►
From YouTube: Kubernetes SIG CLI 20210324
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,
good
morning,
good
evening,
good
afternoon,
depending
on
where
you
are
today
is
march
24th-
and
this
is
another
of
our
sixth
july
bi-weekly
calls-
my
name
is
jeremiah.
My
name
is
mache
and
I'll
be
your
host
today.
So
our
agenda
is
pretty
packed.
So,
let's
let's
go
through
it.
A
B
Well,
so,
following
up
on
our
brainstorm
in
conversation
from
last
week
about
two
weeks
ago,
about
mentoring
and
bringing
in
new
folks
from
underrepresented
backgrounds,
I
started
working
with
the
sig
contrebex
sub
project
for
mentoring
and
they're
working
on
a
virtual
online
recorded
written
new
contributor
workshop
that
they
need
help
with.
So
if
anyone
wants
to
get
involved
and
help
out
with
that,
there's
a
project
board
there
and
the
meeting
is
on
monday
at
9
00
a.m.
B
Pacific
time,
I
believe
so
feel
free
to
hit
me
up
if
you
want
to
help,
but
yeah
they've
put
out
a
call
for
help
to
get
some
of
this
content
written
up
and
recorded.
So
I
think
there's
value
in
us
nurturing
and
mentoring.
Some
more
folks.
C
Okay,
nick
you
want
to
talk
about
the
koi
movement
removed.
D
So
yeah
the
link's
there
thanks
everyone
for
the
help.
I
know
it
took
a
while
a
lot
of
dots
to
cross
and
tease
the
dot
whatever
for
there.
The
I
think
there's
a
couple
of
things
still
to
do
to
sort
of
finalize,
so
we
might
make
a
slack
team
channel
for
kui,
but
that
may
make
sense
to
have
a
place
to
discuss
issues
and
suggestions
and
so
we'll
get
around
to
that
pretty
soon.
D
A
Yeah
awesome-
that
is
very
that's
very
awesome
news.
I
remember
that
when
we
started
the
discussion
I
was
oh,
I
don't
know
like,
maybe
even
two
years
ago,
so
that's
an
excitement.
Development,
yeah.
D
A
And
the
last
link
that
I
pull
in
the
meantime,
we
have
our
six
july
annual
report
ready
for
review
I'll,
be
very
happy
if
you
get
it
a
couple
spend
a
couple
minutes
reading
through.
If
you
think
that
there's
something
missing
in
that
report
feel
free
to
to
comment
I'll,
be
addressing
those
this
or
next
week,
and
that's
if
I
remember
correctly,
we
have
to
close
them
by
the
end
of
month
or
something
along
those
lines.
I
need
to
double
check
the
dates.
D
One
thing
just
started
interrupting:
I
forgot
to
mention
ming
ting
who's
been
a
very
valuable
contributor
to
the
project
is
with
us
today.
So
just
wanted
to
give
her
the
shout
out.
A
Okay,
well,
that
brings
us
to
nicely
to
the
introduction
section.
So
I'll
probably
let
him
speak
since
I'm
guessing
he's.
First
time
in
60
like
holes,
take
away
from.
A
A
I'm
not
sure
if
men
think
you
were
trying
to
say
something.
I
barely
heard
you
talking.
I
mean
if
you're
new,
to
the
call
just
introduce
yourself
where
you're
coming.
What
are
your
expectations?
I
think
nick
throw
you
a
little
bit
under
the
bus,
so
I
just
use
your
name
to
start
off
with
the
introductions.
E
I've
been
using
kubernetes
in
our
company
and
I'm
working
in
one
of
the
kubernetes
platform
things
in
india,
so
I
wanted
to
see
how
I
could
contribute
into
the
core
kubernetes
cli,
so
yeah
as
part
of
that
I
was
exploring
and
got
some
links
over
the
slack
channels
and
landed
up
here
first
time.
Thank
you.
A
Oh
man
thinks
route
wrote
that
he
his
audio
has
problems.
Okay,
no
worries
you
want
to
give
it
a
try.
Another
one.
F
Wow
you're
putting
me
on
the
spot
here,
buddy,
of
course,
hi
everybody,
my
nickname
is
dems.
You
know
I
show
up
in
different
meetings,
sometimes
just
to
get
a
feel
of
how
things
are
going
on
and
get
up.
You
know
sense
of
you
know
how
the
pro
you
know
progress
was
made,
I
guess,
and
in
the
sixth
I
don't
have
anything
space
specific
to
talk
about
today.
I
definitely
was
interested
in
the
topic
that
eddie
was
talking
about
about
mentoring
and
things
like
that.
F
So
kudos
to
that
thanks
for
getting
that
rolling
and
if
you
need
any
help,
please
reach
out
thanks
a
lot.
A
Cool
thanks
a
lot
thomas.
Do
we
have
anyone
else
new
on
the
call,
I
think
I
recall
most
of
the
names,
but
maybe
I
miss
someone.
If
you
want
to
speak
up
now's
your.
B
So
if
folks
don't
know,
lori
apple
is
just
amazing
and
awesome.
B
But
this
will
let
us
kind
of
like
take
the
reins
and
control
the
narrative
there,
and
so
lori
is
going
to
be
able
to
join
us
for
a
future
meeting.
I
don't
think
she
was
able
to
join
us
today,
but
things
like
like
our
cobra
command
separation,
sean
that
we've
been
wanting
to
do
separating
cube
control
commands
from
cobra
like
that's
something
that
we
should
drive
via
kept
and
then
scope
out
and
document.
B
The
work
and
lori
said
that
we
will
be
surprised
at
the
people
that
basically
show
up
and
we'll
do
the
you
know
well-defined
work
for
us.
So
it's
a
benefit
there.
A
I
mean
we
did
something
similar
and,
if
I
remember
correctly
dug
and
a
couple
other
folks
were
driving
the
effort
of
create
commands,
and
it
was
a
a
perfect
example
how
this
can
roll
and
we
got
lots
of
contributors
helping
with
that.
But
I'm
totally
in
support
of
of
doing
that.
If,
if
we
can,
you
know
have
a
because
the
code
recommends
stuff
once
we
have
that,
I
don't
know
one
command
ready.
A
G
So
the
question
I
have
for
this
specific
example
as
a
cap
is
that
it's
we're
basically
doing
a
refactor
with
removing
this
cobra
command
stuff
from
the
options
and
and
keps
as
is
in
the
name,
is
an
enhancement,
so
they
seem
like
they're
in
two
totally
different
lanes.
G
I
guess
I
could
see
how
you
might
want
a
refactor
as
a
as
a
cap,
but
but
they
kind
of
seem
like
in
general,
two
completely
different
spheres.
G
A
I
was,
I
was
rather
thinking
because
here's
the
thing
we
laid
out
several
roles,
what
a
cube
cuddle
command
should
look
like
over
the
past
two
three
years.
A
A
lot
of
the
comments
went
through
the
transformation
up
to
a
certain
point.
For
example,
the
complete
validate
run,
I'm
pretty
sure
the
majority
of
people
is
already
familiar
by
now.
We
could
probably
create
a
cap
not
as
a
targeted
for
this
particular
thing,
but
rather
are
as
a
documentation.
A
What
a
proper
cubecutter
command
could
look
like
and
then
divide
the
work
into
what
this
is,
the
idea
of
where
we
want
to
be
in
we're
not
there
and
slowly
moving
towards
that
direction.
That's
that
would
be
my
take
on
to
it.
B
Yeah-
and
I
I'm
sure
dims
can
also
give
us
a
better
explanation
of
the
caps
for
that
too.
Lori
was
telling
me
basically
anything
that
makes
kubernetes
better
or
changes
things
we
could
do
through
a
cap.
It's
a
tool
that
we
can
use
certain
leverage
to
organize,
just
like
the
the
migration
with
customize
right,
like
that
was
a
big
massive
change
that
we
didn't
have
a
kept
for
that
we
definitely
should
have,
and
the
release
team
has
been
very
confused
about.
B
You
know
such
a
big
change
getting
in
without
a
cap,
so
I've
been
dealing
with
that
behind
the
scenes,
but
yeah,
I
think
I
think
it
just
we
could
make
a
cap
for
everything
right
like
there's.
Definitely
some
changes.
We
do
that
are
pretty
small
scoped,
but
I
don't
think
anything
is
like
out
of
scope
for
sure.
B
G
So
I
mean
there's:
there's
a
non-zero
cost
by
doing
it
with
a
cap.
Right
I
mean
it's,
it's
added
bureaucracy,
and,
and
so
there
there
should
be
a
benefit
to
this
added
cost,
and
and
so,
if
yeah,
I
I'd,
be
slightly
hesitant
to
like
throw
everything
in
a
cap
just
because
you
know
that's
actually
going
to
slow
us
yeah,
it's
going
to
be
pretty
bureaucratic.
H
I
think
we
can
delete
a
lot
of
the
kept
body
like
I
agree
if
you
were
to
fill
out
like
the
rollback
strategy
for
this.
You
know
refactoring
of
code
like
it's
nonsensical
right,
but
the
like.
What
is
the
motivation
for
doing
this,
like
you
could
put
like
two
sentences
in
there
and
then
like?
What
are
the
steps
like
it's?
If
you
just
want
to
like
way
of
documenting
like
here's,
a
body
of
work,
that's
going
to
be
done
and
why
it
seems
it
seems
like
a
reasonable
approach.
F
Hi,
I
just
wanted
to
add
one
observation
here
which
is
like
when
you
need
some
cooperation
with
other
things
on
like
how
to
proceed
with
certain
things
like,
for
example,
I
don't
know
if
we
have
decided
or
designed
how
exactly
cube
cutter
will
ship
once
it
lives
out
of
tree
right.
So
that's
like
a
good
thing
that
we
could
like
write
down
how
that's
going
to
be
done
between
sig
release
and
6
cli.
I
Okay,
so
we
have
like
it's
something:
that's
going
to
show
up
in
a
release
that
needs
a
cap,
it's
something
that
requires
interesting
collaboration
that
needs
a
cap
or
it's
something
that
we're
looking
for
additional
contributors
that
will
need
some
documentation
about.
What
are
we
even
doing
here?
How
do
I
get
involved?
What's
this
supposed
to
look
like
those
are
three
reasons
why
we
need
it.
A
F
So
much
we
can
so
there
is
space
in
cubanetus
community
and
and
to
to
write
down
six
specific
things
too
right.
Sig
architecture
has
a
bunch
of
things
there.
So
that's
another
place,
oh
if,
if
you're
worried
about
like
how
many
approvals
do
we
need,
like
other
people
to
approve
and
like
other
people,
to
take
a
look?
F
So
you
know
if
you
want
to
avoid
that,
then
the
community
is
a
good
space
to
document
this
kind
of
a
thing
which
is
you
know
you
just
need
a
gtmr
pro
from
a
couple
of
people
and
you
are
like
updating
some
documentation
which
will
help
out
newcomers
right.
So
that
kind
of
thing,
so
one
example
that
we
are
doing
right
now
is
like
we
are
trying
to
in
from
seek
architecture.
F
We
are
writing
down
like
feature
gates,
dot
go
you
know
what
are
the
fields
in
there
and
how
are
they
going
to
be
used?
You
know
when
you
go
from
alpha
to
beta
what
happens
so
we're
trying
to
write
that
down.
So
we
are
using
that
space
for
it.
A
B
Yeah
and
then
so
two
quick
things,
so
I
just
dropped
two
links
into
the
chat
I
so
one
of
the
struggles
that
I
had
coming
into
the
project
was
all
of
the
historical
decisions
that
were
made
that
were
just
not
documented
or
shared
anywhere,
and
so
you
know
this
idea
of
like
an
adr,
an
architectural
design
record,
and
that's
how
I
view
like
a
lot
of
the
the
things
that
we
can
do
via
caps
as
well
to
kind
of
leave
like
a
trail
of
breadcrumbs
for
people
yeah.
B
I'm
not
saying
these
need
to
be
like
massive
things,
and
we
definitely
don't
need
to
do
them.
For
you
know
every
single
change
we
make,
but
I
think
that
there's
value,
especially
for
future
people
in
terms
of
like
being
able
to
get
up
to
speed
and
see
where
things
happened,
and
so
I'm
just
I'm
saying
that
we
give
it
a
try,
because
lori
seems
to
think
it
will
help
us
a
lot
and
laura's
really
smart.
B
The
other
link
I
dropped
in
is
this
was
something
that
tim
shared
in
one
of
the
sig
leads
meetings,
and
especially
for
the
the
folks
who
are
further
up
the
contributor
ladder
right
now,
like
the
approvers
and
the
other
leads
the
tldr
of
this.
Is
you
know
if
you
have
a
quadrant
with
level
of
effort
and
impact
on
the
axes?
B
And
you
know
you
have
your
high
effort,
high
impact,
low
effort,
low
impact
that
low
effort
low
impact
is
you
know
we
call
that
snacking
because
that's
like
some
easy,
you
know
tax
that
you
can
do
that
need
to
get
done,
but
like
someone
else
could
do
them
usually,
but
the
high
effort,
high
impact,
is
some
of
the
things
that
only
some
of
the
the
folks
that
are
you
know
higher
up
the
ladder
can
really
drive
and
put
some
thought
into
like
road
mapping
there,
and
so
you
know,
if
we
all
snack
too
much
we're
gonna
starve.
B
A
Things
like
eddie
with
that
one,
I'm
pretty
sure
that
we
will
return
the
topic
once
laurie
is
present
on
the
call.
Maybe
in
two
weeks
let's
say
the
second
topic,
a
particular
review
of
our
move-
bash
completion
to
go.
B
Yeah,
so
this
has
been
open
for
a
hot
minute
mark
has
worked
on
this.
I
don't
think
mark's
able
to
join
us
today,
but
we
basically
if
we
can
get
some
folks
to
dive
in
give
this
some
review
and
test
it.
It'll
help
out
a
ton
mark
put
a
ton
of
work
into
this
and
it's
kind
of
been
slipping
for
every
release,
so
just
putting
out
a
big
call
that
we
definitely
need
to
go
through
this.
So
please
help.
B
A
Okay,
cool
thanks
doug
and
next
one
is
again
eddie
renaming
default
branches.
B
I
just
wanted
to
call
this
out
the
because
cube
cuddle
is
in
a
staging
repo,
we're
not
able
to
jump
on
the
migration
path
early,
so
it
probably
will
happen
towards
more
the
end
of
the
year,
but
mr
bobby
tables
has
been
pushing
and
talking
about
this,
so
we
are.
We
are
looking
into
figuring
out
how
to
rename
the
cube
control
repos
branches,
but
we
have
to
wait
for
staging
to
be
sorted
out
so.
A
Yeah,
I'm
guessing
that
we
know
we
will
most
likely
be
up
with
the
rename
along
the
main,
kubernetes
repo
which,
if
I
remember
correctly,
is
slated
for
december
this
year
somewhere
towards
the
end
of
this
year.
That's
the
target
go
for
for
the
main
repo,
and
I
assume
that
that
might
be
the
case
for
this
engine
because
as
well,
okay,
what.
A
Other
other
repos
are
free
to
pick
up
the
rename
as
soon
as
they're
ready.
If
I
remember
correctly,
the
process
is
reach
out
to
bobby
and
he
will
be
able
to
to
help
with
that
one.
I'm
pretty
sure
that
the
link
that
eddie
put
together
has
all
the
necessary
links
and
contact
details
how
to
proceed
with
that.
A
So,
if
you're
not
directly
connected
with
the
main
repo,
either
for
staging
or
being
part
of
the
main
repo
you're
free
to
move,
I
think
we've
changed
already
some
of
the
repos,
because
my
scripts
broke
a
couple
weeks
back,
so
it's
definitely
possible
already.
H
Phil
all
right,
I
was
just
gonna-
I've
already
communicated
this
to
like
a
number
of
folks,
but
I
just
want
to
communicate
a
little
more
widely
that
I've
been
having
katrina
do
a
lot
of
stuff.
For
me,
she's
been
stepping
into
more
of
a
leadership
position
and
I've
been
redirecting
people
to
her.
So
if
there's
something
you
need
help
with
just
reach
out
to
crew
katrina
directly
and
then
like,
if
it
actually
needs
my
attention,
she'll
reach
out
to
me.
I
Yep
happy
to
help,
please,
please
feel
free
to
reach
out,
on
slack
or
in
github
about
anything.
A
A
Okay,
I
see
nick
passing
on
this
one
jeff.
You
want
to
talk
about
where
we
are
with
customize
and
the
best
access
to
upgrade
the
customize
and
keep
cuddle.
J
There's
no
news,
I
mean
not,
since
the
pr
was
merged
just
looking
forward
to
seeing
it
get
to
get
distributed.
Seeing
people
use
it.
I
don't
expect
any
surprises.
But
of
course
that's
when
you
get
surprised
when
you
don't
expect
surprises,
but
people
have
been
using
the
code
for
a
long
time,
just
not
in
the
context
of
cool
cuddle.
So
it's
not
like
it
hasn't
been
tried.
J
I
see
something
on
the
agenda
for
next
meeting
about
making
plugins
non-alpha,
which
is
a
good
thing.
That's
fine!
We
can
do
that.
Okay,
great!
That's
all!
I
have.
B
Jeff,
do
we
need
to
do
any
sort
of
documentation,
work
for
people
upgrading
for
cube
control.
J
Would
you
like
to
what
do
you
have
in
mind
for
that
there's
documentation
on
the
there's
lots
of
documentation
on
the
coop
cut
on
a
customized
site
and
there's
online
help
that
is
integrated
into
that
was
part
of
the
pr.
So
there's
the
commands.
The
build
command
in
couple
has
online
hub.
So
certainly
I'd
love
to
see
more
documentation.
But
do
you
have
anything
specific
in
mind.
B
J
J
We
that
has
to
do
with
the
gogetter
the
hashicorp
code
and
the
way
it
worked
is
actually
by
jumping
two
versions,
you're
back
to
where
we
were
in
the
initial
version
of
the
cube
cuddle.
So
there
was
no
in
the
original
initial
version
of
coop
cuddle.
Customized
did
not
use
the
hash
go-getter
and
then
customizing
the
wild
standalone
did
and
to
get
it
back
into
coop
cuddle.
J
Underscores
yeah:
that's
what
you're
right
there
is
that
what
that
does
that
doesn't
break
coop
cuddle
that
breaks
customize
the
people
that
were
using
underscores
instead
of
hyphens.
So
what
we're
going
to
do
and
customize
is
add
double
up
in
the
flags.
I
guess
to
include
underscores
so
kubecuttle
uses
hyphens.
I
I
I
J
It's
not
applicable,
it's
not
applicable
here.
No
it
it
does
what
you
just
said
it
does.
It
does
upgrade
fields
so,
for
example,
we're
introducing
a
new
labels
field
in
standalone
customize,
but
that's
not
a
factor
for
coupe
cuddle.
As
it
stands
right
now,
it
will
become
a
factor
down
the
road
if
we,
if
we
deprecate
so
fix,
is
meant
to
deal
with
hard
deprecations
right
now
we
are
not
deprecating
any
fields,
we're
not
removing
fields.
We
had
one
field
removed
that
I
added
back
in
what
fix
does
is
it?
J
It
helps
people
migrate
from
old
fields
to
newer
fields,
newer
fields
that
have
more
power
or
easier
expression
of
features
that
you
want
to
use
and
fix.
Just
does
that
migration?
For
you,
it
reads
the
customization
file
name
modifies.
It
writes
it
back
out
again,
yeah,
so
that
that's
not
necessary
or
that
doesn't
apply
to
coog
cuddle
right.
D
Had
now
question
a
user
experience,
question
about
managed
fields
right
now
on
most
consoles,
you
know
they
have
kind
of
a
yaml
tab
where
they
present
the
raw
dump
of
the
resource
description.
Should
we?
What
should
we
do
with
a
managed
field
in
that
kind
of
context?
Should
we
never
show
them
always
show
them
have
a
flag
to
show
them
put
them
in
a
separate
tab
like
what's
the
policy
we
have
around
managed
fields
from
ux
perspective,.
A
There
is
no
policy,
there
is
a
big
discussion
around
the
topic
of
managed
fields.
A
Yeah
in
the
interim,
what
we
did
is
in
the
camera,
whether
that
was
previous
or
one
of
the
past
releases.
We
hit
the
the
manage
fields
in
the
edit
command,
and
currently
we
do
that
by
default,
for
every
single
keep
command,
keep
cuddle
command
and
there's
a
flag
that
allows
you
to
show
them,
but
you
need
to
explicitly
sense
that,
yes,
I
want
to
see
them.
A
Sorry,
it
will
be
retrieved
from
the
server.
So
if
you
go,
if
you
look
in
with
v9
you'll
see
them
being
get
retrieved
from
the
server,
but
they
are
not
printed.
A
At
all,
I'll
probably
do
something
similar.
What
cupcattle
does
until
we
have
a
an
official
statement
from
the
server
side,
I'll
probably
follow
what
keep
cuddle
did
by
hiding
them
by
default,
but
still
being
able
to
show
them
either
their
separate
tab
or
through.
I
don't
know,
show
me
the
manage
fields
and
you
will
roll
them
up
on
top
for
where
they
would
normally
be.
A
I
I'll
probably
even
make
you
all
know
that
part
of
the
ux
much
better
than
I
do,
but
I
would
try
to
mimic
the
behavior
of
of
what
cuddle
does.
It
seems
reasonable
backboarding
for
tube
cattle.
I
don't
expect
backboarding,
we
rarely
backboard
features
right,
I'll
leave
the
decision
for
koi
up
to
you
and
the
rest
of
the
group.
A
Okay,
do
we
have
anyone
from
crew?
I
see
chris
kim
on
the
call.
Do
we
have
any
update
on
the
crew
site.
G
I
I
had
just
one
quick
thing:
if
we
have
a
a
a
short
minute,
which
is
so
I
so
we
implemented
kep
859,
which
is
we're
adding
extra
coupe
cuddle
command
headers
in
121
as
alpha,
and
it
turns
out
that
the
api
server
so
I
implemented
a
validating
web
hook,
and
it
turns
out
that
the
api
server
it
appears
is
not
forwarding
these
headers
to
admission
web
hooks.
G
So
it
may
be
that,
as
a
team
we
may
want
to
ask.
I
want
to
get
your
guys's
opinion.
Do
we
want
to
ask
the
sig
api
machinery
if
these
extra
x
headers
should
be
forwarded
to
admission
web
hooks?
Otherwise
we
won't
be
able
to
make
any
decisions
in
the
validating
or
mutating
web
hooks
based
on
these
extra
headers,
and
the
use
case
was
if
there
was
an
especially
dangerous
command.
Maybe
you'd
want
to
put
a
validating
web
hook
to
block
that
command,
for
instance,
coupe
cuddle
copy.
G
If
you,
you
know,
the
sres
in
a
particular
company
think
coop
cuddle
copy
is
too
dangerous,
then
maybe
we
would
put
it
in
a
validating
webhook
to
block
it,
but
unfortunately,
as
it
stands
now,
it
appears
we
can't
make
any
decisions
in
admission
web
hooks,
because
these
extra
headers,
indeed
any
headers,
are
not
being
forwarded
to
any
of
the
admission
web
hooks.
Does
anybody
have
any
opinions
on
this?
Whether
or
not
we
should
interface
with
sig
api
machinery
on
this.
A
I
I'll
probably
try
to
dig
up
the
initial
documents
describing
the
web
hooks
and
what
they
said
about
the
headers,
whether
they
have
any
assumptions
about
yes,
no,
they
want
to
for
them.
If
there
is
no
clear
guidelines
that
yes
or
no
we
want
to
forward
them.
I
would
be
open
to
opening
such
a
discussion.
Definitely
oh
yeah.
I
would
I
would
probably
frame
this
as
not
just
this
particular
one,
but
rather
a
broad
approach
of
exposing
the
headers
to
to
the
web.
Hooks.
G
Yeah,
I
didn't
I
I
I
went
through
the
documentation,
didn't
find
anything
about
headers
in
web
hooks.
Of
course
it
talks
about
the
admission
review
and
the
emission
response,
and
you
know
the
format
of
that
v1
beta
1
versus
v1.
I
didn't
see
anything
on
headers,
but
if
anybody
finds
anything
or
knows
about
anything,
you
know.
Please
point
me
in
that
direction.
A
Okay,
cool
any
other
last
minute:
topics.
A
Okay,
cool,
I
think,
with
that
happy
thought
we
can
close
the
call
today.
Thank
you
very
much
all
see
you
next
week
on
our
blogs,
blogs,
prep
meeting
and
we
have
another
sexy
call
in
two
weeks.
Thank
you
very
much
all
have
a
good
one.