►
From YouTube: Kubernetes SIG Service Catalog 20170109
Description
SIG Service catalog touches base for the first time in 2017 and summarizes progress towards their Legos-1 milestone.
Agenda - https://docs.google.com/document/d/10VsJjstYfnqeQKCgXGgI43kQWnWFSx8JTH7wFh8CmPA/edit
A
A
A
A
Okay,
okay,
so
first
thing
proposals,
CFP
responses
submitted
to
cube
con
in
total
for
the
Service
Catalog
space.
There
were
three
proposals
submitted
as
we
have
discussed
in
previous
cig
meetings.
One
was
a
talk
with
myself
in
chip
Childers
from
cloud
foundry,
which
sort
of
the
general
overview
of
open
service
broker,
API
and
Coover
Nettie's
Service
Catalog,
and
then
there
were
two
panels.
A
One
was
with
folks
from
the
cig,
this
cig
and
the
other
was
folks
that
are
committers
on
the
open
service
broker
API,
so
I
am
fairly
confident
that
we
will
get
at
least
one
of
those
in
I
think
that
we
are
supposed
to
hear
back
next
week,
but
in
the
past
q
con
cfp
responses
have
slipped
the.
The
pattern
has
been
like.
A
They
will
say
that
they
will
get
back
to
you
on
date,
X
and
then
they
don't
do
that
a
week
passes
and
then
they
say
they
need
another
week
after
that,
so
maybe
by
the
end
of
January,
we'll
know
for
sure
exactly
which
submissions
were
accepted
in
which
weren't.
But
that
is
the
story
on
it.
So
next
up
is
Aaron,
with
progress
being
made
on
at
F
planning.
D
D
Of
course,
like
last
time,
you'll
have
opportunity
at
the
beginning
of
the
face
of
face
to
say
what
you
want
to
talk
about
as
well,
and
you'll
obviously
helped
shape
that,
but
look
out
for
that
email
on
wednesday
for
those
who
are
coming
to
the
face
to
face,
if
you
don't
know
when
or
where
it
is,
please
ping
me
on
slack
I
will
be
happy
to
tell
you,
and
that
is
about
it
for
now,
so
I
will
turn
it
back
over
to
you.
Paul,
okay,.
A
A
We
have
a
few
poles
that
together
get
us
a
much
of
the
way
towards
the
things
that
we
talked
about
in
Legos
one
and
just
to
to
rehash
what
those
things
are.
Basically,
we
want
to
have
a
working
API
server
that
that
serves
our
V
1
alpha
1
version
of
the
Cooper
Nettie's
Service
Catalog
API,
and
this
this
is
really
sort
of
like
the
linchpin
for
everything
else
that
we
need
to
build
to
get
to
an
MVP
state,
because
that
is
how
all
the
components
and
Cooper
Nettie's
actually
access
the
API.
A
A
Then
a
pull
request
that
that
adds
some
API
registration
and
adds
the
skeleton
of
the
pieces
that
we
need
to
actually
persist
resource
instances
back
into
at
CD
or
into
exedy
and
get
them
back
out,
which
was
submitted
by
myself.
But
is
the
synthesis
of
work
that
Morgan
Bauer
Aaron
at
dais
and
myself
had
done?
And
then
there
is
a
pull
request
from
Jeff
Keeler
at
Red
Hat.
A
That
finishes
the
code
generation
for
the
conversions
fixes,
an
issue
with
code
generation
for
defaults
and
adds
a
test
that
that
does
buzzing
and
round
tripping
of
API
objects
in
our
API
through
all
that
stuff
to
prove
the
proof
that
it
works.
So
that's
that's
big
medicine
because,
after
after
these
are
merged,
I
think
we'll
be
in
a
state
where
it
will
be.
A
There's
there's
not
a
lot
of
work
left
to
actually
cook
the
API
server
up
to
etsy
d
and
be
able
to
write
some
behavioral
tests
and
use
keep
CTL
for
like
raw
crud
operations
against
the
API
server.
So
it
it's
great
to
be
where
we
are
and
I'll
be,
even
happier
in
like
two
or
three
days.
Hopefully,
when
all
that
stuff
is
merged,
any
questions
on
that
before
we
we
move
on.
D
E
Sure
so
so
we
have
a
lot
more
people
working
on
the
repo
now
and
a
lot
of
big
PRS
going
through,
and
it's
resulted
in
broken
functionality
or
broken
building
or
broken
various
things
we
frequently,
probably
more
often
than
things
are
working
and
the
main
cause
of
that
is
that
we
don't
have
it
in
nesting,
either
really
for
the
development
environment
or
for
the
github
PR
gating.
E
I
think
wednesday
of
this
week
to
to
discuss
that
and
finalize
a
strategy
and
then
and
then
we
can
start
implementing
it
right
now
we
have
Travis
which
does
unit
tests
and
we
have
Jenkins
that
is
not
plugged
in
and
and
maybe
has
some
final
bits
that
need
to
be
taken
care
of
before
it
works
for
our
scenario,
so
that
that
could
be
used
for
gating
PRS
through
github.
But
we
just
need
to
come
up
with
a
strategy
for
for
dev
testing,
as
well
as
give
er
testing.
A
A
A
C
The
only
reason
I
focus
on
hyper
cube
is
because,
if
I
understanding
of
mini
cube
is
that
it
stands
up
its
own
virtual
machine
to
run
this
stuff
and
I
I'm
already
running
inside
a
virtual
machine
when
I
do
my
stuff.
But
that's
why
I
like
hyper
cube,
because
it
just
they'll,
use
my
current
soccer
or
set
up
and
just
break
up
containers
to
do
everything.
This.
F
Actually
yeah
Doug
I
can
I
can
follow
up
with
you
all
that
you
want,
but
there
is
a
version
that
doesn't
set
up
a
virtual
machine.
Oh
cool,
okay,
happy
good,
then,
and.
A
E
If
anybody
doesn't
want
to
do
testing,
please,
please
stand
up
now
and
and
then
present
to
your
left.
Yes,
if
I
had.
A
G
No
I
just
wanted
to
you
know:
we've
got
sort
of
these
pr's
that
are
sort
of
yes.
This
depends
on
that.
The
next
one
you
go
to
it.
Oh
this
one
depends
on
that
one,
the
next
one.
Yet
so,
just
what's
the.
G
Constant
evolution,
progress,
kind
of
things
that
were
you
know,
I've,
always
always
making
movement.
You
know,
I,
don't
want
to
be
just
committing
things
to
get
things
committed,
but
what
what
the
thought
process
on
getting
to
16
in,
if,
if
it's
ready
or
if
it's
not
ready
or
what's
the.
A
A
Okay
super,
so
it
sounds
like
we'll
be
able
to
get
that
one
in
today.
If
I'm
hearing
you
correctly
yeah
so
after
that
I
think
I
will
need
to
rebase
to
16
on
master
right,
which
is
cool,
because
I
am
one
of
the
people
that
actually
enjoys
rebasing
code
and
then
I
think
235,
which
is
jeff's
full
request.
If
I
remember
correctly,
I
should
be
actually
orthogonal
to
most
of
those
changes,
so
Jeff
will
need
to
rebase,
but
it
shouldn't
be
that
much
work
like
you,
Jeff
I.
G
Might
have
the
wrong
number
in
there.
I
was
just
I
had
a
tiny
one
that
was
based
on
yours
for
the
sum
of
the
strategy
stuff
to
go
through
it,
and
then
there
was
a
further
question.
I
had
basically
of
what
are
people
sort
of
opinions
on
making
a
PR
to
basically
discuss
things,
and
if
we
don't
want
to
make
a
PR
to
discuss
things,
what's
the
appropriate
venue
such
that
a
it
gets
publicly
exposed
and
B
I
don't
go
around
annoying
people.
I
ask
hey,
we
think
of
this.
What's.
G
A
All
right,
I
think
making
a
PR
is
fine.
It's
easy!
It's
it's
easiest,
I
think
to
to
do
a
review
on
code
than
it
is
to
a
bike
shed
about
English
words,
I.
A
G
Right
now,
that's
you
know,
so
that's
sort
of
a
discussion
PR
and
the
fact
that,
oh
here's
an
example,
it's
basically
just
the
stubs
and
now
I
need
to
figure
out
what
to
somewhat
put
in
some
of
the
things
to
to
know
it,
but
but
and
as
sort
of
a
forward
progress
forward.
Movement
I
just
want
to
get
things
out
there.
Sometimes,
yes,.
D
And
Morgan
I
feel
free
a
to
bug
people
if
you
want
a
preliminary
review
or
just
whatever
I
feedback
before
the
code
is
done.
Okay,
and
I
would
put
in
the
pr
description
to
remind
people
if
they
should
not
be
doing
a
complete
review,
but
just
review
specific
section
or
you
know
the
overall
design
or
architecture
whatever
yeah.
G
A
B
A
Face-To-Face
before
or
after
long
I
think
that
would
be
awesome.
So
it
looks
like
I
am
going
to
be
in
berlin
for
a
openshift
commons
event.
That's
collocated
with
cuba
con,
but
is
the
day
before
I
expect
that
I'll
probably
be
there
for
like
another
few
days,
so
I
would
potentially
be
available
to
do
something
like
that.
Maybe
that's
something
we
should
take
to
the
mailing
list
and
that
policies
coronal
suture,
that
it
could
be
a
great
idea.