►
From YouTube: Kubernetes 1.13 Release Team Meeting 20181029
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
Hey
everybody
welcome
to
yet
another
release
team
meeting
for
those
who
just
joined
I
paint
the
link
to
the
minutes
in
the
chat.
Please
add
your
attendance
in
the
minutes
and
the
usual
disclaimer
I'm,
recording
it
right
now
recording
the
entire
meeting
right
now,
which
will
go
online
in
our
YouTube
channel.
So
please
be
mindful
of
what
you're
saying
with
that:
let's
get
rolling.
A
So,
as
I
said,
we
are
in
week
5
almost
at
the
midpoint
I
think
this
is
where
this
release
gets
really
squished.
It
seems
like
when
I
go
through
the
calendar.
It
feels
like
we've
week.
5
6
7
of
the
usual
cycle
is
almost
between
this
week
and
part
of
next
week,
so
we
are
quickly
proceeding
towards
our
other
milestones
later
next
week.
As
for
announcements,
we
cut
our
alpha
too
early
last
week,
thanks
to
harness
our
branch
management
shadow
was
able
to
successfully
cut
it
and
publish
the
bits.
A
We
still
need
a
dog
to
do
some
notify
steps
at
the
end,
but
other
than
that
it
was
super
smooth
to
get
the
release
both
stage
and
published.
So
thanks,
hem
is
doing
that.
We
have
alpha
3,
which
which
will
go
out
on
Wednesday
this
week
again.
This
is
mainly
to
give
yang
our
other
shadow
chance
to
try
the
pipeline
itself,
so
alpha
3
is
slated
for
Wednesday
and,
as
I
promised
last
week,
I
started
a
release
log
for
113.
A
A
It's
in
no
specific
order
of
flow,
but
there
are
some
divisions
for
things
that
I
kinda
started
keeping
track
of
during
the
release
I've,
given
the
entire
release
team
edit
access
to
it
so
feel
free
to
add
an
add,
add
to
any
of
the
sections
where
you
might
have
more
historical
context
or
more
data
on
how
we
did
it
in
the
past
releases
once
to
last
week
and
go
through
it
towards
the
end
of
the
release.
I'll
formalize
it
a
little
bit
more
clean
it
up
and
then
we'll
see
where
we
can
document
it.
A
Next
up
our
upcoming
dates.
There
was
so
a
beta0
which
is
when
we
cut
our
release
branch
and
create
the
CI,
and
everything
is
scheduled
for
next
Wednesday.
That
is
eleven
six
we've
talked
about
the
status
of
Vita
and
if,
if
we
are
still
going
to
make
the
date
as
we
talk
about
the
CI
signal,
but
in
the
calendar
it's
scheduled
for
next
Wednesday
and
the
tasks
there
I
don't
know
if
Doug
is
on
the
call
today,
but
yeah
we'd
start
we'll
cut
our
branch
and
our
daily
fast-forward
start
from
that
point.
A
B
It
going
everyone
we're
officially
in
freeze
after
last
week,
so
thank
you
to
everybody
on
the
team
that
helped
really
put
together
a
lot
of
minds
to
kind
of
get
that
to
where
it
was
getting
on
all
the
issues
themselves,
as
well
as
the
shadow
stepping
up
and
helping
around
there
as
well.
We
really
knocked
out
of
the
park
on
that
one,
so
there
are
34
current
issues,
of
which
you're
being
tracked
with
inside
there
it's
down
from
three
before
we
went
into
the
freeze
as
well.
B
B
The
other
thing
that
you
had
written
down
here
is
looking
to
the
this
week
and
next
week
is
actually
looking
at
the
open
fears
with
inside
of
each
issue
and
going
through
KK
and
seeing
exactly
where
they
are
on
the
process
and
if
we
need
to
keying
some
of
the
owners,
whether
it
has
to
be
for
code
review
or
for
tests
or
anything
like
that,
just
to
make
sure
it's
consistently
getting
a
little
bit
of
churn
in
there.
I
will
probably
need
a
little
bit
of
help
on
that
next
week.
B
B
And
then
I'll,
let
you
kind
of
get
an
update
on
the
the
QB
dmg
a
because
I
know
that
you
were
pinging
some
of
the
the
issue
owners
and
there
to
kind
of
see
exactly
and
that's
that's
also
a
thing.
That's
happening
in
some
of
the
issues
themselves,
as
they
are
these
umbrella
issues,
and
so
it's
just
getting
to
a
point
now
where
we
have
an
idea
of
what
is
actually
the
the
PRS
that
we
should
be
tracking
for
each
one
of
these
and
how
it's
progressing
for
each
of
these
stages
as
well.
Yep.
A
So
this
is
something
that
I
am
little
bit
concerned
about,
mainly
because
code,
slash
is
just
a
week
and
a
half
away
and
I
I'm,
not
sure
on
the
exact
status
of
where
things
are
I
see
Luba
mayor
on
the
call
I'm.
Not
do
you
have
any
updates
on
the
state
on
the
progress
of
the
issue
itself.
Oliver
there.
C
A
My
I'm
now
my
my
request
was
more
of
is
there
ETA
from
by
when
we
expect
all
of
them
to
merge
and
be
called
done,
because
we
have
about
three
tracking
issues
in
the
enhancements
people.
So
one
of
my
question
was
should
be
closed,
especially
issue
number
296
and
356
in
favor
of
the
umbrella
issue,
which
basically
is
tracking
queue
Backman
going
to
GA.
C
A
A
C
C
A
Okay,
okay
and
then
the
other
part
in
this
whole
thing
is
talks
as
well.
I'm,
not
sure
if
any
of
the
docs
be
ours
have
been
open
at
this
point
and
what
the
story
there
is,
can
a
Kendrick.
Do
you
think
you'll
have
time
to
hop
on
to
the
cluster
lifecycle
meeting
tomorrow?
If
not,
I
would
do
that,
but
I
mean
I'm
gonna,
be
in
a
shuttle
tomorrow.
At
that
time,
yeah
that.
A
Okay,
I
think
I
will
follow
up
with
Timothy.
We
are
email
and
then,
if
need
be
tanned,
we
can
ask
him
to
join
our
release
meeting
next
week
to
give
a
little
bit
more
update
on
the
status
of
things
but
yeah.
This
seems
like
a
black
box
to
me
and
I
would
like
to
get
more
update
and
also
time
box
the
work,
the
remaining
work,
so
that
we
can
make
a
call
for
1:13.
Sooner
than
later,
yeah.
C
So
the
biggest
things
in
terms
of
dogs
are,
we
basically
need
to
change
a
bunch
of
commands
like
we
are
moving
commands
from
an
alpha
state
to
a
GA
state
directly,
and
that
is
something
we
should
address,
but
that's
an
easy
change.
We
don't
have
any
big
PRS
that
for
the
websites
report
that
we
should
write
to
be
honest,
okay,.
A
D
E
D
And
how
it's
being
worked?
That's
really,
the
only
thing
that's
throwing
us
off
here
is
the
appearance
that
failures
mysteriously
crop
up
and
then
it
seems
as
though
nobody,
but
perhaps
yourself,
Luba
Mir,
are
actually
noticing
that
these
failures
happen
and
I'm
sure
it's
because
there's
lots
of
follow-up
work
to
make
sure
that
all
the
T's
are
crossed
and
all
the
I's
are
dotted,
but
from
the
release
team's
perspective,
making
sure
that
things
are
always
running
and
always
green
is
a
really
important
thing
to
us.
So
ya.
D
A
A
E
So,
overall,
we're
improving
a
lot
of
stuff
has
gotten
close
over
the
last
five
days.
I
do
need
to
check
up
on
those
closures,
because
we've
had
a
dismaying
tendency
of
people
to
put
fixes
issue
number.
What
in
there,
PRS,
which
results
in
auto
closures
of
broken
desks
that
aren't
necessarily
fixed
but
barring
following
up
on
that
are
mean
issues
master
blocking
is
looking
good.
We
just
have
that
one
cubed
min
failure,
which
is
new
I.
E
E
Mean
it's
so
improved
right
now,
we've
got
four
test
failing,
although
a
lot
of
the
others
are,
of
course
flaky
as
usual,
which
is
still
a
source
of
concern
now
of
the
ones
that
are
feeling
three
that
are
failing
that
are
just
you
know,
Marcus
is
feeling
too
often
and
or
flaking
too
often
our
gke
tests,
no
two
of
them,
are
G
Kita's.
My
tendency
is
to
ignore
those
from
a
blocker
perspective
based
on
Aaron's
new.
E
E
I'm,
not
you
know,
I've
been
interacting
with
the
SIG's
about
those
issues
trying
to
get
more
effort
on
them
and
it
just
so
I
I.
Don't
honestly
think
things
are
going
to
accelerate
until
we
say
okay
beta
is
supposed
to
be
a
week
away,
but
we're
planning
on
delaying
it
now,
based
on
these
two
tests,
yeah
so.
D
I'm
gonna
ask
you
for
some
data
Josh,
ladies
a
that
people
are
being
slow
to
turn
things
around
you're,
going
to
have
to
start
proving
how
you're
measuring
that
and
with
available
data.
This
is
not
me
like
just
because
this
is
going
to
help
us
as
we
look
to
build
the
reliability
metrics
into
jobs.
It's
something
that
is
gonna
have
to
be
kind
of
hand,
rolled
I,
guess
for
now.
D
Cuz,
like
the
broader
criteria,
are
kind
of
going
to
be
like
when
the
job
started
going
red
and
when
the
travel
in
green,
and
so
that
tells
us
worst
case
like
how
long
did
it
take
the
job
to
get
back
to
green,
but
the
humans
are
going
to
start
to
jump
up
and
down
and
say
well,
like
I
didn't
know,
this
was
a
problem
until
X,
so
we're
going
to
at
least
for
the
shooter.
So.
E
I
first
noticed
that
that
was
failing
25
days
ago,
because
it
was
basically
failing
when
I
started,
to
see
I
signal,
and
did
you
open
an
issue
at
that
open,
an
issue
at
that
time,
one
of
the
sig
leads
assigned
it
to
Joey.
Jim
who's
been
working
on
it.
He
did
some
analysis
of
the
failure
20
days
ago
and
then
submitted
his
first
PR.
E
E
E
D
I
understand
that
so
I
think
what
I'm
asking
for
is
some
view
of
that
in
the
aggregate.
If
you
have
some
kind
of
spreadsheet
or
report
where
you
can
show
us
for
every
failure
that
you've
been
tracking
when
you
first
noticed
it,
my
name
God
first
got
a
response
when
they
first
tried
to
fix
it
and
when
the
problem
was
finally
solved,
I
think
that
those
could
be
some
useful
numbers
to
help
us
get
and
like
who's
being
responsible,
I,
okay,
I'm
thinking
of
we
can
map
those.
D
Eventually,
we
can
map
those
two
if
we
mandate
that
people
have
tests
written
or
learned
sent
to
them
like
cluster
life
cycle
is
doing.
Thank
you
again
very
much.
If
they're
sent
some
publicly
accessible,
Google
Groups,
we
can
start
to
use
that
as
time
and
and
we
spread
the
message
that
people
need
to
actually
be
subscribed
to
those
Google
groups
who
are
going
to
be
in
charge
of
maintaining
tests.
We
can
start
to
kind
of
piece
that
together
ourselves,
but
in
the
meantime
it
sounds
like
you've
been
keeping
a
pretty
good
log
of
this.
D
E
D
Think
that's
something
you
and
I
can
collaborate
on
offline
as
far
as
your
statement
about
the
gke
tests
and
being
and
ignoring
them,
I
agree
with
that.
I
have
had
to
sort
of
float
that
idea
around
internally
and
we'll
be
moving
forward
with
that
today
and
I'll
reach
out
to
you
to
make
sure
that
we
make
make
sure
your
fault
you're,
using
the
right
processes
that
I'm
not
stepping
on
your
toes
that
we've
got
the
right
Doc's
kept
up
to
date.
Things
like
that.
A
E
A
So
that
yeah,
that
just
brings
us
to
the
next
question
as
to
at
what
point
do
we
block
beta,
zero
I?
Think
we've
mainly
we're
tracking
the
cube
admin,
the
scheduler
priority,
an
HPA
one
right?
There
are
a
couple
of
other
intermittent
flakes,
but
we
do
like
to
see
these
three
kind
of
fixed
and
passing
at
least
on
the
GC
jobs.
E
A
E
Well,
I
want
to
say
our
active
test
grid
is
now
working,
so
I
wanted
to
really
thank
Nico
for
that,
because
it
makes
it
easier
for
us
to
sort
of
track
things
on
an
ongoing
basis.
And
so
what
happens
there?
For
people
who
looked
at
it
is
we've
got
like
eight
columns
that
are
automatically
populated
from
open
issues
and
then
another
or
five
columns
that
I
in
the
CI
signal
team
maintained
two
updates
at
us
on
issues
and
and
those
are
now
kind
of
complete
lists.
E
It
doesn't
mean
that
we
have
a
few
old
things
that
we
don't
actually
care
about
in
there,
but
there
wasn't
a
good
way
to.
There
was
a
good
way
to
filter
out
the
old
fence.
We
don't
care
about
and
still
include
build
things
that
we
do
care
about.
So
the
you
know.
So
that's
in
there
for
follow-ups
and
particularly,
were
populating
a
column
that
says
well.
E
A
Okay,
any
questions,
if
not
really
want
to
buck
triage.
Great
okay
nico
is
not
going
to
be
able
to
join
us
today,
but
yeah
we
kind
of
going
forward.
As
Josh
was
saying.
We
also
have
automated
spreadsheet
for
issues
and
PR
tracking
as
well,
which
Nico
put
together,
but
the
the
spreadsheet
is
hitting
some
rate
limits
that
nico
is
actively
working
on,
but
going
forward.
I
would
like
to
kind
of
have
a
blown
status
of
the
bugs
in
here
in
keeping
with
what
we
did
with
the
previous
releases
fitness.
A
It
gives
us
good
insight
into
which
issues
or
which
PRS
might
not
have
complete
labels
and
also
kind
of
helps
us
scrub
them
pretty
easily.
So,
as
of
yesterday,
we
have.
This
is
all
our
open
issues
on
113
that
are
a
the
kind
parka
kind
cleanup
these.
These
exclude
any
feature
issues
flaky
or
failing
tests.
A
We
have
one
critical
urgent,
which
there
is
quite
there's
some
active
traction
on
it
and
we
have
about
11
important
soon
so
this
week,
Nico
and
I
would
be
working
mainly
on
these
important
soldiers
to
get
a
status
on
where
they
are
and
just
to
make
sure
we
let
them
we.
Let
folks
know
that
code,
slash
is
just
a
week
away
and-
and
there
are
about
11,
that
need
labels
to
be
completed.
They
a
them
is
saying
kind
of
priority.
So
this
is
what
we
would
tackle.
A
First,
just
work
with
owners
to
figure
out
if
it's
still
in
scope
1:13,
if
so
at
the
milestone
on
fill
up
all
the
labels
or
removed
from
milestone,
if
it's
not
really
due
to
any
enhancement
work
at
this
point
yeah.
So
that's
what
Nico
will
be
focusing
on.
Is
there
any
questions
or
anything
that
we
should
be
doing
at
this
point?.
F
Hey
yeah
how's
it
going
so
we're
definitely
behind.
We
want
to
be
with
ducks.
We've
got
the
placeholder
PR
deadline
coming
really
soon
and
then
the
actual
docks
deadline,
following
up
soon
after
so
gonna,
be
making
really
big
push
this
week
to
get
a
place
where
we're
happy
yeah
so
behind.
But
you
need
to
make
a
push.
There
also
will
be
generating
the
documentation,
the
January
documentation
for
the
next
beta,
so
that
will
be
available
once
that
comes
out
so
beta
one
great,
so
IB
2
zeroes
next
one.
Yes,.
A
Yes,
but
I
don't
think
we
actually
it's
more
of
a
release.
Branch
cut
and
generating
on
the
release.
Artifacts,
yeah,
Tim,
Tim
pepper,
is
on
the
call.
You
can
correct
me
if
I'm,
but
from
what
I
believe
it's
we
don't
read
it's
not
an
actual,
but
more
of
cutting
the
release
branch
and
generating
the
CI
jobs
and
everything
else.
There.
A
Caleb
I
didn't
see
you
there
cool
so
yeah.
We
think
you
still
have
another
week.
Two
showed
up
those
things
but
awesome
and
yeah.
If
you
need
any
I've,
been
pinging
ish,
hyping
enhancement,
PRS
I've
been
like
kinda
asking
them
for
talks
updates
as
well,
but
if
you
need
help
in
again
giving
up
the
job
of
bugging
folks
for
dogs,
let
us
know
yeah.
G
My
only
update
is
that
so
far
no
release
notes
via
commits
but
I
started
working
with
Steven
on
the
and
enhancements
tracking
spreadsheet
that
he
has
in
order
to
populate
things
like
major
themes
and
stuff
that,
in
the
release,
notes
document
which
you
know
previously
has
been
mostly
a
manual
amalgamation.
But
you
know
well,
since
this
is
a
short
release
and
then
there
is
not
a
lot
of
activity
in
release,
notes
generated
from
commits,
and
it's
early
still,
but
you
know,
might
be
a
good
opportunity
to
expand
document
that
we
generate
with
additional
sources.