►
From YouTube: SIG Cluster Lifecycle - CAPV Office Hours - 2023-09-14
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,
recording
this
is
the
cluster
API
provided
for
vsphere
community
meeting
for
the
14th
September
2023.
This
meeting
is
ruled
by
the
cncf
code
of
conduct,
so
more
or
less
comes
down
to
be
nice
to
each
other,
fill
in
attendees
if
you're
attending
and
I
think
we
can
just
get
through
to
the
agenda
quickly.
A
So
no
news
and
updates
first
thing:
Cocker's
discussion,
quick
one
is
a
PSA
we're
removing
V1,
Alpha,
3
and
Google
Now
Alpha
4.
I'll.
Add
the
link
to
the
issue
here
as
well,
but
this
is
motivated
explicitly
by
cluster
API,
removing
its
V1
out
for
three
types,
which
means
as
soon
as
we
move
to
Taffy
1.6,
which
is
the
next
release
of
cluster
API.
We'll
have
to
remove
these
types
because
we
import
the
V1
has
three
types
in
order
to
upgrade.
A
So
what's
not
clear
at
this
point
is
whether
or
not
this
will
happen
with
cat
V,
1.9
or
1.10,
because
we
don't
know
which
one
will
end
up
importing
copy
1.6,
but
it
will
almost
certainly
happen
with
imported
Capital
1.6.
So
I'll
bring
this
up
at
a
meeting
again
with
bigger
audience
and
definitely
will
need
to
communicate
this
to
the
community
as
it
happens,
but
yeah,
it's
just
a
quick
yeah.
So
so
Richard
do
you
want
to
talk
about
the
ipam
apis.
B
Yeah,
this
was
a
topic
that
I
raised
the
last
meeting
just
trying
to
understand
what,
if
there
is
a
plan
to
get
the
apis
promoted
from
experimental
to
Beta
for
some
context
in
openshift,
we
have
implemented
a
ipam
implementation
into
our
machine
API,
but
our
objective
throughout
that
work
has
been
to
have
an
endpoint
of
getting
to
Cafe
at
some
point
down
the
line.
B
So
we've
been
using
the
Cappy
types
to
facilitate
the
creation
of
the
various
crds
with
the
IP
address,
claims
and
IP
addresses,
and
we
are
interested
and
seeing
if
this
API
can
be
promoted
from
experimental
just
so
we
can
have
some
assurance
that
there
won't
be
some
significant
API
drift
that
we're
ready
to
account
for
down
the
line
and
what
I'm,
also
trying
to
figure
out
is
that
this
is
the
right
form
to
be
raising
this
question.
B
I
I
see
that
Kathy's
been
a
early
adopter
of
this
API
and
seems
to
be
leading
a
lot
of
the
discussion
around
it
and
that's
the
reason
why
I
wanted
to
raise
it
here.
C
Absolutely
worth
a
try,
so
just
based
on
the
context
that
I
have
just
sharing
that
so
as
far
as
I
know,
basically,
the
type
of
thing
happened
that
essentially
was
created
by
the
Telecom
folks.
They
brought
up
to
a
community
with
a
proposing
core
copy,
some
point
that
we
merge
types
there
and
basically
at
some
point
in
parallel,
VMware
or
Cappy,
got
onto
that,
and
then
they
work
together.
I
mean
not
us
too,
but
some
of
our
colleagues
essentially,
you
know
the
point.
C
Definitely
where
I
mean
we
have
them
in
a
capital.
You
know
one
of
crosses,
we
all
know
and
they're
used
in
Cappy
and
what
I
can
also
say
is
they're
heavily
used
in
our
product,
so
I'm
pretty
sure
that
we
have
like
the
same
goal.
We
also
like
to
get
this
to
stable
and
continue
to
use
it,
etc,
etc.
C
What
I,
just
don't
know,
is
essentially
what
the
maintainers
of
the
iPhone
apis
are
think
like
how
ready
is
it
or
not
as
far
as
I
know?
Lastly,
we
talked
about
it
and
think
busy.
Last
time
was
like
two
weeks
ago
or
something
instead,
they
were
saying,
like.
Oh,
we
want
to
maybe
think
about
this
thing
or
another,
but
otherwise
it's
fine,
but
it
can't
really
say
more
than
that,
because
it's
not
really
up
to
me
and
I.
C
Also
don't
really
have
more
information,
of
course,
so,
but
what
we
can
try
to
do
is
did
we
did
I,
try
to
push
this
discussion
in
select
somewhere
just
because
they
know
that
jakub,
for
example,
which
basically
wrote
a
proposal
and
iPhone
types
that
is
pretty
responsive
and
slack,
but
I
I
mean
I'm
only
here
since,
like
two
or
three
months
in
the
meeting
specifically
but
I,
know
he
never
joins
in
this
meeting.
B
C
C
Yeah
sounds
good
yeah,
maybe
maybe
start
a
stack
for
it
and
also
mention
me
there
and
let's
just
try
to
sometimes
it's
just
just
getting
created
with
something
via
slightly
more
synchronous,
communication
and
select
on
issues
sure
and
I'll.
Try
to
help
you
there
a
bit
too
get
a
clear
statement
like
okay.
C
We
can
look
at
certain
things
before
you
want
to
promote,
but
what
are
those
things?
Let's
try
to
get
them
done,
not
just
in
this
fussy,
stator
but
see.
Nobody
knows
what
you're
waiting
for.
C
C
Mean
it's
mightiest
way
to
go.
I
think
the
exact
right
thing
is
basically
the
issue
we
already
have
in
in
core
cluster
API,
so
called
consider
moving
iPhone
provider
out
of
experimental,
essentially
that
I
think
wins
opened.
C
Just
do
you
sorry
I'll
just
post
the
link
in
so
maybe
you
can
open
it.
I
just
want
to
remind.
A
C
Like
what
also
chat
right
now,
just
rereading
what
what
Jakob
was
writing
in
May
foreign.
A
C
B
C
Wrote
something
like
you
don't
know:
if
we're
not
a
provider
shop,
they
want
to
do
like
more
stuff
and
break
things,
but
I
mean
I.
Think
what
you
can
also
consider
is.
Essentially,
you
currently
have
a
live,
API
and
looking
just
at
code
cluster
repair
in
general.
Basically,
everything
is
only
one
bit
of
them
with
corresponding
guarantees,
and
we
also
essentially
have
even
better
to
it
on
the
horizon.
C
So,
basically,
even
if
you
move
it
to
V1
beta1,
we
can
still
do
breaking
changes.
It's
just
like
that.
We
have
to
support
we
on
bit
on
longer
and
I
mean
we're
part
of
even
beta2
to
do
some
sort
of
big
changes.
So
maybe
that's
a
consideration,
but
should
make
it
easier
to
make
the
jump
to
beta
one.
If
you're
saying
okay,
it's
stable
enough
for
that,
and
it
doesn't
mean
that
it
has
to
be
stable
forever.
It's
just
assuming
that
we
are
limited
in
the
sort
of
playing
changes
you
have
to
do.
C
You
need
some
nice
way
to
get
from
beta1
to
Beta
two
yeah,
that
sort
of
stuff.
B
B
Pretty
non
I
mean
non.
B
A
C
Yep,
that's
something
that
we
can
always
do.
I
mean
as
long
as
the
default
Behavior
respect
was
compatible.
We
can
add,
as
many
fits
as
you
want.
It's
really
just
problematic.
If
you
want
to
change
existing
fields,
I'm
not
working
way,
but
even
then
I
mean,
as
I
was
saying.
We
want
better.
One
is
not
frozen
forever,
it's
just
a
stable
for
longer
period
of
time.
C
What
I
can
also
say
is
there
are
some
discussions
in
the
community,
but
we
didn't
get
to
the
point
yet
that
we
actually
did
something
about
it
and
to
potentially
get
rid
of
experimental
in
general.
So
we
have
like
two
things
in
this
API
right:
one
is
it's
V1
alpha
one
and
then
it's
experimental.
You
need
a
feature
of
leg.
C
It
doesn't
make
that
much
sense
to
have
a
feature
gate
for
an
API,
because
I
mean
you
don't
have
to
use
the
API
I
mean
if
you're
talking
about
feature
guides
for
specific
Fields
inside
of
an
API.
That
might
be
a
different
story,
but
if
it's
like,
can
you
use
the
API
or
not?
You
might
as
well
just
not
use
the
API.
You
don't
need
it,
except
if
you
have
like
a
weird
split
between
or
not
that's.
C
Obviously
sorry,
if
you
have
different
personas
like
whatever
operator
are
deploying
cluster
API,
they
want
to
prevent
you
from
running
iPhone.
Then
you
have
a
split
between
whoever
is
running
test
API
and
whoever
is
deploying
series.
But
even
then
you
can
deploy
the
books
and
stuff,
so
even
that
one
so
yeah
that
didn't
get
to
the
point
that
we
actually
did
something.
But
I
could
imagine
that
within
the
next.
C
If
that
person
wants
to
drive
it
gets
to
it
within
the
next
three
to
six
months
we
manage
eventually,
maybe
just
get
rid
of
experimental.
That
still
leaves
the
API
version.
Discussion,
of
course,
which
is
like
a
specific
one,
yeah
yeah
I,
would
suggest.
Let's
start
a
secret
CC
me.
Definitely
there
and
also
Jakob
something,
and
let's
just
see
what
like
the
current.
The
current
takes
are
like
in
September
2023.
C
That's,
it
has
been
a
while,
since
we
got
some
segments.
C
Okay,
so
that's
everything,
sorry,
no!
No!
It's
fine,
yeah
I
just
want
to
say
look
if
you
don't
have
any
other
topics,
but
you
were
just
about
to
say
that
yeah.