►
From YouTube: k8s 1.16 - Week 10 - Release Team Meeting 20190906
Description
Release details: http://bit.ly/k8s116
A
Hello
and
welcome
to
the
1/16
release
meeting
today
is
September
6.
That
is
a
Friday.
This
is
week
10
and
we
are
in
Bern
down
the
end
of
week.
One
of
burn
down
I've
got
a
packed
agenda
today
before
we
get
into
it,
I'd
like
to
remind
everybody
that
this
meeting
is
being
recorded
and
will
be
posted
to
YouTube,
so
be
mindful
of
your
conversations
and
all
conversations
are
governed
by
the
kubernetes
code
of
conduct,
which
says:
please
be
excellent
to
each
other.
I
do
not
have
my
flash
cards
today.
A
A
B
C
Thanks
and
I'll
second,
that
Eddie
definitely
dead
of
the
year.
That's
pretty
pretty
impressive,
so
happy
Friday,
everybody!
We
are
currently
red
and
we
are
red
because
of
a
failing
and
master
blocking,
and
also
in
the
116
blocking
so
I've
linked
the
open
issues
in
the
agenda,
and
there
currently
is
work
being
done
to
investigate
him.
I
believe
that
Valerie
as
just
pings
a
signet,
were
to
look
a
little
bit.
C
Additionality
into
the
reasons
that
they're
failing
I,
don't
a
ton
of
context
on
the
failure
but
I
believe
there
was
some
things
hashed
out
yesterday
in
a
meeting
with
scalability,
and
with
that
meeting
there
was
also
some
known
performance
regression
with
kubernetes
and
go
for
113.
That
was
brought
to
our
attention,
as
well
as
the
ongoing
efforts
between
SIG's
scalability
and
the
release
team
and
CI
signal
to
coordinate
a
little
bit
better.
C
So
with
that
we're
gonna
be
meeting
with
them
monthly
from
here
on
out
for
the
release
cycles
to
kind
of
stay
more
in
sync
with
those
folks,
so
we
have
the
master
blocking
issue,
the
ones
that
are
flaking.
As
far
as
I'm
aware,
they
are
not
an
issue
and
then
we
have
ten
passing
in
the
116
blocking
and
the
116
informing.
We
have
a
one
failure,
stay
KS,
one
that
was
mentioned
yesterday
or
for
the
past
couple
days.
That's
a
known
issue:
it's
not
actually
released
blocking
Matt,
well,
I!
C
Guess
it's
informing
flicking
and
passing
and
then
one
other
note
on
the
Alpha
features.
I,
don't
they
mentioned
it,
but
dims
opened
up
a
PR
to
catch
the
failures
earlier
on
in
the
cycle
versus
later
now
to
hopefully
catch
some
of
these
issues
earlier
in
the
cycle
before
actually
making
any
sort
of
impact
for
master
informing,
we
have
the
AKS
one
failing
as
I
mentioned,
there
is
six
flaking
ten
passing
and
then
you
can
review
the
notes
for
the
release
miscellaneous
as
well
as
the
other
ones
there.
With
that
many
questions.
A
A
So,
basically
the
way
the
feature
gates
are
getting
it.
This
will
fix
all
the
network.
Cygnet
work
problems
in
alpha
that
are
failing.
Basically,
the
end
point.
Slice
thing
is
not
being
activated
correctly
across
all
the
components
which
is
this
new
enhancement
and
the
PR
should
go
and
set
the
right
flags
so
that,
when
cube
up
has
run
the
all
the
component-
and
this
will
have
the
end
point
sliced
thing
enabled
so
I
just
had
a
chat
to
the
the
enhancement
owner
for
endpoint
slices,
his
name's
Robert
Scott
over
at
Google.
A
So
he's
put
it
in
one
place,
but
it
hasn't
affected
another,
and
that's
why
the
Signet
work
piece
is
not
working,
because
it's
not
making
that
flag,
isn't
making
it
a
cube
proxy
which
needs
to
know
about
endpoint
slices
to
actually
fix
this.
So
the
Signet
work
failures
are
related
to
endpoint
slice,
they're,
not
related
to
anything
else.
A
This
enhancement
owner
has
run
up
his
own
test
infrastructure
and
he's
running
the
e2e
tests
off
to
the
side.
At
the
moment,
and
he's
pretty
confident
that
this
will
fix
it,
so
we're
just
waiting
on
a
LG
TM.
If
anybody
can
get
him
that
it's
an
owner
before
this
afternoon,
he's
waiting
for
Bowie
to
to
get
him
LG
TM.
A
A
I've
asked
him
to
preemptively:
do
that
we're
trying
to
just
basically
work
through
this
last
issue.
He's
he
thinks
the
feature
is
stable.
The
problem
is
the
test,
doesn't
turn
on
the
feature
cropper
correctly
and
that's
because
he
needs
a
flag
in
a
bunch
of
components
and
the
way
the
Alpha.
Today's
are
set
up,
it's
not
being
propagated
correctly.
So
if
you
go
and
look
in
that,
PR
he's
putting
a
whole
bunch
of
bash
foo
to
get
things
set
correctly.
A
A
C
If
we
get,
this
resolved,
I
think
we'll
be
feeling
pretty
good.
It
is
somewhat
alarming
the
amount
of
you
know,
testing
failures
or
things
that
were
kind
of
like
yeah.
That's
not
really
released
block
you're
informing
that
probably
could
be
improved,
but
other
than
that,
knowing
that
some
of
these
issues
are
not
too.
Concerning
for
the
overall
release,
I
feel
good
I'd
like
to
kick
it
over
to
George.
If
is
any
further
comments
on
that.
A
Yeah
and
I
just
want
to
say
thanks
to
George
for
teeing
up
the
conversation
with
six
scalability
I
think
it
was
very
fruitful
to
at
least
because
the
last
two
releases
we've
had
some
late-breaking
blockings
called
out.
So
things
seem
to
be
in
good
shape,
we'll
keep
continuing
to
watch
those
scalability
issues
and
master
informing,
obviously,
but
they
didn't
have
any
concerns
and
even
though
they
called
out
that
regression
it's
a
longer
running
regression,
that's
non-blocking
is
what
they
said
so
excellent.
D
Everyone
so
I
start
as
usual
with
some
numbers,
but
first
our
stateís
is
bright.
Green
everything
seems
to
go
very
well
regarded
number
of
open
issues
for
the
Evo
six
in
milestone
here
at
Phi,
which
is
eight
less
than
yesterday.
We
don't
have
any
critical
new
tradition
in
regarding
importa.
Soon
we
have
two
of
them,
which
is
seven
less
then
yesterday
and
we
had
a
very
good
response
rate
from
contributors.
Another
six
each
chart
with
a
lot
to
bring
this
to
bring
this
number
down
right
away.
D
Regarding
PRS,
we
have
four
of
them
open,
and
this
is
6
less
than
were
sweetened
with
out
any
critic
around
and
with
free
importer
service,
which
is
6
less
than
he
had
device
fatigue
and,
last
but
not
least,
is
16
open,
PRS
with
equal
extension
phrase,
which
is
12
less
than
yet
before.
So
we
don't
have
any
upcoming
with
milestones
and
other
stuff.
We
will
continue
monitoring
the
stats.
They
see
how
we
can
help
in
other
ways.
So
this
is
it
from
our
side.
For
now,
do
we
have
any
questions.
A
E
Hello,
everyone
so
for
today
we
are
yellow
and
the
reason
why
I
picked
yellow
is
because
we
still
have
like
in
a
number
of
PRS
to
get
in
before
deadline
on
Monday
and
I
mean
we.
We
are
making
progress
on
them,
but
we
have
a
number
of
PRI
just
like
a
code.
Modification
and
I
should
be
able
to
merge
them
and,
and
also
we've
been
getting
like
a
lot
of
elk
from
the
bird
dog
team,
and
the
shadows
have
been
very
very
like
alpha.
E
In
terms
of
you
know,
working
weeks
you
know
dogs,
ona
and
to
make
sure
the
athletes
review
and
reply.
The
biggest
issue
that
we
have
is
you
know
when
e
time
someone
leaves
a
review.
Getting
you
know
the
the
owner,
the
PR
owner
to
respond
is
always
the
biggest
issue,
but
from
what
I
have
seen
so
far,
which
not
to
them
directly
on
slack
since
two
out
and
once
I
once
loved
the
dogs
once
come
up,
the
minor
things
are
resolved.
E
A
E
The
impact
is,
you
know
we
would
ask
without
the
dots
right,
I
think
it's
something
I
will
be
proudly
axiom.
Everyone
waits
to
get
out
and
released
him
to
decide.
You
know.
Are
we
going
to?
You
know
make
these
he
add
requirement
that
because
of
the
dog
is
not
going
in
gather
to
drop
this
or
we
are
going
to
you
know,
get
the
UNA
online
and
say
we
need
you
to
resolve
this
right
now.
Okay,
yeah
I,
think
it's
something
that
we
have
to
decide
to
get
allottee
my
old
piece.
E
We
don't
get
to
that
point,
that's
my
hope
and
boy.
If
we
do
get
it
up
when
we
have
to
decide
as
a
team
and
I'm
going
to
reach
out
to
in
a
couple
of
you
know
previous
leaks
and
an
axe,
then
how
what
kind
of
reaction
actually
took
place?
I'm
going
to
do
that
today,
you
just
introduce
the
prepare
for
that
kind
of
situation.
E
A
Yeah
I
think
prepare
prepare
yourselves,
probably
the
best
thing
so
that
you
can
bring
what
you
think
the
decisions
are
we
can
make
as
a
team,
because,
obviously
you
get
to
make
the
call
and
we
can
support
you
in
that
call.
I
would
be
interested
just
to
see
out
of
my
own
interest.
What
are
the
16
PRS?
Can
you
just
throw
a
link
in
there
at
least
see
the
the
list
of
them.
E
F
Happy
Friday
everybody
we're
looking
green
over
in
release,
notes,
no
major
updates.
Since
yesterday
we
did
have
a
kind
of
lingering
PR
that
we
have
got
fast
forwarded
with
the
most
recent
updates
and
we're
just
waiting
for
review
to
make
sure
that's,
okay,
I
think,
as
Steven
was
tagged
in
that
as
well
as
we
have,
there
should
be
a
PR
coming
forward
down
there
being
proposed
a
little
bit,
that's
going
to
propose
some
changes
that
are
won't,
add
any
extra
labels
to
prowl
and
and
the
like.
So
that's
right,
we're
at
right
now
any
questions.
G
G
A
Oh
excellent
emeritus
lead
no
updates,
that's
still
the
case,
Josh,
okay,
excellent.
Thank
you
very
much,
which
brings
release
lead,
update,
I'm,
gonna
call
the
status
read
until
we
get
green
on
CI
signal.
I,
don't
feel
I
can
give
a
yellow
or
a
green.
So
thanks
for
the
tireless
work
there,
there
are
a
lot
of
green
in
other
places,
so
just
getting
CI
signal
into
line
and
then
I'll
feel
very
confident.
A
Just
a
reminder.
Monday
Doc's
need
to
be
completed.
Unders
just
mention
that.
So,
if
you
are
a
dog
owner,
make
sure
you
get
them
reviewed
and
merged
in
before
then
RC
one
is
next
Tuesday
and
Code
Thor.
So
we
are
going
to
have
an
interesting
discussion
on
Monday
about
how
things
are
looking
and
make
sure
that
we're
all
all
systems
go
for
RC,
one,
so
I'm
hoping
we
come
in
Monday
with
some
green
on
CI
signal
and
we're
looking
good
and
Hyundai
gives
us
green
on
the
in
the
docs
as
well.
A
So
I
think
things
are
shaping
up
to
be
in
line
and
in
shape.
Finally,
release
117
nominations:
I
put
the
link
there
in
the
agenda
and
I
would
like
to
say
that
I've
nominated
Gwyn
as
the
release
lead
for
117.
So
congratulations,
Gwyn
I've,
put
that
on
the
issue.
Guin
has
done
a
fantastic
job
as
being
released,
lead
shadow
and
116.
So
I
think
she
will
do
a
tremendous
job
of
being
the
lead
in
117.
So
I
wish
her
the
very
best
and
I'm
very
excited
to
have
her
serve
in
that
role
in
117.
F
A
A
Okay,
I'm
not
going
to
walk
the
project
board
today
because
it
feels
a
little
bit
too
disciplined
to
walk
it
every
day
and
say:
what's
your
update?
What's
your
update,
but
we
are
in
burned
down.
I
will
do
it
on
Monday
and
see
if
we're
making
at
least
progress
on
the
things
that
are
are
in
progress.
I
wanted
to
quickly
just
go
through
I
will
just
pop
the
hood
on
share
my
screen.
A
This
is
also
in
the
notes,
but
these
are
the.
These
are
the
four
open
issues
that
need
to
be
merged,
so
I'm
just
going
to
quickly
walk
through
them.
If
they
don't
have
this
one
has
approved
in
lg
TM,
so
I
assume
that
will
once
the
tests,
the
pre
submits
happen,
that's
going
to
go
into
the
Q.
This
is
that
issue.
I
mentioned
that
should
fix
up
the
logic
for
the
tests
looks
like
it's
got.
Okay,
it's
got
LG
TM,
but
not
an
approval.
At
the
moment.
A
A
I
will
ping
Steven
on
it
is
assignees
he
is
not
assigned.
Do
we
have
any,
maybe
I'll
ping
dims
to
see
if
we
can't
get
it
in
the
right
check.
I've
seen
him
around
the
ridges
this
morning,
I
will
take
action
on
get
a
pinging
dims
and
see
if
he
can
provide
an
update
there.
But
it
looks
like
three
of
the
four
will
go
in
two
of
the
four
are
ready
to
go
in
this
one
needs
approval
I.
Just
this
is
Rob
the
one
that
I
mentioned
with
Rob,
so
to
look
at
that
quickly.