►
From YouTube: CNCF Kubernetes Conformance WG - 2018-04-12
Description
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
B
C
B
They
appreciate
the
latest
release
that
has
the
clear
message
at
the
end
about
whether
you're
passing
all
the
tests
or
not
the
essential
processes
are
working
out.
I've
made
a
few
small
updates
in
the
FAQ
to
just
address
questions
that
people
have
asked.
I'm,
also
very
pleased
to
mention
that
Amazon
Web
Services
has
certified
in
the
last
week
or
so,
which
is
essentially
the
last
major
vendor
in
the
space
that
it
hasn't
a
source
they
could
promise
to
do
before
they
went
GA
they've
gone
ahead
and
done
so.
B
From
my
perspective,
the
the
working
group
is
going
well
and
all
those
processes
are
are
underway.
The
kind
of
new
areas
that
we're
beginning
to
invest
in
is
is
starting
to
try
and
improve
some
of
the
test
development.
But,
and
so
we
do
have
this
investment
going
forward
globin,
but
that
hasn't
resulted
in
poor
request.
Yet
and
of
course,
all
those
processes
happen
through
Stig
testing
and
Stig
architecture,
not
over
here
so
could
I
answer
questions
for
anyone
about
the
process
or
issues
folks
have
run
into
or
concerns
or
anything
else.
D
E
E
Sure
so,
we've
been
mostly
understanding
the
whole
project
and
how
the
tests
are
built
and
how
they
are
written
down
and
going
through
all
the
concepts
with
my
young,
and
we
have
this
spreadsheet-
that
Mitra
socialized
with
us
from
the
different
API
endpoints
that
we
should
tackle
first.
So
we've
divided
to
begin
with
with
my
young
bank
is
gonna
tackle
a
bunch
of
namespaces
in
points
like
section
of
it
and
I
will
talk.
E
E
Of
how
this
works,
so
our
step
right
now
is
to
mimic
what
is
there
and
reproduce
that
on
REST
API
like
very
basic
REST
API
testing
like
if
the
initial
status
is
empty,
then,
if
I
ask
for
namespaces,
then
I
should
get
only
a
default
ones.
If
I
have
one
I
see
that
if
I
did
then
I
should
not
see
any
more
and
things
like
that,
so
that
should
give
us
a
lot
of
very
of
clarity
first
to
actually
submit
peers
and
go
through
a
process
and
start
from
the
things
we
know.
B
C
Yeah,
it
sounds
great
thanks
for
the
introduction
nice
to
meet.
You
all.
Thank
you,
so
he
studies
the
plan
to
use
this
meeting
as
kind
of
a
regular
check-in.
Then
the
work
that
you
said
you're
all
doing.
D
E
B
D
Is
kind
of
a
little
bit
more
details
on
what
should
be
a
conformance
test
or
how
we
are
measuring
coverage
to.
We
have
some
touch
point
level
coverage
measurements
added
interesting
for
a
hippie
hacker
I
would
like
us
to
maybe
come
together
as
a
group
to
kind
of
discuss
and
figure
out
what
we're
gonna
do
and
our
next
steps
for
measuring
that
coverage.
So.
C
B
Not
this
working
group
and
so
I
think
it's
totally
fine
for
us
to
discuss
it
or
talk
about
process
because
talks,
but
I
just
want
to
be
clear
that
we're
not
trying
to
take
that
on,
but
I
think
in
particular,
looking
at
how
we're
measuring
and
whether
there's
better
ways
to
measure
what's
really
in
use
and
and
what
should
be
done.
That's
and
such
would
be
immensely
valuable
for
us
to
reach
consensus.
B
D
But
I
can
touch
base
with
a
couple
of
initiatives
from
my
team
to
help
with
conformance
one
is
the
darker
and
darker
based
cube
test
solution
that
can
be
used
to
run
conformance
tests
and
Quinten
on.
My
team
is
basically
working
on
setting
up
a
CI
with
those
kinds
of
conformance
tests,
and
we
want
to
see
how
good
or
how
performant
it
is.
D
He
already
has
a
non-blocking
CI
set
up,
so
I
should
have
more
data
on
that
as
well,
and
I'm
gonna
try
and
ask
the
vendors
to
also
write
a
doctor
in
docker
based
test
to
see
how
comfortable
they
feel
or
how
easy
or
harder
it
is.
There
is
also
some
efforts
with
a
couple
of
people
going
on
around,
like
a
kind
of
like
figuring
out
the
best
framework
to
use
for
a
conformance
test.
They
will
circulate
a
proposal
at
some
point.
I
don't
want
to
step
on
their
work.
D
Just
now
and
Bend
from
my
team
has
been
also
working
with
a
lot
of
the
cloud
providers
to
be
able
to
export
their
confirm
and
stress
test
results
to
discrete,
and
he
has
a
working
dashboard
already
and
a
proposal
that
was
kind
of
circulated
in
the
six
and
I
think
one
or
two
cloud
providers.
Data
is
already
there
so.
B
B
The
surface
area
that
we
do
in
a
in
a
more
efficient
or
more
automated
way,
I
think
would
all
be
would
be
fascinating
areas.
I
mean
I.
Think
that
the
fact
ideally
see
eight
members
like
like
Google,
could
find
a
lot
of
this
work
and
there's
also
potentially
some
funding
available.
If
we
wanted
to
have
contractors
work
on
things,
but
even
just
trying
to
understand
the
landscape
before
we
go
to
the
next
step
would
be
really
helpful.
Sounds.
C
C
Last
time
we
talked
about
profiles,
it
seems
like
this
tool
not
too
much,
there's
no
kind
of
like
driving
need
for
that.
Yet
so
I
guess
we'll
probably
just
table
that
for
now
maybe
just
focus
on
the
quality
and
the
Tesco
region
at.
A
Some
point:
we
should
talk
about
it,
because
the
the
consistency
of
storage
of
cross
providers
is
a
thing
that
people
you
know.
Storage
and
networking
in
other
aspects
will
eventually
want
to
cover
that
I.
Don't
know
when.
Obviously
the
I
think
the
conformists
work
alone
will
probably
take
a
couple
of
cycles.
Several
cycles
we're
nice
to
get
like
good
coverage.
A
C
C
C
All
right:
well,
we
do
have
a
meeting
scheduled
in
two
weeks,
but
then
there's
the
cabinet
in
three
weeks,
so
I'll
probably
suggest
we
we
just
meet
in
three
weeks
for
Copenhagen.
Does
that
sound
good
everybody
that'll.