►
From YouTube: Kubernetes SIG Multicluster 2020 Nov 10
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
B
So,
jeremy,
I
think
you
have
the
only
item
on
the
agenda
today,
so
you
can
take
it
off
whenever
you're
ready.
A
Awesome
I
need
to
screen.
B
Share
well,
you
now
have
that
capability.
A
Perfect
so
yeah,
so
I
think
this
might
end
up
being
shorter,
but
just
kind
of
wanted
to
come
back
to
the
cluster
id
dock
that
we
talked
about
a
couple
weeks
ago
now,
a
few
weeks
ago,
been
a
crazy
couple
weeks.
Time
has
flown,
but
there's
been
some
updates,
so
it
seemed
like
last
time
we
discussed
the
dock
and
and
just
kind
of
for
context.
A
The
the
idea
with
with
cluster
id
was
that
maybe,
instead
of
just
trying
to
create
some
cluster
id
resource,
we
we
look
at
this
as
individual
claims,
so
some
new
crd
that
basically
just
holds
a
bit
of
information
with
a
with
a
name
in,
in
a
few
cases,
well-known
names
that
we
can
use
to
highlight
certain
pieces
of
information
that
we
think
are
are
useful,
basically
everywhere
and
the
two
that
we
kind
of
came
down
to
were
this
id
dot,
kate's
dot,
io
claim
that
would
be
a
clusters
identifier,
so
a
valid
dns
label
and
unique
within
a
cluster
set,
but
but
preferably
unique.
A
You
know
in
a
broader
scope
within
your
organization,
or
you
know
whatever
seems
reasonable
there
and
then
the
cluster
set
kates
dot
io,
which
would
be
the
the
identifier
of
the
cluster
set
to
which
a
cluster
belongs,
and
that
might
be
like
the
id
of
of
that
cluster
within
the
cluster
set
or
it
might
be,
the
id
of
the
cluster
set
itself
shared
between
clusters.
Basically,
this
identifies
to
the
cluster
which
cluster
set
it
belongs
to.
A
I
don't
know
that
we
need
to
go
beyond
that
in
terms
of
like
defining
exactly
how
this
gets
used.
A
So
that's
that's
kind
of
the
background
and
then
one
of
the
open
questions
was
this
idea
of
a
of
a
signature,
so
it
seemed
kind
of
non-controversial
to
have
a
simple
crd
with
a
with
a
name
to
identify
the
claim-
and
maybe
you
know
you
could
add
your
own
foo.com,
whatever
or
or
some
some
personal
claims
to
extend
these
well-known
claims.
You
know
we
wanted
to
leave
that
open
and
some
value
some
claim
dependent
string.
A
That
can
be
whatever
format
is
necessary
for
that
claim,
but
the
I
guess
the
open
question
was
whether
we
should
actually
have
a
signature.
A
So
you
know
it
seemed
reasonable.
I
think
that
there
would
be
some
way
to
validate
that
this
claim
is
real,
but
basically
what
it
what
it
came
down
to
last
time
was:
I'm
certainly
not
an
expert
in
this,
and
it
didn't
seem
like
anyone
on
the
call
really
you
know
was,
was
invested
in
this,
so
we
we'd
reach
out
to
sigoth.
A
So
I
I
spoke
to
mike
denise
and
the
takeaway
was
basically
that
you
know
a
signature
in
this
case
may
not
be
that
useful,
because
anyone
like
what
what
are
we
verifying
it
against
and
there's
some
some
ways
that
could
be
defined,
but
the
takeaway-
and
I
think
he
actually
commented
here-
was
there's
no
real
reason
why
he
didn't
really
see
the
immediate
use
case
or
value
here
and
there's
no
reason
why
we
couldn't
add
this
later.
A
B
And
I
have
the
same
thought
in
my
head
around
signing
in
terms
of
like
what
would
we
validate
it
against
right,
typical,
like
easier
to
add
later
than
to
force
yourself
to
think
of
it
now
and
there's
not
like
a
clear
answer.
A
Exactly
and
and
presumably
it's
set
by
someone
like,
presumably
you
don't
give
everybody
right
permissions
to
these
claims,
and
so
you
know
there's
some
level
of
admin
access
required
to
modify
these
and
it
seems
like
it
would
be
hard
to
make
it
so
that
that
same
admin
couldn't
fake
them.
So
you
know
that's
that's
one
one
side
of
why.
A
Maybe
that
should
wait,
and
then
you
know
in
terms
of
like
a
cluster
set,
if
you
did
have
a
registry
or
something
it
seems
like,
the
registry
would
really
be
responsible
for
for
verifying
this
claim,
not
the
not
the
cluster
itself,
so
yeah
yeah,
nice
and
simple
this
way-
and
there
were
some
comments
too,
on
from
michael
here-
suggestions
that,
like
maybe
we
we
consider
like
a
region.kates.io
to
make
it
easy
to
figure
out
where
cluster
is.
A
You
know
things
like
that
and
well
I
don't
want
to
get
into
that
here
because
I
feel,
like
you
know,
we
could
come
up
with
a
very,
very
long
list
of
potentially
useful,
well-known
names
this.
This
opens
that
door,
so
we
can
start
saying
like
hey
what
are
the
other
useful
values
that
I'd
like
to
know
at
cluster
level.
B
I
think
that
I
like
the
idea
of
exploring
other
ones.
It
seems
like
a
separate
thing
than
cluster
id
to
me.
A
Exactly
but
once
we
have
this,
hopefully
it's
easier
to
to
add
those,
because
we've
got
this
concept
of
like
a
cluster
claim,
and
then
you
know
once
we
get
this
through
and
and
we
start
making
this
a
real
thing.
You
know,
then
we
can
have
other
separate
conversations
about
what
other
claims
would
be
useful
to.
C
Have
so
can
I,
I
think,
plus
your
claim
might
have
been
my
original
naming,
which
you
should
not.
Let
me
name
things
because,
as
I
stare
at
it,
I
realize
we
have
one
other
thing
called
a
claim
and
it
means
something
entirely
different,
which
is
a
persistent
volume
claim.
So
maybe
claim
is
the
wrong
word
here,
maybe
like
fact,
or
assertion
or
statement,
or
something
that
isn't
a
claim.
A
Yes,
I
so
I
wanted
to
basically
get
to
the
point
today
where
we
turn
this
into
a
kep
and
I'll
I'll,
write
that
out
and
basically
take
this
and
put
this
into
a
cap
and
a
big
part
of
that
is
coming
up
with
a
better
name.
I
don't
know
what
the
name
should
be:
I'd,
love
input,
yeah
cluster
cluster
value,
cluster
field.
I
don't
know,
there's
there's
a
lot
and
there's
a
lot
of
options
and
I'm
not
sure
if
any
are
great-
and
you
know
if
that's
the
hardest
part
of
this.
B
My
thinking
too,
and
this
might
be
a
another
good
thing
to
like
sorry,
I
was
reading
the
comment.
It
might
be
another
good
thing
to
do
like
a
community-based
naming
exercise.
A
Yep
yeah
exactly
like
this,
you
know
we
could
spend
a
lot
of
time
on
this
and
I
think
yeah
just
we'll
get
really
good.
I
think,
with
multi-cluster
stuff
about
sending
out
those
surveys,
but
yeah
we
can
figure
that
out
once
once.
We
kind
of
agree
that
this
is.
This
is
a
good
direction
and
I
I
do
think
claim
probably
makes
less
sense
too,
because
without
the
signature
it
doesn't
feel
so
much
like
a
claim,
but
yeah.
C
A
Cool
all
right,
then
I'll
get
working
on
that
and
share
it
out
as
soon
as
it
is
ready.
B
So
here's
one
thing
that
I
think
it
would
be
good
to
cover
in
the
cap
or,
if
not
cover
in
the
cap,
maybe
we
can
just
like
reaffirm.
This
is
a
thing
that
that
we
should
be
thinking
about.
So
we've
had
like
this
threat
of
discussion
over
like
last
several
years
around
like
the
registry
resource
and
though
I
think
our
latest,
like
conventional
wisdom
and
sort
of
proto-consensus
within
this
group,
is
that
we
wanted
to
avoid
a
premature
standardization
on
a
registry
resource.
B
If
we
have
a
way
to
express
what
the
id
of
a
cluster
is,
it
might
be.
You
know
a
fruitful
next
step
to
think
about.
How
would
we
use
that
as
a
coordinate
in
apis
that
need
to
make
references
to
clusters.
A
Yeah-
that's
that's
a
good
point,
so
so
I
here
I
have
these
like
high-level
user
stories.
I
think,
as
part
of
this
kept
and
and
making
it
real,
it's
like
bringing
examples
to
these
user
stories
like
how.
How
does
this
actually
solve
those
problems.
A
Yeah
yeah
yeah,
absolutely
okay.
I
have
some
ideas,
so
why
don't
I
get
this
down?
You
know
into
a
pr
and
then
yeah
like
it
would
be
good
to
have
other
people
chime
in
with
potential
use
cases
and
and
and
patterns
that
they
think
makes
sense.
I
would
love
help
on
this.
A
But
yeah
it
might
be
easier
to
do
in
a
on
a
cap
versus
versus
this
dock.
C
A
B
Excellent
sounds
like
we're
kind
of
talked
through
unless
anyone
has
extra
stuff
for
the
agenda,
but
I
think
that
was
the
only
agenda
item
for
today.
B
Before
we
sign
off,
let
me
just
make
a
call
for
if
you
have
any
demos
you'd
like
to
share
that,
are
you
know
relevant
to
the
conversation
or
something
that
you
know
maybe
is
like
a
next
step
or
further
down
the
road
that
you're
thinking
about,
like,
I
think,
we'd
love
to
see
them,
especially
things.
You've
may
have
already
built
to
solve
a
problem
that
you
face.