►
From YouTube: Kubernetes 1.12 Release Burndown Meeting 20180913
C
A
A
A
B
A
I'll
go
ahead
and
get
started.
It's
two
minutes:
past
I'm,
Tim
pepper
your
release,
112
lead.
As
always.
This
is
a
recorded
meeting
like
all
of
our
kubernetes
community
meetings
that
will
it
is
being
recorded.
It
will
be
uploaded
to
YouTube
for
the
kubernetes
channel
shortly
after
the
meeting
I'm
expecting
a
relatively
quick
meeting
today.
Not
a
lot
has
changed
yesterday,
specifically,
but
actually
a
lot
of
things
have
changed.
So
I'll
kind
of
highlight
those
briefly.
A
The
meeting
notes
agenda,
I'm
pasting
into
the
zoom
they're
in
their
normal
spots
and
the
normal
standing
sort
of
agenda
will
run
through
the
release
burned
down,
so
timeline
still
are
see
next
week
and
release
and
theory
the
the
target
remains
September
25th,
based
on
how
little
has
really
been
resolved.
This
week,
I'm
leaning
towards
making
a
call
on
Monday
to
delay.
A
So
considering
giving
it
a
couple
day
delay
so
looking
at
Thursday
the
27th,
depending
on
how
things
go,
that
said,
let's
dive
into
CI
signal
report,
so
virtually
nothing
has
changed
there
since
yesterday,
except
we
have
an
understanding
of
why
some
things
that
were
expected
to
be
changing
have
not
changed.
So
we're
actually
thinking
that
a
number
of
fixes
that
were
sort
of
pending,
but
it's
first
of
inferred
for
confusing,
raises
hadn't
flowed
in
we.
We
have
a
sense
of
why
that's
happening
and
roadmap
to
fix
this
morning.
A
So
I'm
slightly
positive
that
today,
tomorrow
we're
actually
gonna
see
some
good
changes
there,
but
virtually
nothing's
changed
take
notes.
Biconditional
work
is
ongoing,
but
still
sort
of
pending
getting
some
things
merged
that
had
confused
the
status
Sebastian.
Do
you
have
any
updates
on
any
of
the
the
gke
related
issues.
E
Syria
has
been
assigned
to
the
tmco
pinging,
some
of
the
leads
to
actually
get
some
progress
more
progress
on
this,
unfortunately,
I
mean
we
I,
don't
have
much
information
I'm
new
to
the
team,
so
it's
really
hard
for
me
to
debug,
and
so
that's
why
I
ping
the
league
to
figure
it
out,
because
this
is
working
and
everyone
and
we
need.
We
need
to
know
what
what's
going
on
so
yeah
like
I,
don't
have
much
update
today,
but.
B
A
Okay,
I'm
just
jotting
that
in
the
nuts
the
the
next
one,
then
on
the
the
list,
68
383,
the
horizontal
pod,
autoscaler
and
stability
Sally
has
been
debugging.
There's
a
pull
request
there,
something
to
help
with
debugging,
but
it
stalled
out.
The
tests
are
currently
failing,
but
I'm
I
feel
like
if
this
over
the
next
day
or
two
becomes
our
one
remaining
symptom,
we're
actually
in
pretty
good
shape,
and
we
have
the
the
subject
matter:
expert
on
it.
D
Guess
nothing,
that's
that's
the
main
status
of
the
appellant
or
not
so
it's
been
so
good.
I
think
I
should
say
I
actually
being
very
active
in
the
GK
stuff,
which
is
only
awesome
thanks,
I
and
I.
Hope
we
get
to
that
state
is
very
soon.
One
thing
is
the
couple
of
tools
that
we
use
are
down
like
the
loominator,
which
is
me,
which
might
be
a
problem
forward,
debugging
test,
so
they're.
A
Okay,
so
the
next
one
on
the
list-
68
3,
1
4.
This
is
the
the
storage
related
things
that
Michelle's
been
working
on.
She
thought
she
had
a
fix.
It
didn't
flow
correctly
into
1.11,
and
we've
done
some
looking
trying
to
understand.
If,
for
what
should
have
been
picked
up
when
or
where?
And
Aaron
yesterday
made
a
really
nice
document
that
describes
how
we
figure
out
what
code
should
be
under
test.
A
Hopefully
we
see
that
resolved
and
there's
a
few
other
things
that
seemed
lingering,
where
it
was
confusing.
Why
something
that
should
have
helped
wasn't
seeming
to
be
under
test?
Even
so,
hopefully
that
makes
a
notable
improvement
on
things
next,
one
on
the
list:
67
600
the
DNS
configurations,
so
that
one
is
fixed
on
master.
It's
gone
away,
so
awesome
progress
there.
It's
not
fixed
on
the
112
branch,
but
I
think
that
should
correlate
with
the
other
problems,
we're
seeing
where
things
aren't
going
to
branches
correctly
in
terms
of
the
image
building
and
testing.
A
So
we
just
have
a
lot
of
test
results
that
are
stale
and
hopefully,
that
clears
up
then
as
well.
The
load
balancing
issue,
we've
already
decided,
won't
be
blocking
and
we
do
have
a
fix
in
sight.
They're
looking
at
test
grid
it
looking
at
the
release,
112
all
and
blocking
things
are
actually
starting
to
look
like
they
could
clean
up.
If
we
get
the
storage
fix
in,
we
get
the
DNS
config
fix
in
and
the
some
of
the
taint
nodes
by
condition
things
actually
flow
in.
A
We
may
actually
start
looking
like
we're
in
pretty
good
shape.
So,
for
that
reason,
I
change
the
CI
status
from
red
to
yellow,
just
because
I
feel
like
we
have
some
line
of
sight
on
fixes.
If
over
the
next
24-48
hours,
things
improve,
we
could
actually
be
starting
to
look
yellow,
trending
green,
even
which
would
be
awesome
if
not
on
Monday
we're
gonna,
be
kind
of
getting
towards
pants
on
fire,
yellow,
trending
red
and
delaying
release
to
figure
out
what
the
next
layer
of
the
onion
here
is.
A
A
Moving
on
the
new
features,
I've
got
to
drop
this
for
the
moment
from
the
agenda.
I.
Think
after
last
week's
push,
we've
pretty
much
got
a
handle
on
things.
We've
got
40
features,
I
think
is
the
number
in
and
they're
all
on
target.
At
this
point
aside,
perhaps
from
Docs,
although
zakat
looks
like
missing,
Docs
account
has
now
gone
to
zero.
It's.
A
G
G
G
So
the
way
that
the
document
is
currently
constructed,
every
release
note
appears
in
every
section
for
which
it
was
tagged.
So
if
you
tagged
your
pr4
for
cigs,
we
have
four
copies
of
that
release.
Note
we
had
originally
left
it
figuring.
A
couple
of
duplicates
would
be
okay,
but
it
turns
out
to
be
ridiculous.
G
A
That's
reasonable,
I
would
maybe
say
pragmatically
if
it
does
seem
like
it.
It
if
you
have
some
that
are,
are
potentially
important
for
a
couple
of
the
SIG's
to
see,
maybe
not
just
limited
to
one
but
be
pragmatic
about
it.
If
it.
If
it
seems
like
it,
could
use
that
if
it's
something
that
would
merit
that
sort
of
visibility,
but
I
do
think
condensing
the
overall
length
would
mean
people
are
more
likely
to
read
through
as
opposed
to
skimming
quickly
and
then
you
they
might
see
things
that
been
irrelevant
better.
As.
G
A
G
G
What
I
had
planned
to
do
for
112
was
what
we
had
done
in
111,
which
worked
pretty
well,
which
was
basically
start
with
the
previous
releases
version
section
and
just
update
as
the
PRS
as
we
hit
the
PR.
So
there's
already
two
or
three
or
four
that
I've
seen
that
say
you
know
bumped
this
to
that
version,
etc.
So
that
worked
out
pretty
well
and
111,
so
I
was
gonna.
Do
it
again
in
112
dims.
If
you
have
another
suggestion,
I
am
totally
up
for
it.
C
G
A
A
Communications
Caitlin
I
I,
know
I
gave
you
some
feedback
yesterday
on
key
messages,
I'm,
not
sure.
If
you've
gotten
the
the
info,
you
were
looking
for
also
from
from
Stephen
and
then
I'd
also
looped
in
ice,
also
to
start
giving
her
some
visibility
on
this
stuff
ahead
of
the
next
release,
where
she'll
be
more
involved.
H
A
So
we'll
look
for
that
call
on
Monday
and
I'll,
let
you
know,
okay,
that
is
basically
the
the
end
of
the
agenda.
There's
a
few
things
that
are
tacked
on
the
agenda
for
113
future
miscellaneous
bits
of
stuff
that
we've
given
to
mention
too,
and
each
of
the
prior
meeting
was
just
be
thinking
about
where
we're
headed
next
anything
else.
Anyone
wants
to
talk
about
with
urgency
today.
A
A
All
right,
if
not
I'm,
gonna,
go
back
to
crossing
my
fingers
and
watching
test
grid
and
seeing,
if
there's
any
yards,
that
need
moved
ahead
for
for
fixing
things
are
so
fingers
crossed
with
a
little
luck
today.
Tomorrow,
things
are
gonna
set,
I'll
settle
out
positively.
We
will
see
all
right.
Thank
you
all
we'll
see
you
tomorrow,
bye.