►
From YouTube: Kubernetes SIG Multicluster Feb 2 2021
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).
B
C
Before
you
guys
get
started
just
wanted
to
let
you
know
this
is
my
first
ever
sync
multi-cluster
meeting,
so
I'm
really
looking
forward.
Thank
you.
I'm
really
looking
forward
to
this
welcome.
Thank
you.
Thank
you.
Hi.
My
name
is
sri.
I
work
at
intuit
and
yeah.
We
have
a
huge
kubernetes
setup
over
300
clusters
and
yeah
I
mean
a
lot
of
a
lot
of
the
things
about.
Multi-Cluster
is
something
I
want
to
keep
myself
aware
about,
and
maybe
even
you
know,
be
part
of
community
and
maybe
contribute
in
whatever
way.
I
can
thanks.
B
B
Foreign
all
right,
so
it's
three
minutes
after
why?
Don't
we
get
started
so
hello?
Everyone,
I'm
jeremy,
olmsted
thompson,
it's
february
2nd
2021,
and
this
is
the
sig
multi-cluster
weekly
meeting.
B
A
Chrome
here
we
go
yeah,
so
mainly
I
want
to
revisit
basically
what
we
were
talking
about
last
week.
I
think
several
people
on
the
call
were
there
then,
but
just
to
reiterate
we
were
talking
about
for
the
cluster
id
cap.
When
should
we
be
targeting
a
release?
A
There's
apparently
some
discussions
going
on
about
121
that
we
would
sort
of
need
to
make
some
decisions
about
if
we
wanted
to
be
like
flagged
for
that
and
this
kind
of
spun
off
into
a
conversation
about
whether
we
wanted
to
pursue
going
in
tree
or
a
six
repo
so
and
partially
in
pursuit
of
that
question,
but
also
generally
to
get
the
word
out
a
little
bit
more
about
the
cap.
A
I
went
to
cluster
apis
office
hours
last
week
and
talked
to
them,
and
I
also
talked
to
one
of
the
chairs
of
sig
cluster
life
cycle,
though
their
formal
meeting
isn't
actually
until
next
tuesday,
but
in
general,
I'm
getting
the
sense
from
the
cluster
api
sub
project
and
the
chaop
subproject
that
something
like
this
would
be
useful
to
them.
I
got
some
specific
use
cases
from
cluster
api.
A
The
k
ops
case
was
like
more
tied
to
like
dns
but
seemed
like
it
would
kind
of
be
in
our
purview
at
some
point,
or
at
least
like
our
implementation,
might
be
in
their
purview
at
some
point,
so
I
do
think
on
a
at
a
temperature
level.
I
got
the
perspective
of
like
yeah.
A
We
would
totally
use
your
thing
if
you
have
a
thing
for
for
the
for
this,
and
we
have
some
needs
for
this
type
of
thing,
and
I
got
some
opinions,
I
would
say
about
entry
versus
sigs
repo,
but
I
would
say
that
overall,
it
was
a
little
ambivalent
about
whether
it's
really
required
by
any
of
these
individual
sub
projects
for
it
to
be
entry
for
it
to
be
useful
to
them.
So
I
only
know
a
little
bit
from
basically
those
opinions
I
got
about
whether
you
know
one
is
harder
than
the
other.
A
So
I'm
not
sure
that
I
have
a
definitive
answer
or
like
a
resounding
vote
from
the
outside
of
like
yes
put
this
entry
or
we
won't
use
it.
So
I
think
that
that
decision
is
basically
still
in
our
court
and
so
I'd
like
to
see
if
I
can
figure
out
what
we
would
need
to
do
or
what
consensus
we
can
come
to
to
make
that
decision.
B
Awesome
I'm
curious
if
anyone
here
has
any
thoughts
on
in
tree
versus
editory.
A
Yeah,
and
does
anybody
have
any
experience-
I
guess
too,
of
either
pushing
something
through
or
having
troubles
or
not
troubles
using
something
that
was
entry
and
core
or
not
injury?
That
might
be
worth.
A
B
From
from
my
perspective,
you
know,
I
think,
there's
a
lot
of
utility
either
way.
I
think
it.
You
know
it's
pretty
easy
to
build,
depending
on
a
crd
that
users
would
just
need
to
install
with
whatever
tool
they're
installing
that
uses
it.
I
don't
know
if
there
would
be
any
like
core
components
immediately
taking
advantage
of
of
cluster
id.
B
That
would
be
a
good
signal
for
for
going
entry,
but
you
know
if
it's
if
the
goal
is
to
have
something
that
you
can
assume
available
in
in
every
cluster.
Obviously,
entry
built-in
is
the
you
know
the
easiest
way
to
to
make
that
the
case.
I
know
there's
also
some
discussion
going
on
about
how
to
not
necessarily
have
something
in
tree,
but
still
make
it
part
of
core
kubernetes,
but
I
don't
think
that's
really
been
solved
yet
yeah.
B
Maybe
it
would
be
a
good
idea
to
reach
out
to
some
of
the
api
machinery.
Folks.
A
Okay-
and
I
did
get
one
other
piece
of
advice-
if
the
point
of
kind
of
traveling
to
these
other
cigs
was
to
get
more,
you
know,
community
awareness,
is
there
a
stage
and
potentially
only
if
we
consider
going
entry
where
we
mail
this
out
to
like
the
whole
kubernetes
dev
list?
A
B
I
would
start
there,
I
think,
and
get
their
thoughts,
because
that
might
also
shut
it
down.
Real,
quick
and
let
me
know
yeah
because
I
know
we're
you
know
we're
trying
not
to
pile
on
core
apis.
A
Right
yeah,
I
had
my
understanding
also
through
some
of
these
conversations
was
that
core
is
trying
to
split
itself
up
a
little
bit
more,
and
so
we
might
be
getting
into
the
middle
of
that.
I
guess.
B
Cool
yeah,
why
don't
we
also
come
back
to
this?
Maybe
after
we
talk
about
naming
for
a
bit
because
I
think
yeah
it's
gonna
join
shortly
and
I'm
curious
what
he
thinks.
Oh
sure,.
A
Okay,
so
on
the
naming
point
I
brought
this
up
to
cluster
api
and
you
know
got
some
in
chat,
joke
answers
about
exciting
names.
That
would
certainly
not
make
the
cut,
but
the
sheet
remains
in
its
state
and
we
had
said
we
wanted
to
kind
of
open
it
up
a
little
bit
to
those
groups,
but
then,
if
it
still
stagnated,
we
could
just
start
to
actually
have
the
discussion
of
do.
A
We
need
to
cut
any
of
these
off
and
then
what's
the
means
by
which
I
should
initiate
the
democracy
and
get
get
votes
so
yeah.
So
I
guess
I
think,
if
we
can
make
this
a
little
bit
bigger,
but
I
think,
if
there's
anything
on
this
list
that
we
need
to
like
totally
kick,
there
seems
to
be
like
a
fundamental
concept
of
long
names
versus
shortened.
Somehow
is,
but
otherwise
it's
kind
of
no
like
major
patterns
of
like
difference
between
things
from
what
I
see.
A
Yes,
but
this
is
a
sheet
to
pick
the
one
name
for
the
one
crd
that
we're
trying
to
make
but
yeah
and
as
some
background
since
this
is
your
first
time,
definitely
sorry,
but
we're
trying
to
create
this
crd
to
store
information
about
a
cluster's
id
like
giving
it
some
name
and
then
also
its
membership
in
a
multi-cluster
setup.
A
A
B
A
B
A
cluster
set
right,
you
know
I,
I
was
expecting
at
least
one
name
that
was
kind
of
out
there.
A
B
Yeah,
I
don't
know
that
there's
anything
here
that
that
we
really
need
to
cut,
maybe
just
to
shorten
the
list.
D
B
B
And
then
maybe
it's
just
time
to
share
this
out,
I
mean
clearly
nobody
either.
All
of
these
are
fine
or
right
or
there's
no
real
opinions
right.
B
What
we've
done
in
the
past
is
a
survey
monkey
send
out
to
the
to
the
sig
and
you
could
go
broader.
I
I
can't
remember
if
we
actually
sent
it
to
everybody
or
not,
we
may
have
that.
Might
it's
also
a
good
way
to
get
people
interested
because
hey
hey,
come
give
your
thoughts
and
a
name
also
makes
people
read
the
cap.
B
So
yeah,
that's
probably
just
a
good
place
to
start.
If,
if
some
wild
suggestions
come
in
there,
I
think
last
time
we
added
a
like
suggest
your
own
box
and
some
wild
suggestions
come
in
then
we
can
always
reopen
the
spreadsheet
and
I
think.
A
A
A
This
one
that
it's
so
dry,
I
guess
that
we
may
still
not
get
a
clear
winner,
but
the
only
reason
I
bring
that
up
is
if
we
want
to.
If
we
want
me
to
structure
the
poll
that
it's
not
like
vote
for
your
favorite,
but
it's
like
rank
them,
and
then
we
could
actually
do
something
like
ranked
choice,
voting
or
something.
If
we
wanted
to
later.
A
A
Yeah,
that's
fine
sounds
good,
sweet,
okay!
Well,
I
can
make
a
survey
monkey
and
that
sounds
great.
B
B
D
I
guess
my
gut
would
be
we
if
we
want
it
to
be
universal.
Ideally
we'd
wind
up
entry,
but
I
probably
wouldn't
start
there
not
at
least
until
it's
beta
just
my
own
gut.
A
I
received
like
a
sort
of
like
not
offhand
but
like
very
brief
comment
that,
like
the
process
to
go
entry
and
outside
is
like
totally
different
is
that
are
we
still
in
with
going
all
the
way
to
beta
still
be
the
time
where
we
can
make
that
jump,
or
we
don't
know,
or
is
that
too
late?.
B
I
I
think
you
can
definitely
make
the
jump
at
beta.
I
mean
another
way.
To
put,
it
would
be
that
going
out
of
tree
has
a
lot
less
process,
and
so
you
can
kind
of
get
the
process
off.
I
don't
know
that
it's
different
process,
as
so
much
as
just
there
isn't
process.
We
can
kind
of
steer
the
ship
in
a
six
repo.
A
B
B
We
could
go
alpha
and
in
a
six
repo
pretty
quickly
and
then
and
then
revisit
at
the
beta
timeline.
I
I
do
think
it's
probably
worth
checking
with
the
sig
api
machinery
folks
to
see
what
they
think
and
you
know
what
is
there
even
appetite
in
the
beta
timeline
for
entry
with
the
way
things
are
going,
trying
to
split
out
various
components.
A
A
B
Sounds
good
great!
Well!
That's
that
I
guess
on
that
note,
though,
what
is
left
on
cluster
a
or
aside
from
a
name,
I
guess
from
well,
what's
what's
left
from
from
calling
it
implementable
now
that
we've
got
some
feedback
from
from
other
sigs.
A
A
A
D
The
work
api
repo
it
got
created,
so
we
can.
We
can
begin
to
move
forward
on
that
work
and
I
don't
think
shojin
is
here
or
valerie
or
rainbow
mango,
but
everybody
in
the
group
can
just
know
that
that,
like
we
did
get
that
step
completed
so
that
unblocks
us
from
doing
the
things
we
previously
agreed
to
around
translating
children's
doc
into
a
markdown
and
starting
to
put
some
code
around
this
in
kubernetes
repo.