►
From YouTube: Kubernetes SIG CLI 20230517
Description
Kubernetes SIG CLI bi-weekly meeting on May 17th, 2023.
Agenda and Notes: https://docs.google.com/document/d/1r0YElcXt6G5mOWxwZiXgGu_X6he3F--wKwg-9UBc29I/edit#bookmark=kix.1w649044f53o
A
Hit
record
welcome
to
the
May
17th
edition
of
the
6cli
bi-weekly
meeting
I'm
Sean
Sullivan
I'm,
your
host
I'll
get
directly
into
the
announcements,
so
the
128
release
dates
have
been,
have
been
sent
out
and
let's
go
over
a
few
of
the
important
ones
because
some
of
them
are
coming
up
pretty
quickly.
So
the
production
Readiness
freeze
is
in
a
little
bit
over
three
weeks
on
Thursday
June
8th
for
those
who
are
working
with
caps.
A
Keep
that
date
in
mind,
but
also
especially
keep
the
enhancement
freeze
deadline
in
in
mind,
which
is
going
to
happen
in
about
four
weeks,
and
the
code
freeze
is
in
about
nine
weeks
on
Tuesday
July
18th.
If
you're
in
the
Pacific
time
zone,
the
UTC,
it's
Wednesday,
July
19th
at
1am
and
the
release
itself
should
go
out.
A
B
B
B
He
drives
the
decision
to
push
run
those
meetings
and
he
is
driving
most
of
them
until
this
day,
with
some
help
from
other
folks
and
Eddie
is
also
a
a
primary
person
driving
a
lot
more
folks
into
helping
six
CLI.
So
we've
decided
to
promote
Eddie
to
to
attack
with
role.
So
congratulations
to
Natasha
and
Eddie,
and
thank
you
thanks.
A
So
if
you
haven't
been
to
a
60
line
meeting-
or
you
haven't
been
here
for
a
long
time
and
you'd
like
to
to
raise
your
profile
with
the
rest
of
the
6cli,
this
is
your
opportunity
to
introduce
yourself,
and
this
is
completely
voluntary.
So
if
that's
not
what
you
want
to
do,
that's
fine
as
well.
Is
there
anybody
who'd
like
to
introduce
themselves.
D
A
A
A
A
E
About
the
process,
as
it
was
mentioned,
there
is
a
code
freeze
toward
the
version
release
in
three
or
four
weeks
and
in
the
previous
meeting,
I've
presented
an
issue,
an
improvement
for
cube,
CTL
weight
command
and
I
made
a
newbie
mistake.
I
didn't
assign
the
issue
to
myself
and
one
of
the
contributors
volunteered
to
take
the
issue
and,
and
this
person
submitted,
the
change
and
I
was
involved
in
the
change
to
see
how
it
works
and
I
see
that
the
progress
on
this
work
is
very
slow.
E
So,
after
two
weeks
we
are
didn't,
went
through
many
iterations,
so
I'm
interested
for
a
few
reasons
to
take
this
work
forward
and
hopefully
to
have
it
released
before
the
freeze
and
the
next
fashion
release.
So
my
question
is:
is
there
any
process
or
appropriate
procedure
to
take
this
thing
forward?
Given
them
as
a
core
contributor
has
a
limited
capacity
to
work
on
an
issue?
B
Yeah,
you
definitely
touch
on
a
good
point.
The
the
conflicting
or
parallel
PR
is
a
problem
type
that
I've
been
doing
in
the
past,
but
what
I
personally
tell
everyone
is
that
they
should
double
check
with
the
author
if
they
are
interested
in
working
on
it,
and
only
if
the
author
does
not
respond
then
take
over
a
particular
issue.
Pr
or
anything
like
that.
B
You
clearly
pointed
out
that
there
was
a
your
mistake
that
you
didn't
explicitly
called
out
that
you
are
planning
to
work
on
it,
which
is,
is
a
fair
thing
and
it's
probably
the
safest
to
do,
but
definitely
what
Sean
said
with
regards
to.
If
there
will
be
two
competing
PRS,
the
one
that
is
completed
will
be
the
one
that
gets
merged.
B
There
may
be
multiple
options
why
he
he's
not
pushing
this
forward,
maybe
trying
to
sync
with
him
either
through
GitHub
issues,
by
linking
your
work
or
by
trying
to
paying
them
on
slack
and
get
a
closure
on.
That
topic
is
definitely
worth
worthwhile,
at
least
for
for
you
and
then
then,
at
the
end
of
the
day,
you
can
always
ask
them
to
review
your
PR.
B
Maybe
the
other
person
may
still
enjoy
student
in
following
up
and
and
submitting
additional
changes
on
top
of
the
the
work
that
has
been
merged.
So
that's
roughly
the
process
that
I
I
I've
been
personally
preaching
and
and
as
well
of
trying
to
implement
information
for
myself
but
yeah.
It's
it's
unfortunate.
I
I've
seen
multiple
times
that
there
are
sometimes
people
who
will
pick
up
various
issues
without
even
asking
or
without
double
checking,
which
is
a
problem
for
the
whole
project
as
a
as
a
whole
player
we're
just
humans.
E
Thank
you
very
much.
I
honestly
was
like
having
this
in
mind
when
I
asked
that
I
just
wanted
to
confirm
that
there
are
no
like
well-defined
process
that
I
might
not
work
upon.
But
yes,
this
was
my
idea
just
to
communicate
clearly
with
the
author
of
the
original
PR
to
make
sure
if
he
is
able
to
provide
a
the
resolution
towards
the
the
time
and
see
if
it
is
okay
for
him,
if
I
will
take
it
over
and
collaborate
with
him
to
release
an
adventure.
Thank
you.
A
C
Hi
I
had
a
quick
question
sure
so
I
have
an
open
PR
for
a
particular
issue.
I
was
wondering
if
there's
like,
if
this
is
the
right
Forum
to
sort
of
discuss
it
or
maybe
bring
it
up
with
this
group
or
if
I
should
reach
out
to
you
directly
and
sort
of
like
I,
can
both
tell
you
the
sort
of
the
reason
I
created
the
pr
and
sort
of
what
it
attempts
to
do
so
yeah.
A
C
I'm
happy
to
sort
of
quickly
talk
about
the
motivation
so
like,
as
I
mentioned
earlier,
I
work
on
sort
of
a
kubernetes
observability
tools
and
one
of
the
things
that
we're
hoping
to
do
is
give
folks
the
ability
to
see
that
Cube
cuddle
describe
style
output
for
the
objects
they
have
and
I've
noticed
that
describe
doesn't
have
any.
We
use
informers
to
basically
get
access
to
these
objects
on
the
cluster
and
describe
usually
reaches
out
to
the
API.
C
To
get
the
to
generate
the
output,
it
doesn't
have
any
way
to
use
in-memory
objects,
except
for
this
like
default
object.
Describer,
which
has
it
seems
like
it's
been
sort
of
neglected
for
a
while.
So
I
was
just
basically
adding
all
the
other
describers
to
that
object.
I,
don't
know
if
you
want
to
keep
that
around
or
if
you
want
to
change
that
sort
of
style
and
maybe
move
to
a
different
way
of
doing
it.
A
So
I
just
tagged
it
with
okay
to
test,
and
so
we
can.
We
can
test
it
now
and
so
matcha
you
and
I
are
both
mentioned
as
reviewers.
Do
you
have
anything
to
to
add
I,
actually,
don't
have
anything
to
add
right
now
and
I
I.
Don't
have
that
much
background
and
describe,
but
do
you
want
to
mention
anything
now.
C
Done
yeah
no
worries,
yeah
I
looked
through
the
get
history
for
the
default
object,
describer
and
it
seemed
like
it
was
added
many
years
ago
and
there's
some
debate,
some
potential
debate
on
whether
we
should
keep
it
or
not,
and
then
I
kind
of
just
got
left
there
without
updates
being
made
or
without
it
being
removed.
So
I
understand
if
there's
like
a
reasonable
reason
to
not
have
such
a
default
describer
or
something
of
that
sort,
but
I
figured
if
one
exists,
we
should
make
it
feature
complete
or
something
like
that.
So.
C
B
A
D
A
A
This
is
pulling
out
the
the
actual
kind
of
guts
of
the
command
versus
the
Cobra
part,
and
so
the
idea
is
that
a
lot
of
the
functionality
for
these
commands
maybe
could
be
reused,
maybe
even
in
a
controller,
and
so
we
have
to
pull
out
the
command
line,
interface,
specific
code,
the
Cobra
specific
code
and
separate
them
and
there's
some
general
techniques
that
have
already
been
used.
I
think
the
the
weight
command
actually
was
maybe
one
of
the
first
and
shows
that
the
the
clearest
on
how
to
do
that.