►
From YouTube: 20230308 SIG Arch Prod Readiness
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
All
right,
hello,
everybody-
this
is
kubernetes
sync
architectures
production,
Readiness,
sub-project
Meeting
for
March,
8th,
2023.
A
We
talked
about
a
few
of
these
things
in
our
last
meeting
before
we
decided
to
cancel
it
because
we
didn't
have
enough
attendees,
but
we
did
talked
just
briefly
and
decided
we'd
better
talk.
Well,
there's
a
few
more
people.
So
let
me
share
this
document.
A
All
right,
so
the
first
thing
is
retrospective.
Now
it's
been
a
few
weeks
since
the
enhancements
freeze,
so
I,
don't
know
if
everybody
remembers
but
curious
to
hear
in
particular
from
the
Shadows
how
they
think
when
things
went-
and
you
know
from
everybody
of
course,.
A
Well,
I
guess:
to
get
the
ball
rolling,
so
there's
nobody's
saying
anything:
I
felt
like
it
went
pretty
well,
I
felt
the
shadow
process
was
really
helpful
and
it
was.
There
was
helpful
stuff
from
the
Shadows.
A
What
I
did,
though,
think,
is
that
the
Shadows
need
to
be
more
aggressive,
or
maybe
that's
not
the
right
word
assertive
in
pushing
back
on
people
on
telling
people
hey
this,
isn't
Gonna
Fly
and
that's
okay.
That's
that's
a
learning
process
right
of
learning
how
to
get
building
the
confidence
to
say
say
that
to
people,
but
in
some
ways
it's
easier
than
code
reviews,
because
all
we're
saying
is
like
most
of
the
time.
A
The
mitigation
and
the
resolution
is
for
them
to
add
a
few
sentences
in
a
file.
It's
like
it's
not
we
worth
of
code,
they
have
to
write,
I
mean
it
might
imply
that.
But
you
know
anyway,
that's
the
that's.
The
feedback
I
have
overall
I
feel
like
it
went
pretty
well,
it
was
more
efficient
for
me
for
sure
the
Shadows
did
bring
out
some
issues.
Early
on
and
I
felt.
A
Logistically
are
the
mechanics
of
the
process
were
pretty
good,
but
just
the
main
thing
feedback
I
would
have
is
foreshadows
to
be
more
more
assertive.
B
B
It
it's
okay
to
come
in
and
say
this
is
where
the
bar
is,
it
needs
to
be
crisply
described.
You
need
to
have
some
clear
criteria
that
you're
looking
to
have
met,
but
it's
it's
okay
to
say
well,
this
isn't
what
we're
expecting
we're
expecting
something
like
this,
and
it
can
sometimes
ease
the
pain
if
you're
able
to
point
to
here's
an
example
where
somebody
else
has
done
it.
B
Yeah
I
am
interested
in
how
it
felt
on
the
part
of
the
Shadows
and
if
nobody
talks,
I,
mean
I'll,
know,
probably
choose
a
random
name.
D
I
can
go
first
and
maybe
I'll
stick
to
the
questions.
How
did
it
go?
I
feel
it
went
pretty
well,
considering
we
have
almost
100
I
think
it's
96.
It
hasn't
been
raised
for
the
current
cycle,
which
is
like
the
most
everything
so
yeah
I
think
it's
went
pretty
well
and
I'm.
D
So
glad
like
we
got
the
process
on
kind
of
communicated
well
because
at
the
very
beginning,
it's
kind
of
not
quite
clear
like
if
the
approval
should
wait
until
the
all
the
Shadows
data
pass
or
should
like
they
wait
until
Shadows
see
like
it
looked
good
to
me
or
like
or
if
we
should
like
do
parallel
and
I
think
it
turned
out
like
you're
running
pretty
well
like,
like
we
tried
our
best
to
like
I
guess
have
thank
you
so
much
for
having
like
Shadows
dated
the
first
pass.
D
So
at
least
the
weekends.
But
hopefully
we
could
spot
something
and
help
like
the
process
running
most
mostly
and
another
feedback.
D
I
yeah
I
personally
feel
like
I,
also
visited
in
the
like
Channel
I,
feel
like
the
a
little
bit
challenged
part
while
reviewing
is
to
let
all
the
kind
of
cab
owners
be
aware
of
the
the
the
cap
template
change,
I
think
especially
it
changed
like
in
a
way
it's
quiet
and
especially
when,
like
they
already
got
a
caffeine
and
it's
so
easy
for
them
just
go
ahead
and
update
all
the
fields
and
everything
instead
of
like
pulling
the
new
template.
D
Yeah,
it's
I,
yeah
I
assume
it
will
be
like
I,
expected
a
kind
of
extra
action
for
some
of
the
cap
owners,
especially
for
people
who
trying
to
promote
their
features
into
next
level.
And
there
are
certain
questions
like
cares
more
than
like,
comparing
to
like
they
have
like
a
pretty
new
Alpha
feature,
so
yeah
I
guess
the
communication
initially
didn't
goes
quite
well.
D
I,
like
with
the
release
team
head
on
I,
think
it's
partially
also
like
a
release
teams,
responsibility
to
like
for
the
communication
but
also
like
it
will,
of
course,
cause
extra
time
yeah,
but
other
than
that.
I
think
I
learned
a
lot
and
thank
you.
So
much
also
just
by
going
through,
like
the
questions
like
everybody
raised
on
the
Caps,
like
yeah
I,
think
I
learned
a
lot
personally.
Thank
you.
E
I'm
going
next
I
I'll
Echo
that
I
didn't
expect
the
template
change
and
that
caught
me
off
on
a
few
of
them.
Look
I
had
to
go
back
and
check
again,
I
also
don't
think
I
allocated
or
I
estimated
enough
time
for
how
long
it
was
going
to
take
me
to
get
through
some
of
them
just
felt
like
that.
I
didn't
block
off
enough
time
to
yeah.
A
F
E
Been
approved
since
the
last
time,
so
there
was
a
lot
more
kind
of
like
bookkeeping
that
I
kind
of
expected
so
like.
If
I
do
this
again
next
time,
I
will
definitely
node
like
backload
a
little
bit
more
time
into
things.
I
think
that
suggestion
that
would
be
more
assertive
is
good.
Like
I
felt
like
in
some
of
the
cases,
I
didn't
feel
like
I
I,
don't
know,
I
was
nervous
about
me
sure
more
assertive
or
pushing
people
too
far,
but
that
totally
makes
sense.
A
Any
other
Shadows,
so
so
you
know
a
few
people
joined
after
I.
Think,
just
to
summarize
what
David
and
I
both
said
and
I
think
boy
checking
past
discussions
with
him
said
the
same
thing
that
we
we
would
we're
pretty
happy
with
how
the
shadowing
process
went.
But
what
we
would
like
to
see
more
next
cycle
is
more
assertiveness
from
the
Shadows
to
to
to
set
the
bar
as
David
put
it
and
and
be
okay
with
with
telling
people.
A
That's
that's
not
that's
not
up
to
the
bar,
but
maybe
be
able
to
point
them
and
say
Here's,
here's
one
that
is
up
to
the
bar
here's
or
here's
what
you
need
to
do
to
make
it
up
like
it's
collaborative
process.
We
want
them
to
get
through,
but
we
just
want
it
to
be
at
our
bar.
F
But
yeah
I
will
just
add
one
thing
that
the
bar
that
I'm
using
personally
is
also
like
going
higher
over
Timeless,
so
I
I
think
there
are
like
I,
definitely
see
things
that
I
passed
through
like
in
the
past
Cycles
like
year
ago,
or
something
and
I
wouldn't
pass
now,
and
so
that
also
comes
a
little
bit
with
like
the
more
you
do,
the
more
the
the
more
patterns
you
see,
the
more
you
learn,
the
the
more
you
also
start
to
require
over
time,
at
least
in
my
in
my
case,
but
I
think,
like
I,
think
David.
F
You
also
mentioned
something
similar
last
two
weeks
ago
or
something
like
that
right,
yeah.
B
You
get
more
comfortable
doing
it,
you,
you
also
get
a
sense
for
how
much
is
how
much
prior
Arthur
actually
is,
and
just
over
time
it
becomes
easier
to
say
yeah.
This
is
what
we
do.
Here's
an
example:
we've
done
it
before,
and
I
have
found
that
giving
people
examples
helps
a
lot,
and
so
we
have
been
able
to
to.
C
B
And
I'm
looking
forward
to
I'm
hoping
you
guys
stick
around
and
I
guess.
That's
also
related
to
the
next
item.
When
we
get
there.
F
G
Yeah
I
feel
like
I,
wasn't
quite
as
pushy
as
I
might
have
been,
because
I
felt,
like
I,
was
kind
of
representing
the
process
and
didn't
want
to
be
even
more
pushy
in
prr,
but
it
sounds
like
that.
There's
plenty
of
appetite
for
that
also
for
me,
specifically
I,
think
I
would
have
liked
to
spend
a
lot
more
time
on
it
and
I
aim
to
do
that
next
time.
G
We
kind
of
have
also
had
the
I've
been
returned
to
the
budget
crunch
for
Kate's,
Emperor
and
we'd
have
to
prioritize
things
early
this
year,
but
I
expect
later
this
year,
though
it
will
be
less
of
an
issue
and
yeah
I,
don't
feel
good
about
how
much
time
I
spent
but
I
hope
that
won't
be
a
problem.
Next
time.
A
Cool
yeah
yeah
I
mean
pushy,
is
one
word
like
I
think
we
want
to
be
collaborative
too
right
like
like,
but
holding
the
bar
is
good
and
it's
okay
to
be
be
pushy.
In
that
sense,.
A
H
Yeah
sure
so
yeah
it's
been
a
while
now,
let's
see
my
yeah,
it
has
been
a
while.
But
for
me
my
experience
was
a
little
bit.
Different
I
did
allocate
a
bunch
of
time
for
it,
but
it
ended
up
not
taking
as
much
time
because,
like
the
majority
of
my
caps
were
like
not
actually
doing
anything,
so
I
ended
up
doing
a
smaller
fraction
of
caps
than
I
had
anticipated,
but
the
ones
that
I
was
able
to
review
I
yeah
I
enjoyed
the
process
it
was.
A
Okay,
I
mean
I.
Think
one
piece
of
curious
about,
but
not
hearing
from
any
of
the
Shadows,
is
like
what
we
can
do
better
in
the
shadow
program
like.
Is
there
a
better
way
to
allocate
them,
or
is
there
a
better
way?
You
know
what
didn't
work
for
you
I,
guess
that
we
can
do
better.
H
It
would
be
nice
if
there
was
like
a
guide
to
prr
like
not
just
for
Shadows,
but
for
people
who
are
writing
caps
in
general.
I
think
it
would
be
helpful
for
both
right,
because
the
questions
that
we're
asking
are
questions
that
people
should
be
asking
when
they're
writing
caps
right
right.
A
H
H
If
a
guide
had
examples
like
hey
like
in
this
situation,
this
kind
of
feature
involves
you
know
like
turning
on
this
feature
and
then
what
happens
when
you
disable
this
feature?
Does
it
break
something
you
know
like
questions
like
that,
and
examples
of
you
know
features
that
sort
of
exemplify
those
patterns
probably
would
be
helpful.
H
I
mean
because
all
we
really
have
are
the
questions
and
and
yeah
the
questions
are
great,
but
you
know
I
think
the
value
of
the
reverse
shadowing
was.
We
saw
specific
instances
of
these
things,
but
they're
not
like
completely
articulated
right
like
we
just
if
we
got
lucky.
We
saw
an
example
of
like
this
one
way
this
thing
could
break
or
this
other
way
something
could
break
right.
A
A
Why
the
question
exists
and
there's
like
a
good
answer
to
the
question
versus
a
not
so
great
answer
to
the
question
you
know
like
or
an
example
of
where
I
guess,
that's
kind
of
the,
why
it
exists
an
example
of
where
a
problem
occurred,
because
this
wasn't
done
or
whatever
yeah
I
think
that
would
be.
That
would
be
great.
That
would
be
useful.
D
One
thing
I
want
to
brought
up
is
like
the
process,
wise
yeah
I,
understand
the
idea
on
process
or
like
a
verse.
The
process.
What
we're
trying
to
do
is
maybe
having
Shadows
working
with
the
cap
owners
first
to
get
to
the
pr
section
at
least
the
obviously
ready
to
be
merged
the
kind
of
phrase
until
like
the
approval,
jump
scene
and
do
the
final
pass.
D
But
one
thing
I
noticed
during
the
last
couple
of
Cycles
is
like
at
least
one
third
or
even
more,
of
the
Caps
coming
up
like
in
last
week
or
last
three
days.
It
makes
it
hard
to
like
yeah.
A
D
A
A
But
open
the
ideas
for
sure
all
right
well
feel
free
to
add
some
stuff
to
the
notes.
Documentary
but
I.
Think
David
is.
A
Else
we
want
to
talk
about
in
the
Retro
like
I
feel
pretty
good
about
it.
I
think
the
the
next
thing
is,
unless
there's
something
else,
somebody
wants
to
read
we'll
move
to
the
next
thing.
C
A
C
B
So
I
wanted
to
to
talk
about
the
criteria
that
we
think
makes
sense
for
trying
to
go
from
Shadow
to
approver,
and
this
has
actually
been
sort
of
simmering
for
for
a
long
time
and
I
wanted
to
throw
out
a
few
ideas
and
see
both
how
how
John
and
vojtek
think
about
it
and
also
how
the
Shadows
think
about
it.
B
I
think
that
a
lot
of
what
I
am
interested
in
is
is
breadth
of
experience
for
different
kinds
of
of
caps
and
the
application
of
that
experience
in
I,
don't
want
to
say
contentious
in
cases
where
it
wasn't
all
just
ready
and
there
might
be
some
cordial
disagreement
and
so
I've
thought
about
a
few
categories
and
I
want
to
propose
them
here.
So
one
category
is
reviewing
a
certain
number
of
enhancements
at
an
alpha
to
Beta
transition
that
require
coordination
between
multiple
components
and
the
alpha
to
Beta.
B
Transition
is
important
to
me,
because
that's
the
one
where
the
answers
are
required
and
there
is
an
actual,
it's
actually
possible
for
them
to
have
tried
and
enable
and
disable
flow.
B
So
the
multiple
components
comes
to
mind.
Another
one
that
came
to
mind
is
cases
of
enhancements
that
require
a
version
SKU
consideration,
both
during
upgrades
and
with
the
long-running
case
where
cubeless
runs
skewed,
and
so
that
would
be.
You
know
you
can
have
cubelets
that
run
up
to
n
minus
two,
and
that
has
a
consideration
for
when
and
how
you
detect
what
sorts
of
failures
you
have
and
debugging
and
and
the
like
and
then
the
other
category.
The
last
category
is
enhancements
that
are
outside
of
your
primary
domain.
B
Right,
I,
think
I
know
each
of
you
have
things
that
you're
really
really
good
at
and
then
a
lot
of
things
that
you're
very
good
at
and
then
some
things
you're
not
as
knowledgeable
about
and
and
it
can
be
a
stretch
but
I
think
we
part
of
this
is
being
able
to
effectively
review
things,
components
that
you
are
not
intimately
familiar
with,
and
so
in
my
head.
Those
are
those
are
the
criteria.
I
don't
have
numbers
to
put
against
them,
but
I.
A
Think
another
category
David
is
is
examples
where
thinking
at
scale
is
important,
so,
like
both
scale
in
hey,
here's
a
large
cluster.
A
Where-
and
this
is
maybe
I
don't
know-
this
is
maybe
inherent-
this
is
maybe
like
just
part
of
the
the
process,
but
it's
where
you're,
where
your
feature
makes
you
think
about
how
would
I
deal
with
this
at
a
scale
of
thousands
of
clusters
from
an
administrative
point
of
view,
so
that
that's
one
that
I
that's
one
of
the
areas
at
least
I
find,
and
maybe
it's
just
my
biases
I
find
people.
A
Often
you
know
answer
questions
in
a
way
that
makes
sense
if
you're
only
have
one
cluster,
but
if
you've
got
like
10
000
clusters,
like
I,
can't
use
this
method
of
now.
You
know
when
you
need
a
metric
right
after
supposed
to
like
run
a
cube,
control
command
like
right.
A
A
One
is
well
it's
two,
it's
two
different
types
of
scale.
One
is
scale
in
terms
of
the
scale
of
an
individual
cluster.
So
if
I
enable
this
feature
and
I've
got
a
cluster
with
ten
thousand
nodes-
and
you
know
thousands,
you
know
twenty
thousand
endpoints
in
a
service.
What's
gonna
happen.
A
E
A
And
I
think
definitely
seeing
I
think
some
sort
of
they've
done
transitions
from
you
know.
Two
alpha
or
new
new
feature
Alpha
to
Beta
beta
to
stable
I
think
that
we
should
probably
look
at
making
sure
people
have
covered
all
of
those
areas
too.
A
I
mean
I
agree
that
Alpha
Beta
is
a
big
one,
but
like
Alpha
for
me,
Alpha
the
alpha
ones,
are
the
from
a
prr
perspective.
There's
not
that
much
right!
It's
like.
Can
you
turn
this
thing
off?
Can
you
turn
it
back
on
and
not
Break
Stuff,
but
they're?
A
Actually,
the
most
work
a
lot
of
the
times,
because
it's
the
first
time
you're
seeing
them
and
you
spend
the
most
time
on
them,
and
so
it
like
from
looking
at
how
Shadows
are
thinking
about
things
I
think
it's
useful
to
have
to
like
digest.
Maybe
this
is
just
true
of
a
beta
one
anyway,
but
you.
F
A
A
B
B
I
would
like
to
have
I
would
like
to
have
before
we
do
the
next
cut
cycle.
I
think
what
I'd
like
to
do
is
get
these
I
think
we
have
a
page
on
the
Community
page,
where
we
list
out
what
we
think.
Good
examples
of
criteria
are
and
then
I
think
it
could
be
worthwhile
I'm
not
going
to
say
that
a
shadow
has
to
do
this,
but
I
think
that
in
the
next
cycle
would
be
worthwhile
keeping
track
of,
because
the
big
list
will
be
big.
B
You'll
have
caps
you'll
be
doing
fresh
you'll
have
caps
that
you'll
have
done
in
the
past
and
they're
likely
to
come
back
up,
because
you
know
you
see
the
same
ones.
Transition
between
these.
C
A
They
missed
that
they
missed
the
release
too.
B
Trying
to
keep
track
of
which
categories
the
Caps
would
fit
into
to
make
it
easier.
If,
after
that
cycle,
you
guys
both
feel
that
you
are
ready
and
we
agree.
I
would
be
I.
Think
it'd
be
great
to
be
able
to
see
that
sort
of
listed
out
enumerated
and
we
say
awesome,
I,
don't
know
if
that's
encouraging
or
discouraging
for
the
Shadows
here.
So
if
it's
discouraging
I'll
shut
my
shut,
my
mouth.
B
H
A
So
yeah
David
I
mean
I
I,
agree.
I,
think
that
that's
probably
the
best
way
to
do
it
because
I've
been
having
the
same
thing
as
like.
How
do
we
know
when
somebody
is
ready
for
to
be
an
approver
like
I,
think
having
some
criteria,
kind
of
making
sure
people
have
a
few
examples
of
each
of
these?
That
they've
demonstrated
and
then
I
think
there'll
be
a.
You
know,
I
think
that
there
is
an
assertiveness
Quality,
that's
you
know,
that's
very
subjective,
but
yeah
we'll
see
how
that
goes.
A
I
think
for
for
you
know
it's
a
goal
of
me
and
David
and
boycek
to
get
some
new
approvers
on
back
in
the
next
couple
of
Cycles,
and
so
we
we
want
you
to
to
grow
and
to
succeed
into
the
role
in
order
to
demonstrate
even
whether
you
have
to
grow
it
or
not,
for
it
to
demonstrate
and
ask
that
you're
ready
for
it.
But
yeah
that
sounds
good
to
me.
David,
that's,
a
pretty
good
start.
F
I
need
to
run
in
like
few
seconds.
I
have
another
meeting
today,
actually
and
in
a
minute,
but
one
more
thing
that
I
would
also
consider
like
extending
the
criteria,
or
maybe
we
decide
not,
but
like
one
more
thing
is
isn't
the
is:
is
the
cap
that
like
didn't,
didn't
create,
or
the
cup
the
Capital
Tour
didn't
come
up
with,
like
any
metrics,
I
I?
Think
that's!
F
There
are
very
few
caps
where,
where
like
it,
actually
pops
up
as
as
an
important
topic
for
metrics,
so
it
will
also
be
harder
to
to
to
actually
be
the
approver
or
be
the
shadow
for
those,
because
there
are
very
few
of
them
for
for
metrics
I
think
it's.
F
It
happens
very
often
I'm,
seeing
at
least
like
a
few
per
cycle
that
I
that
are
assigned
to
me
so,
and
it
also
text
a
little
bit
like
of
the
what
we
were
chatting
at
the
beginning,
that,
like
you,
need
to
be
assertive
and
and
kind
of
persuade
that
the
cap
out
or
to
actually
do
something
in
the
Matrix
area.
A
A
C
B
Drop
in
like
two
minutes
so
might
just
be
quick
here,
though,
very.
C
Up
yeah,
it
will
be
quick.
There
is
also
another
mission
called
enhancement
sub
project,
so
I
posted
a
link
to
the
agenda
from
past
meeting,
and
there
were
discussion,
template
updates
in
the
middle
of
release.
I
think
Kristen
took
an
action
item
to
actually
write
a
document
when
template
can
be
updated
and
like
when
it
cannot
be
updated
and
I
also
wanted
to
remind
about
these
two
PRS,
like
one
PR
in
one
issue.
John.
If
you
still
have
super
powers,
please
merge
spr,
it's
very
straightforward,
yeah.
C
Second,
one
is
issues
that
we
discussed
to
I
think
increasing
linked
another
issue.
Somebody
asked
about
the
same
question.
So
if
you
can
update
a
template
right
now,
it
will
be
before
128.
It
will
be
useful.
Thank
you.
C
B
Yep
that
one
was,
if
I
remember
the
right
issue.
It
was
mine,
I
needed
to
clarify
the
difference
between
integration
and
any
tests,
and
when
you
use
one
or
the
other
and.
E
B
A
A
Okay,
so
Kristen's
gonna,
they're
gonna
set
a
policy
to
not
touch
the
template.
Certain
time
period
is.
A
All
right
anything
else
or
shall
we
call
it
a
wrap
today.