►
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
Hello,
everyone
today
is
Wednesday
March
29th.
This
is
the
cluster
API
office
hours,
thanks
for
joining.
If
you're
new
to
this
meeting,
you
can
in
the
agenda
meeting,
notes
and
recording
in
this
document
that
I
will
paste
in
the
chat,
if
you
don't
have
access
to
join
sick
cluster
lifecycle
mailing
list
to
get
access.
A
As
a
reminder,
cluster
API
star
is
a
sub-project
of
sick
cluster
lifecycle
and
we
follow
the
cncf
code
of
conduct
in
this
meeting.
So,
if
you'd
like
to
speak,
please
raise
your
hands,
just
make
sure
you're
respectful
with
everyone
and
with
that
being
said,
let's
get
started.
We
have
a
packed
agenda
today,
so
first
things
first
we'd
like
to
give
a
chance
to
anyone
who's
new
to
the
call
to
introduce
themselves
if
they'd
like
at
the
beginning
of
the
call.
B
Stephen,
hey
how's
it
going
yeah.
My
name
is
Stephen
Cahill,
it's
my
first
time
joining.
The
meeting
was
suggested
to
join
this
group
by
a
colleague
of
mine,
I
work
on
the
technical
support
team
at
datadog,
and
you
know,
I
mentioned
to
a
colleague.
I
was
looking
for
an
open
source
project
to
get
involved
with,
and
he
recommended
this
one
so
have
been
looking
around
at
the
quick
start
ducks
and
you
know,
checking
out
some
of
the
source
code
and
so
just
kind
of
familiarizing
myself
with
the
whole
project.
A
Cool
all
right:
let's
get
moving
so
discussion
topics.
David
sorry
had
the
first
one
go
ahead.
C
Yeah
so
I
know
that
there's
these
there's
this
cluster
resource
set
I
came
across
that
has
properties
of
installing
add-ons
for
cluster
API
and
it
looks
like
this
proposal
hasn't
been
updated
in
quite
a
long
time,
and
so
I
wanted
to
see
from
the
group
where
you
know
what
was
the
background.
What
was
the
desired
usage,
because
I
do
see
a
bit
of
overlap
with
the
cap,
H
add-on,
and
so
so
yeah.
A
If
not,
I
can
try
so
cluster
resource
that
is
currently
experimental,
I.
Think
from
my
understanding
at
least,
the
desire
was
to
eventually
deprecate
it
and
not
graduate
it
out
of
experimental.
A
It
was
initially
introduced
more
as
a
stopgap
solution
in
order
to
be
able
to
get
functional
clusters
of
cni
created
with
cluster
API,
because
the
CMI
needs
to
install
sub
installed
separately.
Sorry,
there
is
no
easy
way
to
do
so,
and
it
had
to
be
manual.
So
the
cluster
resources
were
introduced
as
a
way
to
apply
an
arbitrary,
manifest
or
arbitrary
config
map
or
secret,
or
to
your
plus
workload
cluster.
A
As
soon
as
the
API
server
is
up,
it
was
initially
implemented
in
a
apply
one
strategy,
which
meant
it
would
never
be
reconciled
after
the
initial
apply
so
not
very
resilient
for
long-term
maintenance.
More
of
like
a
a
quick
get
started
approach,
which
is
why
we
started
looking
into
the
add-on
providers
and
other
Solutions
I
think
recently.
There
was
a
proposal
to
make
it
reconcile,
which
I
think
was
merged.
Anyone
can
correct
me
on
that,
but
yeah.
A
So
that
was
added
more
recently
by
folks
who
are
already
using
cluster
resource
set
and
looking
to
use
it
in
the
meantime,
while
other
Solutions
weren't
available.
A
D
Yeah,
adding
on
top
of
that,
so
we
know
that
customer
resource
set
has
some
limitation.
If
you,
we
recently
update
the
proposal
for
supporting
the
sync
mode
that
was
implemented
in
1.4.0,
if
I
remember,
one
so
and
and
the
people
updating
the
proposal
made
a
good
job
in
a
documenting,
the
limitation
of
clusters,
resource
certain
and
basically
the
limitation
of
customers
who
said
are
of
various
nature.
First
of
all,
cluster
resource
set
does
not
offer
a
convenient
way
of
packaging
addons.
D
The
provider
that
you
are
talking
about
is
trying
to
integrated
to
them,
but
there
are
many
other
Solutions
out
there
and
so,
but
at
the
same
time
we
got
a
little
bit
of
pushback
from
the
community
because
folks
are
still
using
taster
resource
set
and
from
a
group
of
forks.
It
seems
bright
enough,
so
we
don't
have
concrete
plan
to
deprecate,
but
we
know
that
it
has
limitations
that
are
hard
or
impossible
to
fix.
D
C
Okay,
yeah
I
mean
I'm
just
getting
a
little
bit
of
mixed
messages,
so,
but
it
feels
like
what
I'm
hearing
from
both
you
and
Cecile
is
that
long
term?
It
sounds
like
this
is
not
the
way
we
want
to
go
and
we
want
to
go
move
towards
add-ons,
and
so
that's
exactly
why
I'm
bringing
this
up,
because
if
we're
going
to
move
towards
Adams
I'd
like
to
try
and
understand
what
what
kind
of
add-on
could
we
do,
that
would
you
know
supplement
or
replace
this
functionality?
That's
here
I'm
curious.
C
C
D
Yeah,
it
is
a
tricky
question
in
the
sense
that
we
know
that
the
first
solution
clusters
was
set
as
limitation,
and
that
is
not
a
good
way
to
fix
the
limitation,
and
so
we
tried
to
get
the
community
to
Rally
around
something
else,
and
this
is
what
we
hope
that
will
happen,
but
it
it
will
take
a
little
bit
of
time.
One
of
the
trigger
of
these
was
the
moving
of
of
was
the
move
of
Cloud
providers
out
of
three
which
trigger
it.
So.
D
We
have
to
give
time
to
the
cluster
API
a
Dom
provider
to
grow
and
to
to
reach
maturity
in
order
to
to
do
so,
they
need
the
feedback
from
the
community,
and
so
it
will
take
a
little
bit
of
time.
So
there
is
no
black
and
white
or
preferred
solution
as
a
copy.
We
kind
of
recognize
that
we
we
need
that
known.
At
the
same
time,
we
don't
want
to
reinvent
the
wheel
and
implemented
another
add-on
manager.
What
why
people
are
using
already
and
are
happy
with
already
existing
add-on
manager.
D
D
So,
if
I
look
this
also
with
the
head
of
see,
Castle
life
cycle,
Tech
lead
I,
think
that
we
should
avoid
duplication
and
we
should
focus
capping
doing
one
thing
well
and
leverage
another
tool
for
the
for
the
top
provisioning
just
provide
a
thin
layer
that
is
orchestrating
Elm
or
k
up
or
whatever,
together
with
the
cluster
I,
think
that's
the
ideal
solution,
and
but
yeah
in
order
it
it
is
open
source.
It
is
community
at
the
end,
it
drives
the
feedback
from
the
user,
etc,
etc.
So
yeah
all
right
that
helps.
C
It
should
be
similar
to
echoing
your
comments
for
briso.
It
should
be
the
bootstrapper
you
know
so
perhaps
there's
an
add-on
of
the
get
Ops.
You
know
bootstrap,
you
know
type
of
thing
and
then
people
can
use
that,
to
you
know,
Hydrate
with
underlying
flux
or
Argo
or
whatever
you
know
to
to
do
the
rest,
but
yeah
so
I'm.
Perhaps
then,
maybe
next
steps
might
be.
For
me,.
C
A
proposal
on
kind
of
the
future
of
these
things,
you
know
a
plan,
would
that
would
that
be
a
good.
D
C
A
Thanks
thanks
for
Brazil
and
David
yeah,
the
add-on
space
is
complex
and
very
crowded
one,
so
good
question,
okay,
so
fabric.
So
you
have
the
next
one
about
the
image
registry.
Actually
you
have
the
next
two,
so
not
giving
you
a
break,
go
ahead.
D
So
then,
thank
you.
So,
first
of
all
we
discussed
earlier
this
several
times.
I
got
another
ping
from
Ben
from
testing
for
us,
slash,
kubernetes
steering
committee.
So
basically,
you
know
that
the
entire
Community
is
trying
to
reduce
the
the
cost
of
the
infrastructure
and
the
one
key
point
for
this
weapon
is
the
immigration
from
the
all
the
image
registry,
which
is
GCI
k,
a
desk
GCR.
I
o
to
the
new
one,
which
is
a
registry
kasio.
D
This
effort,
basically
added
to
accelerate
due
to
the
rising
cost
of
the
infra
and
of
a
big
migration,
happened
the
20th
of
March
so
a
few
days
ago.
So
two
two
calls,
first
of
all,
if
any
one
of
you
is
experienced,
this
problem,
please
report
back.
We
can
help
you
to
get
in
contact
with
folks
with
the
that
are
running
this
effort.
D
B
D
G
Yeah
I
just
wanted
to
add
that
the
guide
for
cluster
API
in
the
how
to
identify
images
and
how
to
mirror
and
also
how
to
use
the
consume
mirrored
images
is
already
merged.
So
it's
part,
it's
already
part
of
the
guide
that
publisher
is
listing,
so
please
go
ahead.
Take
a
look
and
see
if,
if
there's
anything
that
we
would
want
to
add,
on
top
of
it.
A
Cool,
if
not
go
ahead
with
the
next
one
here.
D
Okay,
next
one
is
a
bit
of
bad
news,
so
you
may
know
that
we
are
using
a
library
which
is
called
drone
in
cluster
cattle
for
doing
a
substitution,
and
this
seems
that
this
library
is
not
longer
maintainer.
They
are
not
accepting
nor
PR
nor
issues,
and
this
is
a
problem.
So
we
have
to
look
for
substitution
and
that's
a
fun
one,
because
we
made
a
search
for
Alternatives
and
it
seemed
that
there
are
alternatives,
but
they
are,
they
have
slightly
difference
in
in
syntax
and
so
yeah.
D
We
have
to
figure
it
out
how
to
move
on.
If
you
are
interested,
please
take
a
look
at
the
issue.
There
is
a
back
and
forth
between
me
and
the
pair
and
the
people
that
raised
the
the
concern.
F
A
All
right,
please
take
a
look
at
the
issue
if
you
want
to
comment
async
all
right,
I'm
guessing
this
next
one
is
from
the
release.
Comps
team.
F
Yes,
good,
thank
you,
Cecil
So,
Yesterday,
Cappy,
released
1.4
and
two
new
patch
versions
and
I
would
like
to
just
thank
everyone
who
contributed
to
it,
and
I
would
like
to
thank
everyone
who
worked
on
making
the
release
possible.
F
I
will
move
quickly
along
to
the
to
the
next
topic,
one
of
the
which
is
the
area
labels,
and
this
is
one
of
the
improvements
that
we
are
working
on.
That
was
identified
where
we
would
like
to
discuss
and
have
input
on
which
area
labels
Cappy
should
have
so
that
we
can
start
automatically
assigning
them
to
different
areas,
and
this
helps
with
improving
the
release
notes.
F
So
we
are
looking
for
input
to
to
these
issues.
Thank
you.
A
Awesome,
thank
you.
Oscar
and
I
just
want
to
say
huge
congrats
I
really
seen
really
really
great
work.
All
throughout
this
cycle,
like
you
all,
had
a
not
easy
task
of
doing
this
for
the
first
time
with
no
prior
example
or
prior
art,
and
you
I
think
over
delivered.
So
thank
you
so
much
for
all
your
hard
work
and
getting
us
to
this
1.4
release.
H
Real
quick
on
the
area
labels,
it's
in
a
Google,
doc
and
I,
don't
think
everybody
has
edit
and
so
I
think
we
kind
of
went
with
suggest
for
now.
H
But
if
there's
a
reason
that
you
need
edit
or
want
to
change
something,
let
me
know
I
think
myself
and
Oscar
are
both
I,
don't
know
if
Oscar's
owner
of
the
file
I
don't
know
how
this
works,
but
anyway
we
should
be
able
to
help,
add
and
do
that
kind
of
stuff.
So
it's
just
a
a
running
list.
For
now
we
can
add,
delete
whatever
and
that's
kind
of
why
I
wanted
suggestions.
A
Sounds
good,
is
there
an
issue
associated
with
this
duck
or.
H
A
Right
there,
okay,
perfect
great
awesome
and
I
guess
one
question
so
just
to
clarify
these
area.
Labels
is
the
intended
usage
to
auto-generate
release,
notes
categorized
by
area,
or
is
it
also
to
assign
owners
to
PRS
for
review
so.
H
It
could
be
a
larger
use
case,
but
for
the
release
comms
team,
it's
to
help
us
be
able
to
put
them
into
buckets.
Basically.
So
when
you
look
at
the
release
notes,
they
have
prefix
on
all
the
the
notes
and
in
some
cases
we're
not
sure
what
the
area
should
be.
So
the
ask
here
is
that,
as
a
part
of
the
process
notes
will
be
sorry
area.
H
Labels
will
be
added
either
by
the
the
person
creating
the
pr
or
the
reviewer
and
then
hopefully
long
term
be
automatically
added
kind
of
like
kubernetes
is
at
least
for
release.
Notes
now,
assigning
you
know,
tasks
that
that
could
be
another
topic
as
well.
A
G
And
say
great
nicely
to
the
next
one:
is
we
with
the
1.4
and
the
patch
release
is
out?
We
now
have
1.5
release
team
thanks
folks,
who
all
volunteered
and
congratulations
on
being
part
of
the
next
release
team.
A
Awesome,
thank
you.
You've,
Raj
I
was
not
able
to
open
the
slack
link
on
my
browser,
so
I
just
put
a
screenshot
there
for
now,
so
people
can
see
the
names,
but
you
can.
Click
on
the
so
I
can
congratulate
everyone,
for
you
know
stepping
up
and
volunteering
to
be
part
of
the
really
Steam
and
yeah.
Thank
you.
Everyone
who's
part
of
this
and
keeping
this
going
any
questions
on
that
fabric
store,
go
ahead.
D
Yeah
piling
up
of
the
Congress.
This
was
really
amazing
to
see
this
stepping
in
and
and
see
how
it
delivered
and
I
have
a
greatest
petition
for
the
next
one.
Thank
you,
everyone
for
volunteered
I.
If
I'm
not
wrong
tomorrow,
there
will
be
also
a
retrospective,
and
this
is
part
of
our
improvement
process.
So
your
infrared,
it
could
be
great
if
we
add
a
details
about
tomorrow
or
the
Cosmetic
meeting
and
I
invite
everyone
who's
interesting
to
attend
because
yeah.
D
If
we
learn
from
from
the
first
experience,
we
can
do
better
next
time.
G
Yeah
sure
announced
it
in
the
in
the
slack
Channel.
Where
did
you
want
me
to
add
details
in
the
meeting
doc
here.
A
All
right
yeah
thanks
for
saying
that
for
me
too,
and
no
pressure
next
release
team,
but
fabrizo
has
great
expectations
all
right.
One
question
for
you:
Raj
is
the
Retro
open
to
all.
I
Yep,
just
a
quick,
Service
Announcement
here
the
machine
pools
PR
is
moving
along
on
the
cluster
Auto
scaler
and
hopefully
I'll
give
it
all
the
approvals
this
Friday.
If
there's
no
big
changes
that
are
needed
so
yeah,
please
take
a
look
if
you're
interested
that's
about
it.
A
Awesome
and
just
a
note
also
that
this
goes
hand
in
hand
with
machine
for
machines
in
Cappy,
which
is
still
waiting
for
some
reviews.
So,
if
you're
interested
in
this
area,
please
take
a
look.
I
I
A
Totally
yeah
great
points
all
right,
thanks,
Matt
and
Jonathan
for
working
on
this
any
questions
from
anyone
on
Mission,
Pool
machines
or
autoscaler
Machine
Tools.
J
Thanks
Cecile,
so
I
I
just
had
a
question
around
since,
since
y'all
know
that
we
we
introduced
the
experimental
API
types
around.
The
ipam,
which
were
IP,
addresses
and
IP
address
claims
were
the
two
CRS
that
we
introduced,
I,
think
1.3
or
1.4
was
when
we
introduced
those.
J
There
was
no
specific
contract
as
such,
defined
around
what
around
the
things
that
the
iPad
different
ipam
providers
should
Implement.
So
there
like
for
folks
who
might
not
know
about
this.
There
is
contract
definition
in
the
cluster
API
book
around
the
infrastructure
providers,
the
bootstrap
providers,
the
kcp
providers
and
all
of
that,
so
it's
just
hoping
what
is
the
guidance
to
change
the
contract
for
the
ipam
API
types,
since
it
is
an
experimental
API?
J
A
So
there
was
a
recent
PR
proposal
from
Fabrizio
I
think
that
merged
for
allowing
experimental
features
to
be
exceptionally
immersed
into
patches.
A
A
That
being
said,
we
do
want
to
keep
this
as
a
case-by-case,
like
exception,
only
basis
to
not
make
it
a
habit,
because
with
every
change
that
we
make
in
a
patch
release,
even
if
it's
experimental,
we
increase
the
risk
and
the
difficulty
for
folks
to
adopt
it.
So
if
there
are
too
many
commits,
it
can
be
hard
to.
A
You
know
parse
through
everything,
so
I
think
preference
is
still
to
always
put
stuff
in
a
minor
release
if
it's
possible,
but
if
it's
really
blocking
the
development
of
an
experimental
feature,
then
I
think
it
can
be
proposed
as
a
patch
and
evaluated
Case
by
case.
D
K
D
D
While
the
claim
has
been
processes
is
that
right,
yeah
I
think
that
if
we
don't
have
another
way
to
Signal,
this
is
more
a
bug
of
the
current
documentation
or
that
then,
then,
an
improvement
so
items,
because
otherwise,
how
can
we
detect
when
the
address
is
provided?
So
if
everyone
agrees
that
this
is
a
documentation,
we
can
have
a
better
documentation.
I
I
only
ask
the
courtesy
to
check
with
the
few.
D
If
I
remember,
there
are
only
one
or
two
implementation
of
this
provider,
so
it
should
be
easy
to
to
check
to
have
a
feedback.
Look
that
we
are
not
breaking
anyone
and
also
answering
to
succeed
in
copy.
We
have
only
API
types,
so
I
do
not
expect
this
to
be
an
investing
change.
J
I
think
in
in
terms
of
next
steps,
maybe
I
can
start
or
create
an
issue
and
add
so
at
least
I
know.
There
are
two
implementations.
Currently
one
is
the
in
cluster
ipam
provider,
implementation
and
I
know.
J
Some
of
the
folks
are
working
on
an
infoblox
IBM
provider
implementation,
so
I'd
reach
out
to
to
at
least
these
two
sets
that
I
know
and
ask
for
specifics
around
if
this
contract
change
might
break
something
for
them
or
make
it
harder
for
them
to
accept
or
to
who
want
to
like
a
patch
release
with
this
contract
change,
if
not
the
case,
then
we
can-
or
maybe
at
least
we
can
start
the
discussion
there
and
hopefully
go
from
there
yeah.
J
This
was
just
to
get
an
idea
of
like
if
this
is
something
that
we
could
do
in
the
current
batch
or
in
the
current
patch
release
or
not,
but
yeah
thanks.
A
Sounds
good
and
no
matter
what
we
should
still
start
with
a
patch
or
a
fix
in
the
main
branch
as
the
first
step,
so
we
can
start
there
and
if
anyone
is
in
this
call
or
listening
to
the
recording
and
is
like
wait,
a
second
I
have
an
iPad
provider
and
it
wasn't
listed.
A
You
should
definitely
reach
out
to
Cigar.
Just
so
we
know
about
you
all
right,
Willie.
You
got
the
next
topic
and
last
one
go
ahead.
E
Yeah,
hey
everyone,
I
just
wanted
to
raise
up
a
a
potential
issue
with
cappy1.4
that
we've
been
running
into
on
on
the
camps.
East
Side.
This
is
specific
to
dual
stack
clusters
for
Azure,
because
with
Azure
IPv6
you
need
to
have
both
an
ibv6
and
ipv4.
So
it's
like
it
has
to
be
dual
stack.
E
So
the
issue
is
there's
two
internal
IP
fields
on
our
nodes
and
there
seems
to
be
an
issue
with
SSA
unable
to
being
unable
to
like
apply
a
patch
when
there's
a
duplicate
field
and
yeah
I'm,
not
too
sure
like
exactly
what
changed
in
canopy
1.4
that's
causing
this,
but
it
I
think
you've.
Rod
has
been
taking
a
look
and
thank
you
again
and
just
wanted
to
raise
this
potential
issue.
If
anyone
has
problems
with
like
duplicate
fields
and
cappy1.4,
you've
Raj.
G
Yeah,
so
this
explicitly
happens
only
when
there
is
tool
stack
involved,
because
that
is
when
there
is
a
merge
key
conflict.
When
you
try
to
apply
an
SSA
patch
one
thing
that
I'm
still
curious
about
which
I'm
looking
at
right
now
is
we
don't
so
so
the
musky
conflict
is
in
the
status
of
a
node,
but
we
only
SSH
patch
the
labels
of
the
node,
so
I'm
trying
to
dig
deep
to
see
where
the
actual
problem
is.
But
from
the
preliminary
analysis
I
did.
G
E
Yeah
thanks
thanks
for
that
explanation.
It
definitely
seems
like
we're
going
the
right
direction,
but
yeah
just
a
quick
announcement
in
case
anyone's
running
into
similar
issues
when
upgrading.
D
Just
just
I'd
like
to
repeat
that
the
issue
just
to
see
if
I
got
it
right
so
what's
happening
is
that
we
are
patching
labels
and
we
got
a
an
error
back
because
the
notice
the
node,
the
core
type,
it's
not
a
CR
it
does
not
have
the
is,
is
not
a
capable
to
to
match
it.
G
One
thing
I
want
to
add
is
I
know
that
other
providers
are
also
working
on
integrating
1.4
into
their
into
their
releases
right
now.
So
if
other
providers
can
also
run
dual
stack
tests-
and
let
us
know
if
you
are
seeing
similar
problems,
that
would
be
great
for
us
to
signal
and
identify
and
then
isolate
the
probability
further.
A
Verification
we're
running
this
into
this
with
cabs,
with
itv6
single
set,
not
dual
stack,
but
we
should
run
the
Dual
psych
test
too
to
see
if
we
run
into
the
same
error.
The
reason
we
run
into
it
with
IPv6
in
Azure
specifically,
is
because
there
is
no
true
single
site,
IPv6
Network
for
VMS,
so
the
kubernetes
is
single
stack,
but
the
VM
has
a
dual
network:
ipv4
IPv6,
so
there's
one
IP
for
ipv4,
one
for
IPv6
in
the
node
addresses.
E
A
All
right.
Thanks
for
all
the
discussion
topics,
everyone
last
but
not
least,
provider
updates,
Matt
Boardman.
You
had
the
first
one.
K
Thank
you
Cecile,
nothing,
really
to
add
what
I
have
there.
We
have
two
patch
releases
that
went
out
yesterday.
This
is
two
weeks
of
work.
Sometimes
we
do
it
more
quickly,
so
there
actually
is
quite
a
bit
in
there
and
please
use
those
if
you're
a
capsi
user
and
then
Willie
already
talked
about
the
cappy140
integration.
So
I
don't
need
to
go
over
that.
A
Awesome
cap,
oci
Joe.
H
Hey
sorry,
I
didn't
have
my
camera
on
earlier
I
apologize
yeah,
so
we
cut
release
zero,
eight
zero
I,
don't
know
maybe
a
week
or
two
back
and
I
was
out
of
town
for
a
little
while,
but
we
now
support
okay,
oracle's
managed
kubernetes
I.
Think
I
demoed
that
off.
But
then
the
other
thing
we're
kind
of
excited
about
is
we
we
support
Windows,
at
least
on
our
bare
metal,
so
not
VMS
yet,
but
on
our
bare
metal.
So.
A
Might
be
good
to
get
Oracle
that
folks,
like
opinions
in
there
awesome
and
then
last
one
tab.
H
still
don't
know
how
to
pronounce
it.
Jonathan
go
ahead.
L
A
Great,
what's
the
best
way
for
folks
to
get
started.
L
So
I'm
working
on
adding
some
make
file
targets
right
now
so
that
you
can
run
make
deploy
or
make
install
and
get
that
set
up
for
now,
I'm
also
working
on
setting
it
up
so
that
we
have
add-on
providers
and
cluster
got
all
in
it,
but
that
might
take
a
bit
so
for
now,
I
was
just
gonna
set
up
the
make
file
so
that
you
could
deploy
it
to
your
management
cluster
easily,
so
that'll
be
coming
in
the
next
couple
days.
A
Awesome.
Thank
you.
Any
questions
for
any
of
the
providers
or
any
additional
last
minute
provider
updates.
D
Then
just
thank
you
for
all
the
provider
which
are
working
to
the
copy
bump.
It
is
super
yeah
without
the
provider.
Cap
itself
does
is
not
really
meaningful,
so
yeah
we
need
a.
It
is
super
important.
We
need
feedback
and
then
we
need
a
release
with
support
for
copy
1.4
as
soon
as
possible.
Thank
you.
Everyone
working
on
this.
A
Absolutely
all
right,
if
nothing
else
hope
you
all
have
a
great
end
of
your
Wednesday
or
Thursday.
If
it's
already
Thursday
in
your
time
zone
and
see
you
all
next
week,
thanks.