►
From YouTube: 20211201 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
Hello,
everybody
today
is
december
1st
2021,
and
this
is
this:
kubernetes
production
readiness,
some
project
for
sick
architecture,
all
right
thanks
all.
I
think
we
have
a
pretty
small
agenda
today,
david.
B
Great
okay,
so
I
went
through
the
biggest
there
were
some
eliminate
minor
knits
about,
like
you,
have
no
categories
and
some
changes
around
a
couple
other
labels,
but
the
big
change
was
giving
absolute
numbers.
There's
nothing
too
striking
on
these
first
charts
about
how
you
expect
the
the
null
change
didn't
really
mess
with
much
by
the
way
david.
You.
A
B
B
Anyway,
go
on
the
absolute
counts
on
the
version
rollback,
it
turns
out,
it's
actually
really
hard
to
get
the
google
report
viewer
to
they
won't.
They
won't
show
up
here,
so
I
had
to
add
a
separate
little
chart
which
doesn't
look
as
pretty
as
the
stuff
that
john
put
together,
but
it
shows
you
the
total
numbers
that
we
that
these
percentages
represent,
which
matters
down
below
right,
where
you're
starting
to
build
percentages
of
what
were
the
reasons
down
here?
B
That's
minorly,
embarrassing.
You
cannot
see
it
all
at
once.
So
if
you
scroll
up
and
down,
you
can
see
that
overall,
the
sample
size
is
decently.
Large
right.
You
have
11
production
clusters
where
these
are
their
reasons,
but
in
the
previous
survey
the
numbers
were
quite
small
in
comparison
right.
So
what.
E
Was
the
overall
response
size
in
comparison
like
how
many
did
we
get
for
responses.
B
E
B
D
A
B
I'm
trying
to
remember
it
was
this
page
or
this
one.
It
was
this
one,
the
percentage
where
the
cluster
failed
on
a
patch
upgrade
is
significantly
higher
and-
and
we
have
a
fair
number
of
responses
overall.
C
A
Okay,
so
that's
something
we
should
look
at
because
it's
not
going
to
matter
as
much
here,
because
we
can't
do
a
year
to
year
comparison,
because
we
didn't
ask
that
question
in
2020,
but
I
mean
it's
true.
I
think
it
would
be
useful
to
see
like
is
it
all.
You
know
predominantly
one
particular
version
that
was
bad
or
is
it
you
know.
B
We
know
there
was
one,
but
it
wasn't
it
wasn't
that
it
didn't
only
cause
cluster
death,
but
the
other
thing
to
notice
is
that
the
absolute
number
of
people
who
roll
back
the
patches
they
didn't
didn't
move
up
right.
So.
B
Yeah
and
then
the
last
one
that
we
had
talked
about
was
this
one
I
believe,
and
that
one
is
still
not
not
awesome.
That's
a
lot
of
people
that
enable
alpha
in
production.
B
So
we
that's
the
absolute
number
and
it's
big.
C
B
So
otherwise
it
was
just
cleanup
and
I
don't
think
we
need
to
go
re-go
over
any
of
it.
It
doesn't
change
the
way
these
charts
look
and
clean
up
the
the
no
feels
they
look
basically
identical
to
what
they
did
last
time
went
over
it.
So
yeah
right.
B
I
do,
I
think
I
want
to
cover
I'm
going
to
try
to
keep
it
under
10
minutes
yeah.
Do
you
think
I
need
to
get
down
to
five.
B
B
The
beta
versus
ga
and
alpha
enablement-
those
are
the
areas
I
think
I
want
to
focus
on,
and
the
I
think
I
might
want
to
broach.
The
topic
there
in
that
meeting
of
you
know
is:
is
beta
usage
in
production.
What
we
intend
or
do
we
want
to
take
steps
to
make
that
slightly
harder
to
do,
not
make
it
impossible,
but
make
it
harder
to
do
so.
It's
a
clear,
opt-in,
thanks
right.
B
If
there
are
other
topics
you
really
want
to
cover,
I'm
I'm
open
to
it,
though
so
yeah.
Basically,
I
was
gonna
brush
over
these
charts
related
to
rollback
fairly
quickly.
A
That's
fine,
I
mean,
if
you
have
time
to
look
at
the
version
numbers
that
can't
roll
back.
I
think
that
would
be
helpful.
I
don't
know
that
everybody
answered
those
questions,
but.
B
Yeah,
I
I
will
see
but
chances
of
being
able
to
do
that
in
the
next
24
hours
are
not
awesome.
Oh,
I
guess
there
was
one
other
thing
on
the
versions.
I
was
going
to
go
ahead
and
mention
about
how
the
versions
we
have
fewer
skewing
older
than
n
minus
four,
which
showed
up
in
this
chart,
and
it
is,
I
think,
a
good
sign.
People
are
upgrading
more
often.
A
Okay,
yeah,
that's
good.
I
think
I
I
liked
atlanta's
suggestion
from
last
time,
not
obviously
this
isn't
urgent,
but
I
think
that,
as
we
go
to
do
these
in
the
future,
we
should
code,
those
as
n
and
minus
1
and
minus
2
and
then
it'll
be
easy
to
compare
like
and
do
that
cumulative
right
like
how
many
people
are
on
and
how
many
are
people
are
on
n,
minus
1
and
minus
2,
etc,
or
maybe
from
the
other
direction.
B
E
A
B
B
I
think
the
only
thing
I
might
want
to
wait
for
for
the
next
survey
is
making
sure
that
the
stable
channels
are
for
at
least
some
of
the
major
cloud
providers
are
also
at
122,
because
122
is
the
first
level
where
there
was
the
no.
No
really
we
mean
it
beta's
going
away.
D
Kind
of
runs
that
show
we
already
have
122,
but
it's
it's
currently
in
rockets.
So
it's
it's
okay,
it's
it's!
Basically,
it
has
basically
relatively
low
adoption.
We
are.
We
are
soon
graduating
into
regular.
So
hopefully,
in
like
couple
months
like
I
would,
I
would
expect
that
with
within,
like
the
next
quarter
and
a
half
ish
so
mid
q2,
we
should
have
like
pretty
large
adoption
of
122.
B
Yeah,
I
don't
know
it's
definitely
in
our
fast
channel.
I
don't
know
if
it's
in
our
stable
channel
yet.
B
So
I
would
like
to
see
that
as
criteria
before
sending
out
the
survey,
but
I'm
hoping
it
happens
soon,.
E
B
Stable
there's
a
rapid
and
fast
or
rapid
and
regular
for
gke
I'd
like
at
least
a
couple
of
us
to
have
it
in
stable.
For
I
don't
know
three
weeks
four
weeks
before.
D
E
A
E
We've
the
rest
of
them.
We
all
sent
out
in
q1
yeah.
A
D
B
Weeks
after
we
go
to
stable-
and
you
know
there's
two
of
us-
then
I
think
that'll
be
enough
for
for
what
I'm
looking
for.
A
Okay,
then
I
got
you
on
the
agenda
for
tomorrow
david
and
I
don't
have
anything
else.
I
guess
it's
not
going
to
be
that
long
before
we're
starting
our
next.
Our
next
cycle
of
reviews.
E
B
B
And
well,
I
think
I'm
here
in
two
weeks,
but
I
don't
anticipate
anything
significant
happening.
A
E
Yeah
I,
if
we're
in
I
probably
we
should
just
cancel
the
next
two
meetings.
I
would
say
unless
we
have
anything
burning,
I
oh,
I
did
have
a
thing
which
was.
I
did
my
action
item,
so
I
updated
the
documentation
for
futuregate
use.
I
do
wonder
if
it's
not
very
discoverable,
I
just
double
checked
the
kept
template
and
it
doesn't
link
to
that
documentation.
I
think,
because
that
documentation
was
created
after
the
kept
template
was
put
together.
E
A
That
awesome,
thank
you
very
much
cool
all
right
anything
else,
or
should
we
get
back
45
minutes
of
our
lives
I'll?
Take
the
latter
all
right
see
you
guys
later
have
a
great
holidays
and
vacations.
If
I
don't
see
you
guys
well,
I
guess
I'm
gonna
see
some
of
you
tomorrow,
but
all
right
thanks,
take
care,
cheers.