged-
that's
great,
but
I
and
and
as
well
for
this
qb
types
and
cluster
api,
but
I
think
the
other
ones
should
be
broadcast
to
say
that
these
proposals
have
to
be
merged
by
april
21st.
If
you
have
any
objections,
please
speak
up.
Does
that
make
sense.
F
A
B
C
A
C
C
C
A
I
I
want
only
to
add
a
note
here
that
we
just
finished
at
the
kubert
mean
office
hour
and
the
government
office
hour
lubomir
reminded
me
that
we
have
a
roadmap
to
the
break,
to
to
remove
the
the
cluster
status
from
the
kubernetes
config
map,
which
is
a
a
thing
that
was
replicated
in
kubernetes
two
or
three
years
ago.
So
it's
and
there
is
already
a
replacement
which
is
a
set
of
annotation
this.
I
A
F
Yeah,
so
we
do,
I
mean
we've
been
doing
some
and
we
did
like
put
the
milestones
together
and
I'm
hoping
to
get
that
label
pr.
Merge,
I'm
still
waiting
for
someone
to
approve
it,
and
then
we
can
use
that
to
label
all
the
issues
that
are
in
the
mail
soon
that
are
considered
like
blocking
for
the
for
the
0.40
release.