►
From YouTube: Kubernetes SIG Testing - 2019-06-11
Description
A
A
A
B
F
A
B
E
B
A
B
So
I
think
this
is
mainly
you
know.
Occasionally
we
get
new
people
and
so,
like
somebody,
Shawn
recently
started
on
our
team
and
he
looked
for
you
know
some
good
first
issues
to
work
on
and
found
one
and
did
it
and
so
now
I
think
it's
down
to
zero,
and
so
Aaron
was
sort
of
pointing
out
that
it
would
be
nice
if
there
was
more
than
zero
issues
to
point
potentially
interested
contributors
to
and
and
I
certainly
found.
You
know
not
100%
success
rate,
but
a
fairly
good
success
rate.
B
That's
you
know
a
fairly
good
return
on
my
investment
in
filing
good
first
issues
it
having
them
actually
get
solved
rather
than
me
solving
them,
and
so
I
think
you
know
for
the
more
so
I
yeah
I
think
that
would
be.
You
know,
I
think
it
would
be
in
our
best
interest.
You
know
maybe
next
release
or
in
the
coming
months
to
allocate.
You
know
a
more
consistent
amount
of
time
into
adding
those
issues
to
the
the
repo
and
to
see
if
we
can
get
some.
B
You
know
more
contributions
that
way,
rather
than
you
know,
I
feel,
like.
Maybe
our
group
has
a
little
bit
of
a
bias
towards
you
know,
doing
everything
ourselves
and
that's
obviously
cool
and
super
important,
but
I
think
it's
also
useful
to
sort
of
give
opportunities
for
other
people
who,
like
Prower
or
infrastructure,
to
help
out
and
then
on
the
roadmap.
I
know
you
know,
Steve,
you
have
probably
done
the
most
work
you
and
Aaron
I
guess
have
done
the
most
work
at
having
this
actually
haven't.
B
Read
that
but-
and
so
maybe
sort
of
I-
think
that
maybe
the
rest
of
us
could
get
more
involved
in
you
know
buying
into
that
or
whatever
participate
in
on
it,
and
then
also
you
know
converting
that
into
something
a
little
bit
more
visible
on
github
like
trying
out
using
milestones.
Maybe
we
could
ask
Ben
to
sort
of
I
think
Ben
uses
it
for
kind,
it's
sort
of.
Maybe
he
would
have
some
interesting.
You
know
lessons
to
sort
of
whatever
is
things
to
help
us
figure
out
how
to
use
that
effectively
we're.
D
B
Think
on
the
larger
contributions
you
know,
I
know
I,
yeah
I,
don't
know
if
is
and
I
want
to
see,
but
I
think
that
would
be
I
mean
so
Tim
and
Ben.
Have
you
been
to
find
that
so
you
do
find
that
helps
people
like
in
China
a
contributor.
You
know
what
is
your
your
feedback?
Is
it's
worthwhile
to
invest
in
it
I
think.
A
Like
some
of
the
earlier
work
about
sort
of
try
to
define
some
larger
milestones
for
the
crowd,
like
I,
think
I
timed
that
really
poorly
and
it
overlapped
with
internal
planning,
you
guys
are
doing
installations
it
just
it
just
became
a
real
hassle,
so
maybe
I
don't
know
to
what
extent
we
want
to
try
to
tackle
this
before
the
next.
Like
large
round
of
planning
sessions,
yeah
I,
guess
I
would
I
would
echo
what
vet
said
that
it's
I
think
it's
critical
for
us
to
actually
capture
like
where.
B
A
Yeah
I
think
the
epics
that
I
outlined
and
then
there's
a
couple
that,
like
think
cult
with
a
run
I'll
go
put
it
one
and
John
put
in
one
into
the
the
older
prowl
epics
document
like
I,
guess
my
vision
for
that
was.
We
would
hopefully
get
input
from
a
lot
of
the
the
current
contributors
and
then
how
a
conversation,
what
we
try
to
distill
down,
which
are
more
useful
in
the
short
term.
D
Think
the
key
is
to
be
aggressive.
Do
you
two
things
one
make
sure
that
your
issues
are
sort
of
groomed
for
a
milestone?
That's
the
obvious
one!
So
that
way,
it's
clear
to
external
observers,
but
the
I
issued
consistency
being
able
to
actually
have
anatomic
work
item
that
people
can
take
a
look
at
and
and
be
able
to
execute
on.
D
A
One
of
the
things
that
might
are
like
I
know
for
me
is
that
oh,
the
frustrating
is,
there's
a
lot
of
organizational
pressure
to
use
like
a
specific
bug,
tracker
or
series
of
bug
trackers.
That
does
not
include
them.
You
have
issues.
So
to
what
extent
do
you
find
that
you're
like
duplicating
that
work
internally
for
your
company
as
well
as
in
this?
We.
D
Have
strong
opinions
like
so
I
worked
her
head
a
long
time
and
I
did
bugzilla
Trello
and
github,
and
more
github
and
I
said
never
again
what
he
left
so
yeah,
so
like
the
great
epics,
graphics
and
user
stories
in
whatever
tool
you
want
to
and
then
break
down
that
and
link
to
like
some
project
or
an
epic
board
somewhere
else.
So
that
way,
nothing
should
be
one
for
one
mapping
back
and
forth,
because
that's
just
a
waste
of
time.
Yeah.
A
Yeah
I
think
I'm
I'm.
You
know
I'm
definitely
excited
and
happy
to
continue
to
do.
Work
like
if
we
I
can
try
to
take
some
of
the
take
the
the
things
that
you
have
contributed
to
the
epics
talk
and
put
them
into
milestones
and
maybe
lay
out
a
couple
of
issues.
I
think
some
of
the
ones
that
I
put
in
there
I
have
formatted
with
a
couple
of
like
you,
know
some
sub
projects
or
subtasks,
and
then
that
sort
of
gets
us
like
a
little
bit
along
the
way.
B
D
Got
a
Jim,
sequentially,
okay,
it
just
makes
it
makes
it
easier.
If
you'll
understand
we
don't
even
try
to
like
prognosticator
the
future.
We
have
like
a
current
milestone.
Then
we
have
a
bucket
called
next,
because
we
it
we've
done
a
lot
of
work
in
the
past.
We
said
what
we're
gonna
do
this
in
the
release
after
next
and
that
we
were
basically
just
lying
to
ourselves
right
right
and
so
what
we
do
is
we
just
have
a
next
bucket,
a
current
bucket,
which
is
the
current
milestone
and
in
the
once
you
finish
a
milestone.
D
C
D
For
the
most
part,
it's
works,
I'm
sure
we
could
probably
figure
out
a
way
that
might
be
better.
It
might
be
good
to
have
like
a
broader
survey
from
the
wild
like
organized
from
contributes
or
something
that
says.
Like
you
know,
we
found
that
nine
out
of
ten
new
contributors
find
that
X
Y
Z
is
better
yep
I.
A
Guess
the
biggest
thing
for
me,
unlike
that
approach
for
some
of
the
work
that
we're
doing
it
tested
for
it's
like
hard
to
nail
down
I
mean
some
of
it-
is
like
directly
tied
to
a
release,
but
I
mean
historically
figuring
out
where
pry
was
moving
in
the
next
three
months
is
good.
That's
a
very
intentional
problem.
A
D
A
D
Of
a
lot
of
times
will
put
thing
in
the
milestone
and
we'll
say:
Help
Wanted
you're
like
we'll
get
to
it.
We
want
to
get
to
this,
but
there's
no
guarantee
if
no
one's
committed
to
it
right.
So
so
it's
the
priority
on
the
priority
chain
is
lower,
typically
down
the
list.
Anything
that's
p0
p1
is
usually
has
somebody
committed
to
it
and
then
things
that
are
further
down
the
list.
We
want
to
get
to
it.
If
you
have
time,
but
no
one's
actually
committed
to
it.
A
So
I
guess:
how
do
we
I
mean?
How
do
we
want
to
structure
that,
like
my
thumb
with
the
uptick
stock
previously
was
to
sort
of
get
ideas
on
where
people
think
the
most
value
would
go
for
the
project
for
things
to
focus
on
next
and
then,
once
we
had
a
sort
of
you
know
general
agreement
on
some
of
the
big
picture
items
we
could
then
see
who
wanted
to
trade
somewhat
Eric
when
you
said
that
this
would
be
a
good
time
like
within
the
next
month
to
start
thinking
about
the
next
planning
cycle.
A
A
B
You
know
it
felt
like
we
had
already
had
a
plan,
and
so
it
wasn't
I
feel
like
it
would
be
more
useful
to
in
this
next
release.
Come
up
with
a
you
know,
open
source
plan
that
we
are
all
participating
in
before
we
have
already
ossified
Google's
plans.
Yeah
so
I
feel
like
now
will
be
a
good
time
to
try
and
come
up
with
the
open
source
plan
and
use
that
to
influence
our
internal
planning.
I.
B
B
Yeah
I'm
I
mean
I,
don't
know
about
you,
but
I
sort
of
feel
something
I
mean
as
long
as
it's
not
like.
If
there's
you
know,
if
there
are
two
things
to
do,
if
there
are
you
know,
I
would
tend
to
be
somewhat
biased
towards
as
long
as
what
people
are
working
on.
Is
it
in
conflict
with
something
that
we
want
to
do
you
know
sort
of
I
feel
like
in
some
ways
you
know
the
prior.
B
A
And
potentially
like
that,
planning
session
could
be
a
place
for
somebody
to
convince
somebody
else
that
XYZ
isn't
yeah
yeah
yeah
like
I'm
happy
to
drop
whatever
I've
got
planned.
If,
if
something
else
is
someone's
got
my
dealing
that
you
know
and
so
I
think
yeah.
If
we,
because
if
we
don't
have
that
sort
of
cross-pollination
of
ideas,
then
basically
we're
all
taking
to
the
table.
What
we
want
to
work
on,
you
know
running
it
through
whatever
process
and
then
putting
it
until
I
started.
That
makes
it
more
open,
but
I'm
not
sure
they.