►
From YouTube: Kubernetes Community Meeting 20160505
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
Discussion of Community Committed Feature tracking; SIG-Clusterops to start making reference architectures; SIG-OpenStack report from OpenStack Summit ATX; SIG-Node Feature summary; Notices
A
As
I
do
every
week,
I
want
to
mention
that
we
are
recording
this.
This
is
the
Cooper
Nettie's
community
meeting
for
May.
Fifth,
it
is
also
the
grand
asia
time
zone
friendly
experiment.
So
we
will
see
how
many
people
we
end
up
bringing
in
for
for
this
meeting
to
see.
If
we
want
to
continue
to
do
this,
as
we
were
looking
at
once
a
month
to
do
I,
Asia
friendly
times
on
meetings
that
makes
it
hard
for
some
of
our
European
friends
and
west
or
east
coast,
us
friends,
alright.
A
So
one
more
quick
call
up
for
andreas
to
see.
If
he
is
here
all
right,
we
may
not
have
a
demo.
Does
anybody
have
something
that
they
wanted
to
demo
quickly
is
a
starting
point
for
today,
because
our
scheduled
demo
or
doesn't
seem
to
be
with
us
yet
I
usually
use
this
first
10
minutes
as
a
demo
space
in
order
to
let
people
trickle
in,
if
they're,
here
a
little
bit
late,
all
right
well,
not
today.
A
The
big
agenda
item,
then
that
I
have
for
today
is
talking
about
the
community
committed
features
for
1.3
and
the
fact
that
we
have
been
tracking
and
reporting
on.
We
TJ
Gulf
Gutterman
from
google
have
been
tracking
and
reporting
on.
The
four
big
features
that
we,
the
Google
team,
is
working
on
and
we
haven't
had
a
good
insight
into
what
broader
community
is
working
and
I
would
like
to
introduce
Mike
Rowe,
who
just
turned
on
this
video
and
is
the
new
VP
of
engineering.
Yes,
at
core
OS
yeah.
A
The
new
VP
of
engineering,
a
core
OS
and
who
has
graciously
volunteered
to
dig
in
and
investigate
what
1.3
is
going
to
look
like
as
a
what
trial
by
fire
to
get
started
there
at
core
OS
and
give
us
some
dashboard
possible
views
into
the
community.
Work
is
being
done
so
let
Mike
show
us
what
he's
been
working
on
and
start
getting
feedback
from
the
rest
of
you
around.
What
we
need
to
fill
that
in
yes,.
B
Hi
everyone
so
I,
don't
know,
maybe
I'm
just
new
to
the
projects
I
I
I've
been
in
Korea
for
a
couple
of
months
now
and
kind
of
my
observation
was
that
there
it's
a
big
project
right.
Was
there
a
lot
of
documents,
Google,
Docs
and
git
hub
vika
documentation
flying
around,
but
it's
very
hard
to
understand
kind
of.
What's
eventually
going
to
go
in
one
two,
three
or
not
I
kind
of
lucan
sources,
dogs,
some
of
them
mentioned
Fisher's
at
p0
or
p
1,
and
then
according
to
get
up
proposals,
the
priority
has
been
changed.
B
C
B
Me-
and
this
is
the
example
of
the
doc
doc
was
shared-
is
me
right,
kind
of
control,
plane,
external
and
turns
out
a
lot
of
things
which
you
are
p
0
accordin
to
get
carbon
a
longer,
20
and
so
I
started
compiling
the
dashboard
like
that.
Let
me
know
if
it's
too
small
it
to
like
and
kind
of
I
want
an
opinion
from
everyone,
especially
from
the
leads
from
cigs,
if
it's
kind
of
useful
and
if
they
can
help
me
maintain
in
their
parts
of
the
dashboard
right.
B
Basically,
the
idea
of
the
dashboard
to
monitor
which
see
it
belongs
to
so
it's
easy
to
filter
out
right.
So,
if
I
missing
some
I'm
happy
to
add
more
then
describes
the
item,
the
link
to
proposal,
or
in
some
cases,
use
it
to
make
kind
of
it's
a
big
item.
Right
is
the
status
of
all
the
items,
but
I
already
p
want
it
to
be
0.
Ideally,
if
I
can
track
the
github
of
dev,
lift
or
dev
leads
approver
because
kind
of
our
experience.
It
is
from
Korres
site,
sometimes
it's
hard
to
track
down.
B
B
A
My
kaftan
are
chad
is
there?
Is
this
only
for
sig
related
efforts
are
for
all
major
major
items
that
are
to
land
in
13.
We
would
love
it
for
to
be
for
all
major
items
that
are
to
land
in
13.
The
this
cigs
are
where
a
lot
of
that
effort
and
a
lot
of
the
documentation
has
bubbled
up
from.
So
that's
why
we
focused
it
as
pulling
together
and
collating
and
collecting
the
sig
related
commitment,
but
if
there's
other
stuff
that
needs
to
be
on
there,
please,
let's
find
out
where
it
is
and
find
out.
B
And
kind
of
also
it
in
conjunction
to
where
it's
home,
who
is
the
person,
can
help
me,
maintain
the
kind
of
the
current
status
right,
I'm,
basically
playing
a
high-level
project
manager
for
external
community
by
compiling
things
together,
but
I
need
tech,
leads
and
individual
items
to
help
update
to
the
correct
status
and
the
current
assumption
kind
of
its
era.
Provided
me
write.
A
server
at
me
is
that
they
sig
lead
mail
list
and
group
is
the
one
which
I'm
going
to
be
buggin.
B
A
B
A
B
A
B
B
I
mean
it
right
now,
I'm
doing
it
manually
kind
of
what
as
I
said
right.
The
first
attempt
is
to
do
it
manually,
run
it
by
cichlids
and
clean
it
up
and
then
start
putting
automation
and
correlating.
It
is
what's
actually
in
github
and
hopefully
x,
1
dot,
4
kind
of
right.
Once
we
have
some
shape
of
stipulated
format,
we
can
generate
all
kind
of
graphs
and
reports
out
of
it
to
just
visualize
also
how
many
open
in
progress
issues
with
happen.
How
realistic
it
is
that
they
will
all
converge
in
the
next
three
weeks.
B
Wait
and
I
think
it
would
be
useful
for
community
to
usually
as
well
right,
but
then
yeah
so
for
one
dot
for
probably
really
kind
of
annually.
We
are
working
in
parallel,
is
Brian
grant
and
probably
stroll
a
person
to
work
on
owners
with
proposal
and
automate
a
lot
more
of
that
and
kind
of
track.
Who
are
the
owners
populated
spreadsheet
and
what's
the
status
based
on
github.
B
B
A
Yeah,
making
this
as
transparent
and
also
simple
as
possible
is
the
key
point
of
this.
There
have
been
concerns
recently
of
things
getting
all
the
way
to
looks
good
to
me.
Awake
and
one
other
person
look
at
it
and
then
then
getting
a
whole
bunch
of
work,
rework
to
reexamined
in
a
way
that
changes
directions
so
trying
to
get
to
a
spot
where
we
have
clear
owners
and
I
believe
that
Brian
grant
put
up
a
list
of
owners
by
directory.
Recently
hey,
yes,.
A
A
This
is
awesome
like
thank
you
for
pulling
it
together.
If
there
are
more
comments
or
people
want
to
see
it,
Wow
node,
signode,
so
dawn
10
who's
on
here
is
saying
that
the
signal
team
actually
tracks
and
prioritizes
all
the
features
from
the
whole
community
around
note.
So
she
can
help
you
with
all
of
the
node
related
stuff.
Also.
A
B
A
E
A
E
A
All
right
I
will
take
that
as
a
no
or
you
are
on
mute
so
far,
fantastic,
so
I
see
sig
business
being
added
into
here.
If
they're,
so
cluster
ops,
you
just
gave
us
a
little
bit
of
an
update.
Thank
you
rob
if
there
are
other
cigs
that
want
to
do
updates,
we
can
do
that
now.
You
need
cig
leads
any
Sigma's.
You
want
to
talk
about
something
going
on
in
this
egg.
F
F
Thank
you.
So
we
had
our
like
diminished
community
had
its
own
personal
place,
space
on
OpenStack
summit,
so
of
from
under
the
Swiss
d
we
provided
for
for
the
interested
people,
some
information
or
that
in
OpenStack
and
convenience
corporation,
some
interesting
for
them
of
them
information
or
they
intimidate
so
yeah.
F
It
was
really
useful
for,
like
both
communities,
for
instead
communities
for
people
who
are
that
the
work
in
his
open
stick
and
are
interested
in
grenades
and
for
people
who
are
already
familiar
with
convenience,
but
like
to
dive
into
like
deep
integration
between
cabinets
and
help
instead,
so
I
hope
to
like
to
continue
reduce
the
similar
activities.
So
like
dis
imelda
for
when
summits
and
on
the
simulator
on
so
yeah,
it
was
amazing.
I
hope
that
was
useful.
However,
one
who
attended.
A
A
Yet.
But
there
have
been
some
interesting
discussions
and
the
OpenStack
mailing
list
and
the
network,
the
OpenStack
sig
mailing
list
and
the
network
sig
mailing
list
talking
about
what
IBM
is
trying
to
do.
Making
network
happen
from
open
stock
components
and
Cooper
Nettie's
happen.
Containers
and
Cooper
Nettie's
happen
in
sort
of
a
hybrid
space,
so
that
that's
another
thing
that
came
out
of
the
OpenStack
summit
last
week,
which
was
and.
F
Yeah,
that's
correct,
so,
first
of
all
way
we
are
going
to
integrate
like
to
start
working
on
deep
integration
of
enabling
possibilities
to
run
cabinets
on
OpenStack.
Like
really
first-class
provide
like
a
heavy
car
appointment
based
google
cloud,
we
will
continue
our
edible
airstrikes
places
so
on.
So
we
are
going
to
enable
the
similar
possibilities
to
run
on
a
pure
OpenStack
environment.
Key
ok,
the
next
cluster
like
for
elevation
purposes
but
like
for
getting
started
here.
Dececco
and
the
second
positive
we're
working
on
is
deep
integration
with
Scala
community.
F
We
had
to
breathe,
not
a
brave
one.
Our
discussion
with
Alex
forward
from
Korres
durand
OpenStack
summit,
possibly
the
most
people
from
communities
community
at
the
on
the
open,
stick
summit,
I
suppose
a
habit
and
alert.
So
we
had
like
a
productive
discussion
to
Alex
and
yes,
so
we
are
going
also
to
work
deeper
on
enabling
possibilities
to
run
of
cabinets
cross
as
a
foundation
for
OpenStack.
F
So
like
hugs
peleona
keynote
heuristic
Simon
showed
us
like
the
demo
of
tectonic
project,
like
mirantis
guys
shown
some
demo
get
started
for
a
working
on
mineral,
marantis
and
yeah.
So
this
is
like
early
important
post,
because
comedians
provide
may
provide
really
great
possibilities
to
be
used
as
a
foundation
for
opens
and
as
we
hope
so.
F
F
Yeah,
that's
not
a
common
one,
but
two
yeah.
It's
like
an
interest
in
concept,
so
we
were
also
discussing
it
as
well
and
I
would
be
with
tomatoes
so
yeah
everything
is
going
to
be
really
successful,
is
going
to
be
interesting,
Pepin
in
community
community
or
getting
up
and
stick
it
in
nearest
weeks
and
months.
A
G
Can
if
nobody
chiming
I,
can
click
any
update
to
a
sick
note.
That
would
be
often
on.
Thank
you.
So
we
have
the
reason
we
actively
discuss
an
internally
and
with
our
community
partner
on
the
continent,
random
interface.
So
we
one
side
of
things.
We
are
working
very
actively
with
the
rocky
latest
integration
as
an
authority
of
the
runtime
packing
to
the
Copernicus,
so
make
another
progress,
so
it's
kind
of
almost
finished
on
the
1.0
release.
G
Meanwhile,
we
want
to
enable
other
a
runtime
provider
like
a
for
example,
when
those
hyper
and
all
the
other
end
also
there's
the
oci
RNC
continuity,
wednesday
plasticity
a
lot
of
people
behind
this
kind
of
things.
They
want
to
do
something
part
time
or
all
those
kind
of
things
you
need
to
the
company
Renta
and
each
of
those
implementations
represent
the
group
of
new
users.
So
we
want
to
make
that
it's
easy
for
them
to
pack
into
their
own
rent.
I'm
so
we
didn't
have
the
some
discuss.
G
We
will
publish
it
well,
hello
will
draft
a
dog
and
and
the
sheer
waste
among
the
community,
and
so
we
have
to
discuss
in
deep
discussing
with
the
with
the
a
red
hat
container,
runtime
fox,
and
we
also
have
discussing
with
the
chorus
people
and
the
Durst
design,
discussing
also
with
the
high
pursue
the
sickle.
No,
the
asia
community
meetings
and
also,
we
recently
natural
conclusion
on
the
eviction
on
the
note
and,
if
there's
the
out
of
resource
situation.
G
So
how
we
are
going
to
this
idea,
though
this
is
the
first
emotion
we
are
converging
on
the
kind
of
design
and
on
the
resources,
which
is
the
memory
and
the
disk.
How,
when
there's
the
out
of
resource
before
we
to
make
to
make
sure
no
the
reliability
and
how
I
can
excited
that
you
to
protect
our
node.
So
we
also
actively
working
on
the
automated
a
talker
because
stalkerish
you
a
darker
darker
still.
G
It
is
the
default
one
time
for
us
and
in
production,
so
we're
activity
in
evaluate
those
darker
one
tackle
and
a
talker.
When
that
11
and
there's
the
some
discussing
and
yesterday
one
of
the
developers
and
email
asked
the
feedback
on
the
talker
1000
neighbor,
and
so
if
you
are
interesting
about
all
the
talker
as
the
runtime
pleased
with
I
to
the
our
github
this
for
every
dockery
nice,
we
have
the
validation
issues
and
all
the
problem
integration
problem
rendered
each
washer
we
actually
we
are,
we
make
sure
every
single
issues.
G
We
opted
there
and
provide
a
solution
and
all
maybe
we
just
have
to
solve
those
integration
issues.
So
I
think
there's
all.
We
recently
also
draw
a
conclusion
on
the
GPU
for
the
first
version,
GPU
support
and-
and
that's
premature-
that's
pretty
much.
What
we
did
recently,
which
is
not
a
small
amount
that
you
do
thanks,
hope
we
can
t
iva
all
we
promise
for
the
1.3.
Oh
one
more
thing,
so
we
talked
to
a
conclusion
on
the
note,
the
I
think
that's
the
issue
I
filed
for
the
know
the
problem
api.
G
We
audit
your
conclusion
for
the
first
worship,
so
we
decided
not
to
introduce
new
type
of
the
API
object.
We
are
using
the
event
and
a
condition.
So
please
take
a
look
because
I
updated
with
the
detailed
information
and
then
we
have
the
part
type
of
food
well
engineer
and
have
the
prototyping
and
we
could
leave
the
dymo
if
we
want.
So
that's.
A
A
So
my
hope
is.
We
can
get
this
mailing
list
bootstrapped
out
as
well.
Then
we
can
start
pointing
people
there
for
user
discussions.
We
still
want
to
keep
the
questions
that
are
searchable
being
pointed
out
to
stack
overflow
that
we
have
a
nice
clean,
searchable
answer
for
people
who
are
looking
looking
for
answers
to
question
any
thoughts
or
comments
on
the
cover
native
users,
mailing
list.
A
A
The
users
are
having
and
work
and
give
answers
to
those
those
questions
before
they
have
to
come.
Asking
on
the
mailing
lists
so
links
in
there
again
and
and
hopefully
we
will
get
some
further
feedback.
There
was
an
email
out
to
the
Cooper
Nettie's
dev
mailing
list
a
couple
of
weeks
ago,
and
there
wasn't
a
lot
of
feedback
or
flurry
on
it.
So
I
figure
out
we'll
try
here
as
well
as
anybody
interested,
particularly
in
the
docs
mailing
list.
A
A
F
A
A
If
you
have
interest
in
that
or
our
game
to
help
move
some
of
the
existing
wiki
content
into
this
new
repo,
so
that
it
can
be
better,
updated
and
maintained
by
a
broader
group.
That
would
be
super
super
helpful.
So
please,
let
me
know
if
you
have
interests
in
helping
with
that
work
and
then
also
finding
out,
or
you
know,
helping
us
flesh
out
how
that's
going
to
work
for
the
different
special
interest
groups
as
well.
A
There's
some
conversation
about
whether
we
use
to
get
hub,
wiki
portion
of
that
repo
or
we
set
up
directories
and
just
use
markdown
files
in
those
directories
for
each
of
the
different
special
interest
groups
and
so
on.
I,
don't
think
I
saw
Aaron
on
Aaron
Criken
Berg,
but
he
was
one
who
was
actually
advocating
to
fix
the
wiki
problem
and
I
will
have
to
tell
him
we've
gone
and
gotten
started
down
that
path.
Oh
there
is
an
errand,
but
is
it
Aaron
Criken
burr
not
going
to
tell
me?
A
A
All
right,
it
sounds
like
we've
got
one
well,
thank
you
for
joining.
That's
super
also
will
have
to
connect
and
chat
a
little
bit
more
offline
Rob
had
also
a
question
about
lens
for
Peter
days.
Yes
and
Diane,
do
you
want
to
just
go
ahead
and
give
us
an
answer
on
that
one,
and
then
I
will
do
one
other
thing
which
I
forgot
to
put
in
the
notices
so
Diane
you
game
to
make.
C
Sure
Tom
just
cut
and
ryans
on
the
I
can
see
Ryan's
on
there
somewhere
too.
We've
got
two
talks,
as
most
people
know,
the
open
shift
up
streams,
cougar
Nettie's
and
so
we've
got
to
tocs.
Steve
is
going
to
do
a
talk
on
the
basics
of
containers
and
how
to
run
them
on
openshift
and
if
either
even
cook,
coo
coo
benetti's
in
the
hood.
C
So
there's
gonna
be
a
lot
of
Cooper
Nettie's
talk
in
that
one
and
then
ending
the
day
will
be
Ryan
jarvinen
who's
going
to
do
another
talk
on
openshift
and
containers
from
a
slightly
different
perspective
and
then
there's
a
third
talk.
I
believe
that
all
is
calm
that
Vincent
bats
and
Ryan
are
doing,
which
is
more
on
container
standards
than
it
is
focusing
on
Cooper
Nettie's.
A
Cool
thank
you
to
broaden
out
what
is
happening
at
oz
con,
and
this
is
super
awesome
for
those
of
you
who
might
be
attending
and
who
might
want
to
participate.
We
have
a
lot
of
things
going
on
around
Cooper
Nettie's
at
oscon,
so
there
is
a
two-day
workshop
on
Cooper
Nettie's,
Cooper
Nettie's
from
scratch
to
production
in
two
days.
That
is
happening
on
monday
and
tuesday,
and
then
we
have
a
let's
say:
I'm,
trying
to
find
there's
a
session
on
Cobra
Nettie's
here
at
some
point,
I
think
it's
just
the
two
day
workshop.
A
A
Then
the
more
exciting
piece
is
both
CN,
CF
and
and
Cooper.
Nidhi's
and
Google
specifically
have
gone
ahead
and
pulled
together
a
hackathon
that
is
happening
so
the
Auslan
contribute
hackathon
is
thursday
from
11
to
six
pm
is
happening,
and
we
will
have
a
bunch
of
people
working
from
google
who
will
our
core
contributors
at
the
CNCs
at
the
hack
up
hackathon
the
cabrona
decency
and
CF
hackathon
on
thursday,
and
we
have
pulled
together
a
list
of
things
that
we
want
help
on
and
want
to
work
on.
A
So
there's
a
whole
list
of
automatable
tasks
are
on
documentation,
there's
all
of
the
tasks
or
issues
that
are
labeled,
Help
Wanted
and
all
of
the
issues
that
are
labeled
bounty
in
the
queue
brunette
ease
repository,
and
then
there
are
some
sample
update
issues
that
need
to
be
labeled
so
that
they're
identified
in
this.
But
we're
trying
to
find
a
list
of
things
that
people
can
work
on
and
then
from
that
draw
a
first
and
new
decree
to
Coober
Nettie's
onboarding
contributor
experience,
so
that
we
can
make
that
experience
better.
A
We
haven't
done
a
great
job
of
publicizing
this
yet
because
it's
just
come
together,
but
they
expect
to
see
some
tweets
and
so
sharing
those
will
be
helpful
and
I
will
send
an
email
to
the
mailing
list
to
talk
about
to
talk
about
how
anyone
who's
attending
oz
con
can
help,
and
if
any
of
you
is
interested
in
participating
in
the
hackathon
and
helping
guide
or
onboard
new
or
existing
contributors
and
work
through
issues
with
them
and
try
and
help
them
get
involved
in
the
project.
That
would
be
super
helpful.
A
H
A
So
we've
got
a
lot
going
on
at
oz
con.
Is
there
anybody
else
who
is
attending
and
who
wants
to
step
up
we're
also
having
office
hours
in
Google
booth
for
all
of
the
open
source?
Google
projects,
not
all
of
them?
Let
me
be
clear:
all
of
the
one
thing
in
cloud
so
grp
see
cooper,
Nettie's,
a
Apache
bean
and
tensorflow
will
all
have
office
hours.
So
if
any
of
you,
as
contributors
and
community
members
want
to
come,
participate
in
office
hours
in
Google
booth,
that
would
also
be
super
awesome.
A
E
E
A
A
I
think
that's
pretty
reasonable.
To
do
so
will
at
least
to
do
the
first
meeting
in
June
as
a
Asia
friendly
time
zone,
and
we
will
see
how
this
all
goes.
But
yes,
it's
a
knockout,
a
your
head
said
knots,
not
such
a
good
time
zone
for
you.
It
really
knocks
out
asia.
The
same
way.
The
morning
time
zone
really
hurts
in
in
asia
all
right,
yeah
on
europe
and
asia.
That's.
F
A
A
All
right,
thank
you
all
for
joining
my
grand
experiment
of
this
time
and
we
got
a
little
under
half
the
number
of
purchase
since
that
we
usually
do
so.
We
usually
hover
between
50
and
60,
and
actually
it's
about
thirty
percent
between
and
we
have
20
today.
So
that
is
also
there's
also
an
interesting
piece
to
keep
track
up,
we'll
give
it
one
more
in
June
and
we'll
see
how
that
one
goes.