►
From YouTube: Kubernetes 1.14 - Week 11 - Burndown 2019-03-20
Description
A
B
B
B
A
B
B
C
B
B
B
D
D
Okay,
so
hi
everybody
I'm
Aaron
of
cig
beard
today
is
Wednesday
March
20th.
You
are
all
at
the
kubernetes
114
burned
down
meeting
we're
all
going
to
publicly
adhere
to
the
kubernetes
code
of
conduct
and
not
be
jerks
as
we
are
recorded
and
posted
to
YouTube
later
today's
cat
t-shirt
is
Schrodinger
Schrodinger
and
is
many
cats
misbehaving
like
they
just
can't
wait
to
get
out
of
the
box
like
I?
Just
can't
wait
for
these
to
actually
happen.
So,
let's
move
on
to
enhancements
from
Claire.
E
F
Sounds
good
not
too
much
has
changed
since
yesterday
there
was
a
peer
merged
one
of
the
two
pairs
that
were
interested
interesting
to
us,
that
deflect
storage
upgrade
tests.
There
is
another
that
I'm
hoping
will
merge
before
we
release
in
Monday
and
it
addresses
the
race
condition
in
the
cupola
volume
manager.
F
F
The
other
issue
that
I'm
looking
at
is
one
on
a
flaky
test
that
is
related
to
pod
reconciliation.
From
what
I
remember
there
was
a
comment
by
recon
is
the
github
name?
I
can't
remember
his
real-life
name.
He
said
he
would
look
into
whether
there
would
be
a
problem
there
with
district
installation
happening
as
often
as
they
expected
other
than
that.
We're
pretty
steady,
114
blocking
is
is
passing.
There
is
a.
There
have
been
two
failed
transfer,
scalability
in
master
blocking
starting
today.
My
plan
is
to
look
at
master
blocking,
but
not
master.
F
D
D
Recommend
not
looking
at
master
if
you
can
help
it
just
because
there
are
actually
I,
don't
know
how
many
PRS
emerges
lately,
but
there
are
now
a
number
of
PRS
master
right
that
just
will
never
end
up
in
the
least
114
unless
they're
intended
to
be
cherry
picked.
So
the
state
of
master
is
now
very
different
from
the
state
of
114
and
I
would
not
try
and
rely
on
it
for
meaningful
signal.
Oh.
F
D
We'll
get
to
it,
we
get
to
test
in
for
a
guess.
Okay,
any
other
questions
for
Maria
cuz,
like
super
awesome
to
actually
see
both
master
blocking
in
1:14,
blocking
with
no
red
on
them.
I
mean
it
happened
for
a
little
bit
yesterday
and
then
Tim
pepper,
said
something
out
loud
and
well.
Here
we
are,
but
it
does
seem
like
things
have
been
improving,
which
is
always
great.
D
D
D
G
D
Yeah
for
what
it's
worth,
that
that
flake
about
volumes
failing
or
flake
bug,
whatever
you
want
to
call
it
body
was
failing
to
clean
up
when
the
couplet
restarts
due
to
a
race
condition
between
the
actual
and
desired
state
like
I
talked
to
offline,
and
there
was
some
consternation
that
hey.
Maybe
this
is
a
plate-
that's
been
around
for
a
while,
and
we
want
to
be
mindful
of
solving
it
correctly,
rather
than
potentially
like
trying
to
fix
this
one
thing
and
causing
other
implications.
D
H
Basically,
the
problems
are
because
historically,
we've
been
creating
the
new
jobs
based
on
the
old
release
instead
of
master
and
I.
Think
the
while
is
going
to
audit
all
of
the
changes
that
that
were
possibly
affected.
Because
of
this
we
might
need
to
accommodate
accommodate
that
change
in
our
automation
plan,
but
other
than
that
we
should
be.
We
should
fix
all
of
the
jobs
today
as
soon
as
we
get
to
know
what
needs
to
be
done.
Okay,.
D
From
a
work
tracking
or
a
XI
tracking
perspective,
should
we
have
an
issue
open
to
cover
that
audit,
or
do
we
feel
like
that
is
really
close
to
Maria's
work
to
try
and
hawed
it?
The
like
so
I
believe
Maria
is
trying
to
audit
like?
Could
we
have
the
114
stuff,
look
like
the
master
stuff
as
a
source
of
truth
and
I?
Think
what
we
ran
into
here
was
that,
could
we
have
the
114
job
configs
use
the
master
job
configs
as
their
source
of
truth,
because
it
appears
in
both
cases.
D
I
J
K
We're
also
green
we're
pulling
in
new
changes
as
they
happen.
I
did
add
the
most
recent
pull
requests
for
the
release,
notes
tool
on
github,
but
we're
ignoring
that
for
the
most
part,
because
we've
discovered
that
the
generated
notes
don't
keep
things
in
order
in
the
same
order
from
generation
to
generation.
So
it's
nearly
impossible
to
keep
it
updated
with
the
Google
Doc.
But
Jeff
has
added
some
fixes
to
the
release,
notes
tool
that
addressed
that
so
he's
tracking
the
changes
in
his
own
branch
of
bringing
them
in
from
there.
K
So
that
will
hopefully
go
into
the
release,
notes
tool
very
soon
and
let's
see
dependencies
were
updated
last
night
and
we
have
to
give
a
big
shout
out
to
join
Leggett
for
doing
a
lot
of
work
for
us
on
the
release,
notes
team
so
for
anybody
else
that
still
needs
to
look
at
the
release.
Notes.
Please
do
so
we're
gonna
be
editing
them
continuously
until
the
release
day.
D
So
I
am
trying
to
mostly
just
keep
the
next
couple
of
days
as
open
as
possible
to
be
as
responsible
as
as
responsive
as
possible
to
any
media
requests,
as
well
as
anything
that
arises
from
a
wall.
So
that's
why
I
like
scheduling
times,
probably
the
best
way
for
me
to
get
a
good
commitment
to
you
as
far
as
I
know
we're
all
on
track.
There
Honus
release
branch
management.
L
Sorry
about
that
hello,
everybody
release,
branch
management
is
green.
We
cut
the
1:14
rc1
yesterday.
The
next
cut
would
be
their
release,
except
we
decide
for
some
reason
to
cut
the
Narcy
at
some
point.
I
think
it's
not
planned
I'm
open
and
happy
to
do
that.
I,
don't
know
how
much
sense
it
makes
like
home.
How
much
feedback
do
we
get
for
our
c1
and
therefore
might
consider
cutting
our
c2,
but
anyway,
we
can
decide
on
on
that
now,
as
we
stop
doing
branch
fast-forwards,
that
we
need
to
take
a
look
at
the
cherry
picks.
L
L
D
It
is
ultimately
at
your
discretion
because
you
own
the
branch
I
think
that
what
would
probably
be
cool
is,
if
we
chatted
about
them
like
give
a
heads
up
that
you've
noticed
one
in
sake
release
we
can
chat
about
it
there
and
then
come
to
a
decision
on
the
phone
request
as
needed.
I
did
ask
people
when
I
sent
out
an
email
about
the
way
things
work
now.
Is
that,
regardless
of
what
they're
trying
to
do
to
land
something
into
114,
they
come
and
talk
to
us
in
sake.
Release
anyway,.
L
Yeah
looks
good
to
me,
I
guess
I
guess
controversial
stuff.
We
can
also
then
discuss
on
the
daily
burn
down,
which
we
still
have.
Yes,.
D
All
right
wait,
so
I
love,
you
all
I
trust
you
all
you
do.
You
do
great
work.
So
if
it
is
your
opinion
that,
like
this
looks
pretty
it
probably
really
needs
to
go
in,
then
I
trust
your
judgment
and,
like
worst
case,
if
we
have
to
revert,
we
have
to
revert
that's
okay,
but
if
you're
really
uncertain,
then
you
are
always
welcome
to
wait
or
a
decision
from
me.
D
C
Them
I
guess
the
first
thing
I
would
mention.
Is
the
mechanism
just
to
affirm
for
Honus,
you
go
in
the
github
UI
and
you
apply
the
label
manually,
it's
not
a
command,
but
then
the
process.
Yes,
historically,
branch
management
after
release
was
on
a
single
person,
so
they
were
kind
of
sadly
left
to
their
own
choice.
Making
for
113,
Alexandre
and
I
have
been
trying
to
weigh
each
of
them
together.
Have
a
brief
chat
and
I.
C
We
don't
want
this
or
yes,
we
want
it
now
and
yes,
we
want
to
defer
it
to
the
patch
team
and
there
it's
been
kind
of
the
gut
of
of
everybody.
So
I
think
just
ask
questions
early
and
often
and
I.
Definitely
in
the
112
release.
There
was
one
where
I
I
had
a
nagging
worry
and
I
minimally
mentioned
it
and
then,
like
a
week
later,
it
blew
up
and
I
regretted
not
having
talked
about
it
more
so
just
like
that,
there's
no
harm
and
conversation.
D
D
D
I
D
Right
so
my
gut
is
telling
me
it's
not
bother
cutting
an
additional
RC,
but
I
think
we
reserve
the
right
to
change
our
minds
about
that
tomorrow
and
since
we're
having
tomorrow's
meeting
a
little
bit
earlier,
I
think
that
gives
you
enough
time
to
do
the
needful
if
you
need
to
start
a
cut
after
that
meeting.
That's
not
very
honest.
D
L
I
D
D
So
thank
you
for
bringing
up
an
idea
of
documenting
how
we
could
do
better,
because
this
is
the
part
of
the
release
life
cycle,
where
I
think
we
start
to
breathe
a
little
bit.
We
start
to
reflect
on
how
we
did
what
we
did,
what
we
liked,
what
we
didn't
like
and
there's
this
retrospective
document
available
for
us
to
brainstorm
in
in
anticipation
of
the
release
going
well
and
going
out
the
door
on
Monday.
Assuming
that
all
happens
is
scheduled.
D
The
plan
would
be
to
take
over
the
community
meeting
about
Thursday,
so
I
think
that
would
be
Thursday
March
28th.
The
Thursday
is
right,
I
don't
know
about
the
day,
and
so
that's
where
we
can
talk
about.
You
know
what
we
think
went
well
what
we
think
we
could
do
better,
and
so
this
sounds
like
a
perfect
topic
for
that.
So
I
dropped
a
link
to
the
doc
in
the
chat.
A
link
to
the
retro
doc
as
far
as
I
know,
is
also
available
at
the
top
of
the
meeting
notes
and
on
the
release
page.
D
So
I
encourage
any
and
all
of
you
who
have
thoughts
too,
to
contribute
there.
The
only
other
main
thing
that
I
have
I
sent
out
requests
for
some
information
on
the
release
team
mailing
list.
Thank
you
to
those
of
you
who
have
responded
already.
I
would
like
to
get
as
much
of
that
information
as
soon
as
I,
possibly
can
because
of
the
lead
times
required
for
that
information.
D
So
if
you
haven't
checked
your
inbox
lately
or
if
you
don't
have
the
kubernetes
release
team
Google
Groups
setup
to
automatically
mail,
you
maybe
go
check
it
out.
If
you
are
on
the
release
team,
it's
got
a
big
action
required
thing
in
the
title
and
so
because
of
all
the
fine
work
that
everybody
else
has
done,
even
though
we
do
seem
to
have
these
perpetual
flakes
going
on
and
I'm
like
a
little
weary
about
the
jobs
and
how
we
created
them.
I'm
overall
feeling
good
enough
to
call
the
release
status.