►
From YouTube: SIG Network Gateway API meeting for 20230905
Description
SIG Network Gateway API meeting for 20230905
A
B
B
So
basically,
every
provider
that
wants
to
support
can
just
follow
up
this.
It's
pretty
pretty
straightforward
and
there
are
examples
and
code
looks
yeah,
so
it
should
be
really
really
easy
to
add
a
new
provider
and
there's
also
the
ability
for
implementations
to
kind
of
initiate
clients
read
their
resources.
Everything's
covered
here
so
yeah
I
just
want
I've
spoke.
B
I've
spoke
in
the
past,
with
Sanjay
and
John,
so
I
kind
of
want
to
raise
it
up
again.
We
really
encourage
implementations
to
implement
it,
because
once
Gateway
is
GA,
who
anticipated
a
lot
of
people
will
need
help
converting
their
resources.
So
yeah
have
a
look
and
I'm
available
in
slack
for
whatever
needed.
A
That
is
awesome.
Thank
you,
leor.
Well,
we
have
you
and
you're
talking
about
Ingress.
The
Gateway
I
know
you're
working
towards
an
initial
release.
What's
between
you
and
that
first
release
on
Ingress,
the
Gateway
yeah.
B
So
there
is
a
PR
in
flight
for
editing
the
whole
readme,
which
was
old,
so
yeah.
So
this
one,
it's
an
interview
and
except
from
that
I
want
to.
There
is
another
local
branch
with
release
process
documentation.
B
I
I
yeah
I
have
one
problem
with
the
kubernetes
release,
generator
go
to,
which
is
not
working
for
some
reason,
so
I
kind
of
follow
up
offline
with
you
after
oh
yeah,
once
it's
there,
I
yeah
I
want
to
release
the
first
one.
The
first
version
yeah.
A
Awesome,
that's
cool
thanks
for
all
the
work
on
this
Candace.
You
got
a
hand.
C
Just
had
a
quick
question
for
lelor
is
there?
Is
this
a
sort
of
a
runtime
live-action,
something
that's
watching
as
ingresses
are
created,
or
is
this
any
ingresses
that
exist
is
going
to
just
convert
them
all
yeah.
B
So
currently
it's
not
a
runtime,
it's
a
binary,
so
we
basically
run
it
and
you
either.
You
know
it's
either
consume
your
Cube
config
or
if
you
give
it
a
file
because
we
saw
it,
some
people
also
wants
to
transfer
from
files.
So
yeah,
the
probably
what
I
want
to
do
is
I
want
to
release
it
and
release
a
blog
post
with
it
with
the
GAA
release
and
then,
as
people
come
out
on
board,
do
it
probably
will
get
a
lot
of
feedback
and
then
we
see
all
right.
B
A
Yeah
Candace-
that
is
a
great
question
and
I've
actually
I,
had
a
conversation
with
someone
else
about
something
very
similar
to
that
of
well
I,
don't
know
if
this
exactly
where
you
were
going,
but
you
could
theoretically
use
some
of
the
code
here
to
you
know,
run
a
controller
that
just
understood
Ingress
and
gate.
You
know
just
implemented
Gateway,
but
also
implemented
Ingress
by
translating
it
to
Gateway,
maybe
an
interesting
Next,
Generation
or
future
feature
of
this
project.
B
So
so
yeah
have
a
look,
I
think
it's
pretty
pretty
straightforward
after
the
pr
I'm
I'm
waiting
for
so
we
have
Ingress
engine
next
sorry
supported
mostly,
but
we
kind
of
work
to
add
the
ability
to
read
custom
resources
to
enable
more
providers.
So
yeah
I
wanted
at
least
two
implementations
supported
before
kind
of
releasing
a
blog
post
on
the
first
release,
so
I'm
trying
to
work
with
people
to
implement
it.
A
All
right,
David,
I,
think
you're
back
now,
maybe
I'll
hand
it
back
off
to
you
and
see
if
demo
works.
D
D
Now,
if
you
can
see
my
screen.
D
Okay,
yeah
I'll
I'll
give
a
short
demo
here
of
what
this
project
is
where
it
came
from
and
where
I'm,
where
I'd
like
to
take
it,
and
so
this
is
give
API
State
metrics.
D
So
if
you're
not
familiar
with
State
metrics
or
Coupe
State
metrics
it's
about
getting
Prometheus
metrics
and
for
different
kubernetes
resources
so
that
you
can
say,
build
up
dashboards
or
alerts
around
the
state
of
different
kubernetes
resources.
So
in
this
case
it's
all
about
Gateway
API
resources.
So
it's
it's
Gateway
provider
agnostic!
Shall
we
say
so
the
repos
here
there's
a
bunch
of
information
in
here,
but
rather
it
can
go
into
any
of
that
detail.
D
I'll
focus
on
just
highlighting
some
of
the
metrics
and
then
a
few.
The
dashboards
that
I've
included
in
the
in
the
repo
as
well
do
jump
in.
D
After
any
questions
or
reach
out
to
me
after
as
well
by
the
way,
so
in
Prometheus,
first
I'll
just
give
you
an
idea
of
what
metric
what
these
new
metrics
are
and
what
they're
for
so,
if
I,
just
hopefully
the
drop
down,
is
appearing
in
the
screen
share.
But
it's
Auto
completing
here
with
a
bunch
of
different
new
metrics
here
so
Gateway
API.
D
It
is
about
seven
or
eight
different
ones,
for
gateways
a
few
for
Gateway
classes
and
about
four
or
five
for
HP
Reds
and
so
they're,
the
ones
that
are
implemented
at
the
moment.
So,
if
I
take
a
look
at
say
just
Gateway
API
Gateway
info-
and
we
can
see
that
there's
three
results
here,
so
I
have
three
Gateway
instances
in
this
cluster
at
the
moment
and
we're
get
getting
back
some
basic
information
from
this
metric.
So
what's
the
name
of
this
Gateway?
D
What's
the
namespace
and
some
custom
resource
information
here,
too
I'll
just
jump
jump
over
to
terminal
real
quickly,
just
to
show
what's
in
this
cluster,
and
that
will
make
it
a
bit
more
and
relevant
by
show
the
dashboards.
D
So
if
I,
just
let's
see
I,
think
I
have
one
Gateway
class
in
here
at
the
moment,
so
just
an
istio
Gateway
class
there.
Let's
look
for
any
gateways.
Three
different
gateways,
so
I've
set
this
up
to
be
like
I,
guess,
a
medium-sized
Enterprise,
so
I
have
an
external
Gateway,
an
internal
one
for
like
running
the
company.
I
didn't
have
an
internal
one
focus
on
developing
stuff
from
my
team
of
developers,
so
three
gateways,
maybe
a
bunch
of
different
Services.
D
D
So
here
we
go
a
bunch
of
different
hosts,
a
bunch
of
different
services
and
just
to
have
something
we
can
visualize.
So
I
jump
back
to
Prometheus.
We
have
a
look
at
say
the
history
routes
we
can
get
to
create
a
timestamp,
for
example,
as
a
metric
for
all
of
these
another
example.
What
we
have
we
have
the
the
hostname
info
coming
back
as
well
as
a
metric.
So
in
here
for
example,
you
have
a
label
here
for
the
hostname.
D
So
how
is
this
useful?
Well,
the
dashboards
is
just
one
way:
I
thought
it'd
be
good
to
show
that,
so
let
me
jump
over
to
to
grafana.
D
D
If
we
jump
into
Gateway
classes
first,
we
can
see
that
we
just
heard
one
so
as
expected
and
over
on
the
right
here
we
can
see
any
instances
of
that
Gateway
class
and
what
namespace
they're
in
and
so
on,
so
I've
included
a
couple
of
kind
of
what
you
call
them
quick
links
up
here.
So
I
can
jump
between
gateways
and
HTTP
Reds
up
here
and
just
go
back
to
classes
or
you
can
click
on
an
individual
Gateway
here
to
drill
down
to
that
specific
Gateway.
D
D
I
can
see
the
the
addresses
from
the
status
block
of
my
gateway,
whatever.
What
are
the
attached
routes,
the
actual
HP
routes
that
are
that
have
a
parent
reference
after
this
Gateway,
unlike
with
Gateway
classes,
the
gateways
I
can
also
drill
down
to
the
individual
route
as
well,
so
how
to
look
at
my
accounts
route.
That
brings
me
to
the
HTTP
route
page,
and
we
have
that
filtered
view
here.
D
So
that's
that's,
that's
more
or
less
what
I
wanted
to
show
and
kind
of
leave
it
open,
then,
for
any
any
questions
or
any
thoughts.
I
do
I,
do
want
to
kind
of
revisit
the
dashboards
and
maybe
take
a
different
approach.
So
there
could
be
more
say:
Gateway
API
provider,
specific
elements
to
it,
because
this
is
all
very
generic,
so
yeah
I'll
hold
there
I
see,
there's
a
couple
of
hands
up,
so
I'm,
not
sure
who
was
first
was
Candace.
C
Hi
sure
I
had
a
question.
This
looks
pretty
good,
actually
I
want
to
mention,
but
I
had
a
question
about
sort
of
how
I
remember
seeing
in
one
of
the
dashboards
there
was
a
healthy
and
accepted
statuses
and
I
was
wondering
whether
you
thought
about
you
know
having
this
sort
of
plotted
out
as
a
as
a
line
graph
over
time
sort
of
so
that
you
can
see
you
know.
C
Maybe
one
was
okay,
but
the
other
two
were
dead
or
something
I
don't
know,
or
you
know
how
much,
basically
counting
how
much
up
time
you
have
for
for
Gateway
I
thought
that
would
be
a
cool
thing
to
see,
but
otherwise
so
I
guess
it's
almost
like
the
same
question
I
had
before
I'm
interested
in
like
kind
of
like
what's
happening
now
with
the
status
of
the
Gateway
and
what
was
happening.
C
You
know
an
hour
ago
a
day
ago,
that
kind
of
thing
I
think
that
would
be
really
cool,
but
that's
the
only
addition.
That's
the
only
comment
that
I
that
I
had
otherwise
looks
pretty
good.
D
Thanks
yeah,
that's
that's
a
really
good
suggestion.
Actually
I
recall
seeing
something
similar
when
working
on
some
Kafka
stuff,
looking
at
the
state
of
character
instances
and
that
was
actually
very
valuable,
so
yeah
I
don't
think
it's
that
tricky
to
do
on
grafana
as
well
so
yeah
in
terms
of
healthy
State,
I.
Think
I
just
said.
If
it's
accepted
today
on
the
program,
State
we're
good
it's
healthy
and
we
could
even
quite
easily
graph
that,
like
I
was
saying
so.
A
C
Definitely
I
have
a
little
grafana
experience,
so
maybe
I
can
point
you
the
the
widget
that
you
need
excellent.
D
That'll
be
very
helpful:
New
York.
B
Yeah,
just
just
yeah
well
done
it
looks
really
good
I
suggest
one
one
small
thing:
I
suggest
you
might
keep
an
eye
on
a
gap
for
supported
features.
It's
probably
gonna
be
very
soon,
so
it'll
be
really
cool
to
represent
here.
It's
a
field
under
status,
so
it'll
be
really
good
code
to
represent
here.
What
are
the
supported
features
the
Gateway
support,
so
maybe
this
URL
will
be
more
useful
for
people
other
than
just
reading.
D
Cool
yeah:
that's
a
good
idea.
I'll
I'll
I'll
take
a
look
at
that
Gap
and
see
where,
where
that
data
appears,
if
it's
under
the
status
block,
we
can
easily
pull
it
out.
There
are
some
if,
if
there's
kind
of
an
array
of
things
within
the
status
or
sorry,
if
there's
nested
arrays
of
things
in
the
status
that
can
be
a
bit
trickier
and
there's
an
upstream
change
in
Cube,
State
metrics
I
depend
on
to
pull
in
that.
D
A
Well,
this
looks
really
great.
Thank
you.
Thanks
for
the
demo,
I
I
had
seen
just
a
little
bit
of
this,
but
I
didn't
realize
how
extensive
it
already
was.
So
really
really
cool
I
appreciate
what
you're
working
on
there
I
noticed.
This
was
under
the
quadrant
organization
on
GitHub
I.
Don't
know
enough
about
that,
but
it
looks
like
it's
doing
some
other
Gateway
related
things.
Do
you
have
any
more
context
on?
Yes,.
D
Yeah,
so
the
quadrant
project
I
think
I.
Think
we
probably
if
it's
if
it's
okay
with
the
people
here
would
in
the
future
call.
Maybe
we
could
do
a
small
few
slides
on
what
exactly
is
quadrant,
but
just
to
give
an
overview.
Now,
it's
it's
focused
on,
say,
policy
attachment
to
Gateway
API
providers.
So,
for
example,
all
policy
is
rate
limiting
and
we're
also
looking
at
some
multi-cluster
aspects
of
gateways
as
well.
D
D
So
that's
that's
kind
of
where
the
the
metrics
and
the
dashboards
piece
came
in
from
our
site.
We
were
like
okay.
If
you
have
a
lot
of
different
gateways
and
they're,
potentially
on
a
lot
of
different
clusters,
this
is
going
to
be
Troublesome
to
see
what's
going
on,
so
that
that's
why
I'm
looking
at
the
dashboards
piece
and
pulling
information
from
like
a
central
panus,
so
that's
kind
of
a
high
level
overview
but
yeah.
D
If
we
might
see
if
it's
okay
to
present
a
bit
more
at
some
point
in
the
future,
if
that's
okay,.
A
Yeah
that
sounds
very
interesting.
I
had
not
paid
enough
attention
to
what's
going
on
there,
but
it
sounds
very
cool.
Shane
also
mentioned
in
chat
the
multi-cluster
solution
so
yeah.
It
sounds
like
a
lot
of
very
relevant
things
to
Gateway
would
love
to
learn
more
about
your
experience
and
yeah
very
cool.
Okay.
Thanks,
awesome,
well,
I'll
keep
on
moving
through
the
agenda.
Then.
Unless
are
there
any
more
questions
on.
D
A
Comments
all
right,
I'll
keep
on
going,
then
all
right,
so
very
very
unfortunately,
it
looks
like
we
are
going
to
need
a
patch
release
of
0.8,
so
0.8.1,
unfortunately,
is
going
to
have
to
exist,
but
thank
you
very
much
to
Travis
who
called
this
one
out
and
caught
it.
We
had
made
one
mistake
in
the
cell,
validation
and
honestly
I
didn't
think
it
was
a
mistake
until
Travis
correctly
pointed
out
that
it
was,
and
we
were
unfortunately
missing,
test
coverage
for
this
very
specific
example.
A
Unfortunately,
the
cell
validation
released
in
0.8
does
not
allow
that
to
happen,
and
so
because
again
it
was
in
the
listener.
Uniqueness
validation
and
you
know
reasonably-
did
not
consider
these
two
to
be
unique,
but
unfortunately
that
was
a
mistake.
So
there
is
a
fix.
Thank
you
to
Frank
for
getting
this
fix
in
pretty
quickly.
A
We're
gonna
have
to
cut
an
0.8.1
release
pretty
soon
as
a
result,
but
I
think
we'll
wait.
A
few
more
days.
Please
try
and
break
Gateway
0.8
to
try
and
find
edge
cases.
You
know
trying
to
avoid
releasing
a
large
number
of
patch
releases
here
when
our
Focus
was
really
on
getting
1.0
out,
but
I
I
recognize
we
do
need
to
have
an
0.8.1
sometime.
Pretty
soon
just
would
love
to
make
sure
it
catches
all
potential
bugs
we
may
have
released
in
0.8.
B
Yeah
I
I
think
I
have
to
I,
asked
this
on
slack
last
week,
but
I.
Don't
so
I
think
if
I'm
not
mistaken,
so
I'm
not
sure
if
them
I
I
see
that
the
mirror
multiple
mirror
apis.
Wear
change
and
the
Web
book
was
changing
0.8
but
I
think
it
wasn't
com,
so
I'm
not
understanding
the
status
of
it.
If
it's
in
there
or
not.
So
if
it's
like
kind
of
partially-
and
we
might
want
to
make
more
clarifications
in
the
one.
A
B
There
is
yeah,
but
that's
going
to
be
in
the
triage,
but
there
is
the
conformance
test
which
we
do
patch
conformances.
So
if
we
could,
if
we
could
make
it
same
patch,
so
it'll
be
good.
A
Yep,
that's
that's!
Actually,
a
good
call
yeah,
so
anything
like
that,
like
the
multiple
mirror,
conformance
that
that's
something
that
really
should
get
in.
So
if
there's
anything
like
that,
that's
really
just
again
an
improvement
on
existing
Behavior,
not
an
actual
API
change
would
love
to
squeeze
that
into
0.8.1
as
well.
A
Cool
okay,
now
next
up
is
1.0.
That
is
coming
way
lately
too
quickly.
We
have
I
think
what
Shane
and
Nick
and
I
have
said.
Is
that
we're
really
trying
to
have
everything
merged
by
end
of
month
so
that
we
can
start
the
API
review
process
at
the
very
beginning
of
October?
So
that
does
not
does
not
leave
us
that
much
time
to
get
things
in.
So,
let's
run
through
the
release
blockers
real
quickly,
there
are
14
open.
A
Fortunately,
a
lot
of
these
have
open
open
issues,
open
PRS
for
them
or
someone
in
assigned
I'm,
pretty
sure,
I'll
double
check
with
Nick,
but
I'm
pretty
sure.
Oh,
this
is
actually
a
PR,
so
let's
good
the
rest
have
PR's
or
are
assigned
this
one
is
the
very
end
of
you
know,
so
we're
pretty
close,
but
I
I'd
encourage
you
if
you're
looking
for
ways
to
get
involved
that
will
be
very
impactful
and
help
us
I
think
we're
all
looking
forward
to
getting
a
ga
release
out
the
door.
A
A
So
with
all
that
said,
anything
you
can
do
to
help
if
you
see
PRS
that
are
associated
with
these,
if
you
can
help
review
them,
if
there
are
some
that
are
open,
that
don't
have
PR's
right
now,
if
you
can
help
that,
that
would
be
very
helpful
yeah
just
this
is
this
is
top
of
mind
for
us
in
the
coming
month,
as
we're
trying
to
rush
to
get
1.0
out,
I
also
wanted
to
cover
some
gaps.
A
B
If
you
go
to
the
last
comment,
I
think
yeah,
it's
just
a
few
minutes.
I
think
it's
one
hour
to
admit
yeah
this
one's
clarification,
yeah,
yeah
and
what
time
yeah.
A
B
There
is
a
there
is
a
change
follow
up
on
this.
There
is
a
change
to
http
request,
header,
modifier
or
response
header,
modifier
conformance
name.
So
if
you
want
me
to
get
it
in
8.1
I
can
it
should
be
not
related
like
it's
following
the
formatting
rules
we
defined
here,
but
I
mean
it's
just
gonna
clean
up
so.
A
B
A
Cool
then
next
step
is
and
I
just
these
are.
These
are
the
ones
that
are
are
top
of
mind.
For
me,
this
HP
out
route,
timeout
I,
don't
see
anyone
on
the
call
that's
been
working
while
Flynn
has
been
working
to
you
to
push
this
one
forward,
but
anyone
else
who's
been
been
working
towards
us
feel
free
to
speak
up.
I
think
this
is
just
about.
This
is
just
waiting
on
myself
or
Nick,
or
somebody
to
review,
but
other
reviews
are
also
very
very
helpful
here.
A
But
again,
additional
reviews
are
definitely
very
welcome
on
this
one
and
then
finally,
there's
one
from
Candace
on
TLS
I
would
love
to
see
this
one
make
it
in
as
well.
I've
been
talking
with
Candace
some
on
slack
and
I
think
this
is.
D
A
Very
close,
just
a
few
changes
left
on
this
one
I
think
I
don't
know
Candace.
Do
you
have
any
any
updates.
A
Cool
cool,
yeah
I
would
like
I,
think,
there's
a
decent
chance.
This
one
merges
this
week
so
yeah
very
excited
to
get
this
one
through.
It's
been
a
very,
very
long
process.
Thank
you
for
your
patience
on
this
one
Candace
and
for
continuing
to
push
and
get
this
one
in
and
yeah.
This
is
a
pretty
key
feature
that
we
want
to
have
representable
in
Gateway,
so
I'm
excited
for
this
one
cool.
C
A
D
A
Cool
some
really
quick
one
last
thing
for
me
is
kubecon
I,
don't
know
how
many
people
on
this
call
are
going
to
be
there,
but
it'll
be
helpful
to
maybe,
if
you
just
want
to
use
the
raise
hand
feature
or
something
like
that,
if
you
know
that
you're
going
to
be
able
to
make
it
we're
trying
to
find
some
way
to
do
some
kind
of
low-key
event
celebration,
something
because,
ideally
this
kubecon
is
where
Gateway
goes
GA
and
there
and
the
whole
point.
A
The
whole
thing
with
Gateway
is
more
than
any
kubernetes
API
before
it.
It's
on
the
you
know
this
is
this:
API
has
been
you
know,
developed
collaboratively
with
just
a
lot
of
different
people,
other
kubernetes
apis.
It's
been
a
few
people
that
have
largely
pushed
it
forward.
A
Add
this
API
in
particular,
has
been
the
strength
of
the
community
and
everyone
contributing
and
so
would
love
to
find
some
kind
of
way
to
recognize
that,
but
just
as
a
starting
point
would
be
helpful
to
to
understand
who,
if
anyone
is,
is
going
to
be
able
to
be
there.
So
we
can
get
a
get
a
high
level
understanding
of
who
we
can
celebrate
with.
Basically
so
seeing
a
couple
hands,
ain't.
A
Gateway
party
yeah
yeah
yeah,
it's
early,
November
I,
think
yeah
November
6th
looks
like
the
right
start
date
in
Chicago.
A
Cool
all
right
and
and
I
should
also
say,
I
plan
on
being
there,
but
you
know
obviously
nothing.
Nothing
is
completely
certain,
but
that
is
definitely
my
intent.
A
Cool
all
right
next
up,
I
had
just
a
few
things
in
triage.
Let's
see,
I,
don't
remember,
okay,
this
one!
Thank
you,
I'm,
guessing
Leora
added
this
one
or
maybe
I
I,
don't
know
who
did
but
anyway,
this
one
is
just
needing
an
approval.
I
think
yeah.
A
B
A
So
yeah
anyone
who
is
a
conformance
approver
can
approve
this
one
and
get
it
and
then-
and
we
already
have
implementations
of
this
I
think
you
said
it
was
contour
and
istio
so
far.
Is
that
correct.
B
B
Oh,
that's,
it
is
test
it
is
tested
with
Contour,
as
you
see
for
the
pr
description.
Okay,
so,
okay,
yes,
I,
don't
know
what
is
to
state
but
I.
Think
ACS
also
supported
so.
D
A
A
For
a
conformance
test,
I
well
yeah,
actually
we
have
been
adding
conformance
release,
notes
so
yeah.
It
probably
should
have
a
release
note
here
and
then
the
the
other
thing
I'm
not
actually
going
to
edit
this
live,
but
sometimes
what
I've
seen
happen
is
you'll
get
just
a
general
note
for
anyone.
You'll
get
just
a
code
block
like
this,
which
looks
the
same
to
any
reviewer
but
is
not
actually
effective
and
is
not
picked
up
by
our
release,
notes
generator.
A
A
Okay,
next
up
is
which
one
oh
yeah.
This
is
Nick's
handling
for
indistinct
parent
refs.
This
is
one
that
just
needs
a
review.
A
But
this
is
again
trying
to
say
hey
if
you
have
more
than
one
parent
ref
referencing
the
same
parent
and
those
are
not
distinct,
they're
overlapping
in
some
form,
then
you
should
really
be
considering
that
an
error
and
raising
this
condition,
and
not
just
the
spec
change,
but
a
corresponding
conformance
test
change
as
well.
A
So
this
one
needs
needs
review.
I,
guess
this
is
the
high
level
one
but
I
think
it's
a
good
change
that
we
want
to
include
in
1.0
and
then
finally,
this
last
one
is
redirect
loop
protection.
Somebody
pinged
on
this
pretty
recently
and
I'm.
Trying
to
remember
I,
know:
we've
discussed
this
before.
Does
anyone
remember
what
our
conclusion
was
here?
I
feel
like
the
conclusion
was
that
this
was
this
did
not
that
that
we
did
not
have
or
implementations
did
not
have
a
good
way
to
consistently
solve
this.
A
But
I
I
can't
remember,
or
there
were
too
many
too
many
potential
edge
cases
to
really
cover
anyone
happen
to
remember
any
more
context
on
this.
One
I
would
love
to
update
it,
because
we,
it
would
be
good
to
have
some
kind
of
guidance,
whether
it's
this
is
something
we
can't
provide
or
we
have
a
path
forward
and
it
just
hasn't
been
prioritized,
don't
see.
Travis
on
this
call.
A
Based
on
silence,
I'm
going
to
go
ahead
and
say
no
one
else
remembers
either,
which
is
unfortunate,
but
that
would
appreciate
some
some
comments
or
thought
on
on
this
one
as
well.
Just
so
we
don't
lose
track
of
it.
A
And
with
that
I
think
we
have
made
it
all
the
way
to
the
end
end
of
our
agenda
today,
any
last
items
before
we
give
everyone
everyone
some
time
back.
A
Cool
good
good
luck,
I,
don't
I,
still
don't
know
how
you're
going
to
update
the
invites,
but
I
I
wish
you
luck
in
in
figuring
out
our
our
funny
calendar
schedule.
Thank
you,
yeah
yeah
cool
all
right!
Well,
thanks
everyone
for
meeting
today,
it's
been
very
helpful
and
yeah.
If
you
have
any
exercises.
Reviews
on
these
PRS
are
very
welcome.
As
always,
and
with
that
I
hope
we'll
see
some
of
you
at
kubecon
and
some
of
you
next
week
as
well
have
a
great
rest
of
your
week.
Cheers.