►
From YouTube: Kubernetes 1.10 release burndown meeting 20180312
Description
A
Good
morning
everybody
it
is
Monday
March,
12
2018.
This
is
the
penultimate
week
of
the
community's
1.10
release,
and
this
week
we
will
be
commencing
with
the
Monday
Tuesday
Wednesday
Thursday
Friday
meetings
and
probably
Saturday
and
Sunday
meetings
too,
because
that
seems
to
seems
to
happen,
hopefully
not
too
much
of
that.
But
so
we
are
going
to
run
through
the
schedule,
real,
quick
and
talk
about
where
we're
at
and
where
we
need
to
be
so
this
week.
A
This
is
lovingly
known
by
myself
and
others
as
crunch
week,
which
means
that
we
need
to
get
a
lot
done
this
week,
as
I
mentioned,
we're
going
to
be
beginning
Monday
through
Friday
burn
down
meetings.
Wednesday
is
the
end
of
code
freeze
at
6
p.m.
Pacific
time
I'll
be
working
with
probably
cold
at
the
Google
team
to
help
lift
that
and
make
the
changes
to
the
munzur
so
that
the
requirements
for
labeling
and
whatnot
are
lifted.
A
And
after
that
point
we
basically
get
into
the
cherry-picking
business
for
1.10
and
technically,
because
master
branch
will
be
open
at
that
point.
For
111,
that
is
the
technical
beginning
of
1.11,
but
not
the
actual
recognize
early
start
simply
because
we
haven't
identified
a
release
lead
yet,
and
they
just
we
need
to
get
the
schedule
and
all
that
stuff.
So
there's
a
bit
of
a
lag
the
following
week:
we'll
have
the
the
retrospective
and
stuff
and
that
sort
of
wraps
up
the
110
cycle
and
marks
being
a
plan
11.
A
Does
this
seem
like
we're
on
track
for
release,
vote
amongst
the
team
and
have
a
discussion
around
that
just
so
that
we
can
prep
for
any
kind
of
changes
that
we
might
need
to
do
in
terms
of
the
marketing
activities
and
other
things
that
are
scheduled
around
release
day,
which
is
the
following:
Wednesday?
Yes,
so
again,
Wednesday
that's
the
day!
That's
when
we're
planning
on
doing
it.
A
Timing
is
a
little
bit
open.
It
depends
on
how
comfortable
we're
feeling
in
terms
of
making
the
final
decision
I
am
more
than
happy
to
delay
if
it
means
that
we
have
better
signal
and
better
quality
of
release.
If
there's
not
a
compelling
reason
to
delay,
then
we
won't
and
we'll
just
get
it
out
and
start
getting
the
ball
rolling,
but
typically
the
the
release
is
cut
again
around
5:00
or
6:00
p.m.
Pacific,
depending
on
how
the
that
goes,
the
process
takes
of
it.
A
A
Also,
next
week
we
will
be
managing
media
engagements,
we'll
be
handing
off
the
110
release
to
the
patch
manager,
and
so
basically,
after
after
the
release
is
cut,
the
patch
manager
becomes
the
go
no-go
for
any
additions
to
the
release
branch,
and
that
ceases
being
my
job
so
and
then
we
will
hopefully
have
a
111
release
leader
identified
soon.
That
will
take
the
reins
for
111.
A
Lastly,
next
week,
we'll
just
prep
for
the
retro.
So
if
there's
again
anything
that
you
think
of
as
we've
gone
through
this,
that
could
have
gone
better
or
things
that
you'd
like
to
see
done
differently,
definitely
hit
that
retrospective
document
and
there's
a
link
to
it,
as
always,
at
the
end
of
each
burndown
meeting
minutes
and
if
you're,
following
along
after
the
fact,
those
meeting
minutes
are
available
at
bitly,
slash,
gates,
110
burned
out
so
without
further
adue
metrics
review.
A
I'm
open
issues
in
the
milestone
currently
set
at
18,
and
that
is
down
for
from
Friday
critical
issues
at
18,
down
to
other
markers,
are
plus
or
minus
one.
So
they're,
not
really
a
big
deal
at
this
point.
There
are
number
of
milestone,
needs
attention,
issues,
and
so
what
I'd
love
to
do
is,
if
you
could
just
run
through
those
and
make
sure
that
if
there's
something
that
you
know
about
from
a
Sikh
perspective
that
needs
to
be
updated
or
if
you
need
to
get
consensus
on,
adding
it
to
the
milestone,
then
please
do
that.
A
I
have
also
looked
at
those
and
tried
to
update
them
as
I
go,
but
for
the
most
part,
that's
not
necessarily
my
responsibility.
It's
about
the
the
SIG's
making
sure
that
they're
comfortable
with
where
issues
are
at
and
they
should
or
should
not
be
the
milestone
open
PR
is
the
milestone,
are
currently
a
sixteen
that's
up
one,
which
is
fine.
The
the
things
that
are
out
there
seem
to
make
sense.
That's
AB
eleven
closed,
so
we're
actually
moving
things
into
the
the
milestone
which
looks
good.
There
are
currently
some
critical
PRS
out
there.
A
There's
four
I
put:
it
needs
attention
on
the
fixed
default
auditing
options
because
that's
been
out
there
for
a
while
and
I
want
to
make
sure
it's
getting
worked
so
important.
Soon
again,
there's
some
more
things
there
to
look
at
I,
put
a
put
a
link
to
those
in
the
mini
minutes.
If
you
want
to
look
at
them
and
some
test
blocking
both
of
the
test
blocking
PRS
need
attention,
so
they're
not
actually
in
a
state
where
they've
got
the
right
milestone
labels
in
place.
A
So
whoever
needs
to
look
at
those
that
would
be
super
great
just
so
that
if
they
need
to
be
in
there
they
are,
and
if
they're
not,
then
we
pulled
up
and
long-term
stuff.
Not
really
an
issue
and
test
great
desk
dashboards,
we'll
talk
about
as
we
move
forward
so
I
don't
see
Caleb.
So
we
can
skip
branch
updates
branch
health
unless
somebody
wants
to
add
something
there.
A
D
A
Okay,
that's
hot
okay,
many.
E
E
A
B
A
E
A
A
B
A
That's
actually
I
think
honestly,
that's
the
right
answer
and
Christoph
was
just
speaking
in
terms
of
preserving
Google's
interest
there.
So
then,
if
you
can,
if
you
can
track
that
down,
that's
actually
my
preference
I
would
prefer
to
do
that.
Well,
dude,
thanks
cuz!
We
definitely
don't
want
to
accidentally
do
something
that
is
going
to
impact
cheeky
cool.
Okay,
do
you
want
to
continue
through
the
stuff
that
you're
working
on
there,
son
yeah.
B
A
B
Yeah
from
the
last
update
they're
saying
this
was
beautiful
and
updating
fluent
point.
Oh
point
14
and
I'm
not
sure
sig
has
a
action
item
under
the
issue.
Maybe
we
should
we
worried
the
friends
when
they.
A
A
A
B
A
If
you
don't
get
any
traction
there
can
you,
let
me
know
at
all
and
I'll
go
hassle.
People.
B
B
F
B
F
Know-
or
maybe
this
that's
gonna
work-
this
test
is
just
slow
now
yeah,
but
that
that
went
in
through
the
exception
process,
so
maybe
Janet
could
help
put
that
reviews.
F
F
E
So
we
need
to
get
the
Cuban
artists
dashboard
problems
fixed
first
and
then
we
can
run
the
tests
on
this
PR.
But
then
the
question
is:
if
we
do
want
to
treat
this
as
a
release
blocker
at
the
three
112
of
eight
CD,
then
how
much
time
do
we
need
test
so
drawn
like
a
few
days
before
we
can
share,
because
this
seems
like
new,
eight
CD
versions
are
like
always
I.
E
E
A
Okay,
I
kind
of
feel
like
this
needs
this
@cd
thing
and
just
coordinating
this.
We
need
to
have
somebody
dedicated
to
it,
or
somebody
that's
willing
to
take
point
on
this,
because
this
feels
like
a
big
enough
issue,
that
this
could
really
interfere
with
their
will
the
release.
So
anybody
feeling
magnanimous
anybody
want
to
help
leap
weren't
on
this.
B
I
can
check
with
John
man
understand
the
issue
first
and
I.
Probably
you
also
want
to
grab
him
into.
We
have
meeting.
A
F
E
A
C
A
B
See
a
great
s
Weiss,
there
are
still
a
few
like
when
test
breakage
from
1:9
we
act
or
somewhat
emphasis
back
to
annoying
was
that's
when
that
branch
is
open,
but
I'm
not
too
worried
about
that.
Because
those
tests
are
passing
in
110,
we
know
they
will
fail
in
one
night
my
real
concern
is
squeeze.
The
dungaree
test
is
busted,
that's
fine.
You
see,
yeah
looks
like
a
stick.
Storage
and
suggest
issue
is
like
failing
me
to
believe
a
stateful
set.
B
A
A
H
F
H
Well
want
to
pick
that
up
in
the
next
class
board
and
tag
we
cut
sounds.
F
F
G
F
H
F
A
I
Informed
by
the
last
three
ducks
rank
those
Wranglers
for
the
last
three
releases.
The
talks
are
in
unusually
good
shape,
who,
as
well
as
attested
by
the
number
of
open,
docs
pr's
and
a
couple,
have
come
in
over
the
weekend.
That
surprised
me.
I
would
like
at
some
point
this
week
to
touch
base
with
Igor
to
make
sure
that
we
have
our
accounting
properly
for
features
but
I'm.
Thinking
that
our
these
look
like
small
things
and
I'm,
also
thinking
that
our
release
notes
will
take
care
of
things
so
yeah
we're
looking
at
good
shape.
I
My
biggest
concern
this
week,
I
may
need
to
pull
in
some
consulting
expertise.
I
am
slammed
with
not
one
but
two
simultaneous
releases
for
hefty,
oh
and
so
getting
the
generated
docks
out.
I
may
need
some
help,
but
I
have
volunteers
waiting
to
help
out
so,
okay
I
think
that
part
is
also
covered.
Okay,.
A
I
I
A
C
Just
getting
awesome
you
so
it
looks
like
some
of
the
features
are
changing
a
little
bit
which
we
expected
for
the
blog
and
so
I
just
want
to
make
sure
I
know
you
are
within
there
Jase.
C
C
C
A
D
A
C
Perfect
all
right:
well,
then,
I
will
move
forward,
often
an
email
around
only
Ford,
so
hovering
everybody
makes
their
their
edits
to
the
release
blog
and
then
my
recommendation
on
reaching
out
to
media
and
timing
of
interviews
and
timing
of
story
just
make
sure
nobody
has
any
concerns
and
then
we'll
move
forward.
On
our
end.
J
So
last
week
we
identified
that
there
is
this
regression
with
updating
at
CD.
Two
three
point:
two
point:
whatever
from
three
point:
one
point
ten
and
that
caused
an
increase
in
the
memory
usage
of
at
CD
2x,
and
we
also
are
observing
that
it.
It
led
to
increased
pod,
startup
Layton
sees
so
I
did
up
by
section
and
once
we
identify
that
we
after
discussing
is
a
bunch
of
folks.
J
Yes,
so
currently
I
am
doing
a
second
round
of
bisection
for
a
regression
that
came
after
this.
So
yes,
hopefully
I'll
be
able
to
point
out
this,
the
for
DPR
sometime
today
or
tomorrow.
It
depends
on
how
lucky
I
am
with
the
bisection
and
yes,
so
I
hope
this
wouldn't
be
any
significant
reward.
This
won't
need
any
significant
trivet
of
feature
or
so,
but
yeah
I'll
need
to
check
it
out
and
yeah.
A
Thank
you
so
much
for
looking
into
that
super
super
helpful.
If
there's
anything
that
I
or
anybody
in
the
team
can
do
to
help.
No,
no,
please
let
us
know
surely
do
we
do
have
an
idea
about
when
this
whole
issue
might
be
resolved.
Do
you
mean
just
your
gut
I
know
that
obviously
bisection
is
really
tedious,
but
yeah.
J
It's
a
bit
painful,
but
so
I
would
kind
of
refrain
from
like
saying
it
to
confidently,
but
I
think
if
there's
just
this
and
there's
no
more
regression
on
top
of
this,
it
probably
should
be.
We
should
be
at
least
knowing
the
problem,
if
not
debugging
it,
but
I
mean
it's
not
fixing
it
by
let's
say
by
Wednesday
or
maybe
tomorrow,
okay
and
yes,
I
just
remember.
Actually
there
is
another
issue
which
probably
is
not
this
serious
but
I.
It
still
matters
a
release.
Blocker.
Let
me
just
quickly
find
that
one
yeah.
A
J
J
Looking
into
so
I
could
I
didn't
manage
to
find
time
to
play
both
this
and
the
other
ones.
I
just
started
looking
into
this.
This
is
probably
not
that
serious
in
the
sense
that
this
is
being
seen.
Flaky,
like
some
of
these
failures,
are
being
seen,
flaky
ly,
but
yeah.
There
might
be
something
behind
this
as
well.
So
ok
I
will
give
a
better.
J
A
K
A
C
C
That
we
do
can
base
their
stories
on
that
and
a
lot
of
times
they
want
to
see
what's
in
the
what's
in
the
release,
so
that
way
they
could
get
permission
from
their
editors
to
conduct
interviews.
That's
usually
kind
of
that
initial
staff
to
be
able
to
set
up
interviews
for
you
and
you
get.
The
story
is
rolling,
so
I
just
want
to
make
sure
that
we
get
that
blog
in
a
good
place
by
Wednesday,
so
I
can
I
can
send
it
under
embargo
to
media
and
start
doing
interviews.
Okay,.
A
C
Do
so
I've
asked
you
and
you
are
in
Caleb
just
to
kind
of
look
through
the
blog,
the
upfront
portion,
the
title
and
the
future
sections
just
to
make
sure
that
everything
is
accurate
and
presented
in
the
way
that
we
want
it
presented.
I
have
Caitlyn
updating
the
five
blog
series,
such
as
listen
with
Iran,
that
the
release
team
section
has
been
written.
I
just
wanna,
make
sure
that
that's
how
we
want
to
prevent
it.
You
are
filling
in
the
project,
velocity
area
and
I
know.
C
We've
had
a
comments
Jean
on
that
and
then
in
terms
of
you,
study,
highlights
and
ecosystem
update,
tubecon
updates
and
how
you
get
involved.
Those
sections
are
done,
and
then
Caitlyn
is
just
filling
in
the
webinar
with
the
date
and
time
and
the
link
for
people
to
click
on
that.
So
a
big,
a
lot
of
it
is
written.
C
I
just
want
to
make
sure
that
you
know
it
is
that
source
of
truth,
and
everyone
is
happy
with
that
and
then
I
also
send
it
over
to
a
partner
to
see
if
she
had
any
additional
input
and
if
she
wants
it
to
go
over
to
to
Brian
and
Tim.
So
this
is
my.
This
is
my
main
focus
right
here
as
making
sure
that
everyone
is
comfortable
with
the
draft
for
me
to
send
that
under
embargo
to
media,
okay,.
I
C
I
and
I
won't
and
I
won't
send
it
until
I
have
sign-off
from
you
saying
yep.
This
looks
good.
We
are
comfortable
with
this
messaging
because
if
a
reporter
poll
sentence
from
sentence
verbatim
into
their
story,
I
want
to
make
sure
that
everybody
is
happy
with
that
and
I
think
that's
a
really
good
outcome,
because
they're,
you
know
really
staying
on
on
the
message
that
we're
getting
out
there.
Okay,.
A
L
I'm
still
here,
okay,
so
the
release
notes
are
in
a
state
right
now
that
if
you
put
a
gun
to
my
head
and
said
you
must
have
these
out
in
two
hours,
I
could
release
them.
I
could
just
basically
strip
out
a
bunch
of
extra
information
that
I'll
put
in
right.
Now,
if
you
go
and
look
at
the
release,
notes,
it's
easy
to
get
terrified
because
they're
93
pages
long,
but
of
those
93
pages,
probably
I,
don't
know.
L
A
L
A
L
L
L
L
A
That
all
right
I
will
I
will
look
at
it
at
that,
and
so
you
are
the
blocker.
Yes,
since
you're
the
the
sort
of
process
holder.
So
what
I'll
do
is
check
with
you
privately
offline
and
will
synchronize
some
time
windows
that
work
and
then
I'll
try
and
rally
around
getting
people
to
show
up
for
that.
All
that.
L
A
A
A
Okay,
as
always,
thank
you,
everybody
for
your
time,
I'm,
hoping
that
this
the
meeting
tomorrow
will
be
shorter
and
we'll
try
to
make
these
as
efficient
as
possible.
Thursday.
This
meeting
collides
with
the
community
meeting
if
you're
on
the
release
team,
the
core
release
team
I
would
expect,
hopefully,
that
you
attend
this
meeting
and
not
necessarily
the
community
meeting
and
we
will
go
from
there.