►
From YouTube: Kubernetes Community Meeting 20190207
Description
We have PUBLIC and RECORDED weekly meeting every Thursday at 6pm UTC.
See https://github.com/kubernetes/community/blob/master/events/community-meeting.md for more information!
A
A
As
an
image
earlier,
we
still
could
use
a
note-taker
for
the
community
meeting,
so
if
Google
Docs
is
behaving
for
you,
please
volunteer
to
take
some
notes
for
the
meeting.
People
do
count
on
the
notes
to
catch
up
on
these
when
they
miss
them,
particularly
if
they're
in
a
time
zone.
That
makes
it
hard
to
attend
this
meeting.
A
B
So
we
are
going
to
speak
about
it
keeps
che
and
we
are
so
Mario
Lori
dohui's
leader.
She
takes
on
developer
tools
at
Haddad
and
myself.
The
vendor
know
who
is
project
manager
at
that
as
well.
So
this
week
we've
just
announced
the
ladies
of
cada
dia
spaces,
which
is
a
productized
version
of
Eclipse
chief
for
for
openshift,
and
we
are
going
to
give
a
quick
introduction,
an
introduction
to
Eclipse
chair
and
open
source
community
ID.
B
So
what
does
that
mean
exactly
and
what
is
eclipsed
she?
In
fact,
it's
a
developer
tool
that
is
running
on
kubernetes
and
it
takes
the
production,
containers
and
augment
them
with
development
tools,
and
it
makes
them
seem
shareable.
So
we
cinch
as
the
developer
a
space
apart
which
are
providing
all
the
developer
tools
you
need
to
edit
so
code
to
be
leaked
to
test
it
and
to
run
it
and
all
the
dependencies
needed
to
work
on
the
project
on
the
project
are
packaged
into
container.
B
So
inside
of
a
chi
develop
our
space,
we
are
providing
the
web
IDE,
and
this
web
IDE
has
compatibility
with
via
SCADA
extensions,
but
those
will
be
packaged
as
containers
to
provide
the
dependency
that
they
need
to
turn,
in
fact,
so,
for
example,
if
you
take
the
Java
extension,
it
will
be
packaged
in
a
container
with
property
decades
put
a
melon
and
grader.
So
this
really
simplifies
a
lot.
The
setup
and
the
configuration
of
the
developer
environment
and
the
other
aspect
is
that
it
understands
a
cube
now
kubernetes
application.
B
C
C
And
if
I
do
get
God
I
have
one
fault
year.
You
see
that
that's
that's
actually
the
chase
server.
So
that's
the
the
web
application
that
allows
us
to
start
workspaces
and
when
you
start
jail
when
you
sell
it
locally
or
you
run
it
in
the
cloud.
The
first
thing
you
will
see
as
the
dashboard.
So
that's
the
dashboard
so
a
year
where
you
can
select
some
technology
to
start
developing
workspaces.
So
you
see
we
support
a
lot
of
languages,
so
everything
will
run
in
a
container
one
or
more
container
defensive.
C
Your
of
depends
on
what
application
you're
going
around
and
what
we're
going
to
see
now
is
actually
a
node.js
application
that
will
use
MongoDB
to
persist
data
so
I'm
not
going
to
use
the
dashboard
here.
The
this
this
che
dashboard
I'm
gonna,
show
something
different.
That's
what
we
call
the
factories,
so
you
can
it's
it's
a
link
that
you
can
put
on
where
you
want
actually,
and
you
can
just
select
it
and
it
will
create
the
development
environment
for
your
project,
so
you
can
put
it
on
your
github
readme.
C
C
The
editor
part
that
we're
gonna,
see
in
a
few
seconds
is
actually
a
non
GIS
application
that
is
running
in
in
a
dedicated
container,
so
it
eats
its
that's
a
component
of
che
that
it's
really
similar
to
view
code.
That's
something
that
we
can
change.
So
it's
it's
pluggable,
that's
the
editor
that
we
provide
by
default,
but
we
can
also
have
we
have
we
used
to
have
another
editor
based
on
Orion
on
Eclipse
Orion.
So
that's
that's
the
one
that
this
one
is
based
on
Monaco
instead.
C
So
that's
a
like.
The
experience
is
really
similar
to
a
desktop
IDE,
so
we
have
a
common
palette.
For
example,
you
can,
if
you
want
to
change
the
team,
so
it's
light
should
be
easier
to
to
read.
We
have
you
can
do
like
code
navigation
or,
for
example,
let's
see
where,
let's
find
all
the
reference
of
this
variable,
for
example
here
so
whatever
you
want
to
do,
you're
used
to
do
is
that's
supported
via
the
language
server
technologies.
C
So
here
there
is
a
node.js
language
server
that
is
helping
with
the
support
of
the
JavaScript
language,
so
it
provides
a
good
completion,
find
usage
of
POD
etcetera.
So
once
we
have
done
that,
what
what
else
we
can
do
is
that
we
can
run
our
our
application.
So
we
have
in
the
the
definition
of
the
of
the
workspace
environment
is
based
on
the
application
app.
C
The
up
EML
that
actually
provide
the
definition
of
the
kubernetes
resources,
so
che
is
actually
using
these
definitions
to
start
a
DB
running
in
a
container.
Will
he
create
store
the
service
that
we're
using
exactly
the
same
definition
that
you
can
can
find.
So
what
we
can
do
now
is
that
we
can
start
using
the
common
of
that.
We
can
run
a
task
to
start.
Do
not
yes
and
we'll
be
able
to
yeah.
C
So
that's
the
application,
that's
onaji
as
I
started,
so
we
can
access
the
application
and
we
can,
for
example,
using
that's
a
simple
application
to
a
degree.
For
example
software
engineer
Mario,
let's
save
it
looks
like
yeah,
you
need
to
reload
it,
but
yeah.
That's
that's
just
running
the
application,
but
we
can
also
do
bug
the
application,
so
it
has
the
back
support
as
well.
So
let's,
for
example,.
C
C
The
save
function,
for
example
here
so,
let's
stripe
like,
for
example,
p.m.
and
move
on.
Okay,
so
you
see
that
the
application
now
is
doesn't
stop
this
in
post
because
we
have
reached
the
breakpoint
and
we
can
see,
for
example,
and
pleae
the
variable
you
can
see
the
attribute
and
the
name
is
the
run.
So
it's
yeah,
it's
pretty
powerful
yeah
and
you
have
the
usual
control
that
you
have
in
a
90.
So
let's
resume
it
yeah.
As
you
see
now
we
have
two
employees
all
right,
so
that's
that's
it
for
them.
C
We
don't
have
a
lot
of
time
to
to
go
through
all
the
features,
but
let
me
just
show
you
what's
what
as
a
plant
so
from
the
browser
we
have,
we
have
contacted
the
che
server
that
has
created
a
che
workspace.
This
che
workspace
is
actually
running
in
a
pod
and
yes
started
a
MongoDB
up
runtime
and
a
node.js
app
runtime
as.
C
C
A
Thank
you
very
much,
and
if
people
have
questions
for
Mario
or
seven,
please
ask
them
in
the
group
chat
or
contact
them
via
the
contact
information.
So
thank
you
very
much
for
presenting
that
next
up.
Thank
you
week
we
have
the
weekly
update
on
the
114
release
Aaron.
Where
are
we
hi
I'm
Aaron
estate,
beard?
I'm?
Sorry,
no
cat
t-shirt
today.
Well,
maybe,
okay,
real
quick!
Here's
glance!
It
scattered
pattern
is
the
sixth
planet
out
there.
This
is
the
sixth
week
I've
been
doing
these
updates.
A
So
there
you
go
Wow,
but
I
got
a
turn
video
off,
because
my
connection
doesn't
seem
to
be
all
that
great,
so
I'm
also
not
gonna,
be
sharing
my
screen
or
anything
like
that.
If
you
take
a
look
at
the
meeting
agenda,
you'll
see
links
for
all
the
things
I'm
going
to
mention.
We
are
in
week
five
week.
Five
is
the
one
week
of
the
release
that
doesn't
have
any
scheduled
milestones,
so
this
is
a
relatively
quiet
week.
A
The
next
milestones
you
should
be
aware
of
is
week
eight
to
February
25th
the
end
of
this
month.
We're
gonna
start
burned
down.
That's
an
increased
cadence
of
meetings
where
more
community
members
will
be
invited
to
show
up
and
we're
really
going
to
start
focusing
on
making
things
more
stable
a
week
after
that
so
March
seventh
week,
nine
code
freeze
will
go
into
effect
in
between
those
two
expect
to
see
a
lot
of
brace
yourselves
can't
resist.
Coming
me
on
the
enhancements
front,
we
have
passed
by
the
enhancements
extension
or
extension
deadline.
A
So
at
this
point,
if
you
are
trying
to
get
an
enhancement
into
the
milestone,
you
need
to
follow
the
exception
process.
I
just
went
through
this
morning
and
reconciled
the
state
of
all
the
issues
in
the
milestone
and
the
feature
spreadsheet.
So
we
have
20
alpha
enhancements,
13
beta
enhancements
in
six
stable
enhancements
slated
for
this
release.
That
seems
like
a
lot
we'll
find
out
how
that
evolves.
A
As
we
get
closer
to
code
freeze,
the
next
steps
from
an
enhancements
perspective
are
to
go
through
and
actually
review
all
of
the
test
plans
to
make
sure
that
those
are
actionable
review.
Their
graduation
criterias
make
sure
that's
fact
and
make
sure
the
considerations
for
upgrade
and
downgrade
are
worthwhile
totally
fine.
Having
that
be
fluid
for
a
little
bit,
but
as
we
get
close
to
or
to
burn
down
we're
gonna
want
that
to
be
kind
of
locks
down,
so
that
the
goal
posts
aren't
moving
from
a
CI
signal
perspective.
A
The
CI
signal
report
hasn't
been
updated
this
week,
so
I'm
just
going
off
of
what
I
can
see
in
test
grid.
We
are
not
very
happy
from
an
upgrade
test
perspective,
though
thankfully
it
seems
to
be
due
to
a
limited
number
of
tests
and
the
master
blocking
board
is
having
some
issues
so
look
for
CI
signal
and
buck
triage
to
come,
knocking
on
the
doors
of
the
people
responsible
for
fixing
those
things
if
they
haven't
been
knocked
on
already
other
than
that,
like
I
said
it's
week,
five,
it's
a
quiet
week!
A
So
that's
all
the
updates
I
have
from
if
I
have
the
spare
time
like
from
an
enhancements
perspective,
recognizing
that
this
is
all
been,
maybe
a
slightly
bumpy
ride
or
not
as
well
communicated
as
some
people
would
have
liked.
Does
anybody
want
to
give
any
feedback
right
here
about
how
the
process
has
gone
for
them?.
A
Okay
figuring
that
I
might
have
gotten
that
wonderful
silence.
In
response
to
that
question,
we
do
have
a
retro
dock
that
is
opening
to
the
open
to
the
community.
So
if
you
want
to
put
feedback
in
the
retro
dock,
that's
the
retro
dock,
so
we'll
review
that
as
a
community
and
then
also
yes,
Steven.
Thank
you
a
reminder
that
caps
as
a
process
are
owned
by
8
p.m.
cest
p.m.
is
the
place
to
take
all
discussions
about.
What's
bad
about
caps,
what's
good
about
caps,
how
we
should
make
them
better.
A
D
So,
hey
everybody,
I'm
Frederick
I
work
at
Rahab,
and
this
is
the
second
cementation
update.
So
since
our
last
update,
I
gave
a
little
bit
of
an
update
about
this
topic,
but
we
were.
We
were
working
really
hard
on
a
lot
of
performance
improvements
on
one
of
the
community.
Sub
projects
called
cube
state
metrics.
This
is
a
component
that
essentially
takes
kubernetes
objects
and
turns
them
into
Prometheus
metrics.
So
one
just
as
an
example,
one
example
could
be.
D
We
take
the
deployments,
object
for
example,
and
expose
the
expected
amount
of
replicas
and
the
actual
amount
of
mech
replicas,
and
then
you
could,
for
example,
do
some
alerting
on
that
kind
of
data
and
because
kubernetes
can
have
lots
of
objects.
This
can
put
some
load
on
this
particular
component,
so
we
did
a
pretty
large.
This
particular
release
featured
a
large
rewrite
and
refactoring
of
the
codebase
in
order
to
do
some
massive
performance
and
resource
improvements.
D
So
if
you're
interested
in
the
details
about
this
I
gave
that
at
the
last
update,
but
now
we
have
actually
released
the
stable
version
of
this
and
another
huge
thanks
to
max
in
them
for
doing
the
majority.
The
overwhelming
majority
of
this
work,
who
because
of
this,
is
now
a
member
of
the
kubernetes
Congrats
to
the
mass
and
the
second
thing-
and
this
is
probably
the
biggest
thing
that
sig
instrumentation
is
doing
for
1.14-
is
the
metrics
overhaul
enhancement.
D
Essentially,
what
this
features
is
we're
going
through
all
the
metrics
that
would
kubernetes
that
fraternities
is
currently
instrumenting
and
we're
making
sure
that
everything
aligns
with
the
instrumentation
guidelines
that
have
been
written
roughly.
We
introduced
them
roughly
two
and
a
half
years
ago,
and
since
then,
a
number
of
violations
have
sneaked
into
the
code
base
so
we're.
Essentially,
this
is
a
cleanup
and
making
sure
that
everything
goes
by
the
guidelines.
D
We
don't
do
not
formally
have
any
stability
and
compatibility
around
metrics.
We
do
intend
to
change
that.
Eventually.
This
cap
is
not
doing
that,
but
it's
kind
of
hopefully
leading
up
to
it.
It's
a
separate
discussion
to
have,
but
we
do
recognize
that
a
lot
of
people
are
using
a
lot
of
these
metrics
that
we're
touching
so
1.14
is
kind
of
going
to
be
a
deprecation
release
where
both
the
yi,
both
the
deprecated
metrics,
as
well
as
the
new
metrics,
are
going
to
be
present.
There
are
some
there
are
barring
changes.
D
D
D
So
this
is
kind
of.
However,
it
can
be
used
by
any
any
system
that
is
compatible
with
prometheus
metrics.
However,
we
are
actually
not
using
this
with
Prometheus
we're
just
implementing
the
same
kind
of
format.
Support
into
the
metric
server
for
this
and
the
really
the
change
that
we're
introducing
here
is
that
these
metrics
are
then
going
to
be
served
through
the
stats
endpoints
provided
by
the
container
of
runtime
interface
instead
of
through
C
advisor
yeah.
D
So
that's
that,
if
you're
interested
in
more
of
the
details,
the
second
campus
actually
is
super
well
researched
and
benchmarked
and
everything.
So,
if
you're
interested
in
something
like
that,
I
recommend
reading
that
they
would
put
a
lot
of
thought
into
that,
and
that's
really
it
if
there
are
any
questions,
I'm
happy
to
answer
those.
Otherwise
we
can
go
on
to
the
next
cig.
A
Okay,
it's
about
it,
make
sure
the
slides
are
linked
in
the
meeting
notes.
Please
well!
Thank
you
very
much
for
presenting
on
on
your
caps
that
really
helps
people
actually
keep
track
of.
What's
going
on
in
development,
okay,
so
our
net
speaking
of
caps,
our
next
sig
is
sig
PM
steven
augustus.
You
want
to
go
ahead.
E
E
We
gave
an
update,
I
believe
at
the
beginning
of
January,
so
a
lot
of
us
will
be
rehash
for
some
people,
but
let's
kick
it
off.
So
here's
what
we
were
working
on
last
cycle.
We
kicked
out
a
cig
TM
survey,
which
we
also
kind
of
used
as
the
impetus
for
the
the
cube
con
Seattle
intro
and
some
of
the
deep
dive
talks.
So
we've
been
there
there's
been
kind
of
a
massive
effort
around
revitalizing
the
kept
process.
So
some
of
the
things
you
might
have
seen
come
across.
E
Ke
dev
are
the
fact
that
the
the
kept
meta
implementation
is
kind
of
in
progress.
So
there
is
a
meta
kept
talking
about
how
we
can
improve
the
process.
That's
kept
one
a
there
are
links
to
it
later
in
the
slides.
We
built
a
project
tracking
board
for
the
kept
implementation.
So
all
of
the
stuff
that's
laid
out
in
that
cap
are
actually
put
up
on
a
project
board
and
a
an
umbrella
issue
detailing
all
of
that
work.
E
So
there
has
been,
there
have
been
questions
around
what
enhancements
are
and
some
confusion
there
so
enhancements
just
to
just
to
lay
it
down
here.
Enhancements
are
a
term
to
encompass
a
set
of
a
set
of
kubernetes
enhancement
proposals
and
features
right.
The
idea
of
the
the
cap
is
that
it
is
the
steel
thread
that
captures
an
idea
from
essentially
from
the
the
ideation
state
to
implementation
state
right.
So
it's
the
the
I
have
a
great
idea.
E
Can
we
do
this
and
moving
through
acceptance
and
alpha
beta,
stable,
GA
stages
and
and
capturing
all
of
that
stuff
in
one
place?
So
that's
what
the
cap
is
right
that
operates
over
the
enhancement,
which
is
meant
to
be
a
multi
release
set
of
work.
Essentially,
so
that's
that's
a
little
different
from
a
feature
which
is
essentially
supposed
to
be
scoped
to
a
single
release
cycle.
So
the
idea
here
is
by
rebranding
what
features
were
before.
E
We
have
an
opportunity
to
capture
features
in
a
single
release
cycle
and
then
bring
those
thoughts
from
the
features
into
a
single
thread,
which
is
the
cap
right.
So
some
of
the
some
of
the
kind
of
administrivia
around
doing
that
included
included
categorizing
the
open
caps
and
design
proposals
as
kind
cap
creating
new
labels
for
that
making
sure
that
everything
was
tagged
properly
and
and
the
approvers
and
editors
were
updated
on
that
renaming.
E
The
kerbin
Nettie's
features
repo
to
kubernetes
enhancements,
so
that
happened
a
little
while
back
and
starting
to
and
moving
all
of
the
caps
that
were
in
the
kerbin
IDs
community
repo
over
to
kubernetes
enhancements
right.
So
the
cube
con
talks
where
we
talk
about
caps
and
and
quite
a
bit
of
detail
or
there,
the
introd
I've,
as
well
as
the
talk
that
jason
I
did.
Jase,
Caleb
and
I
did
on
enhancing
kubernetes,
so
what's
been
happening
with
caps
and
the
kubernetes
the
kubernetes
release
cycle.
E
So
we
kind
of
said
this
time
around
that
we're
going
to
require
caps
starting
the
1:14,
we're
going
to
require
caps
for
code
or
process
and
release
changes
that
are
landing
in
1:14
right.
So
there
was
a
discussion
thread
that
happened
around
the
enhancements
freeze
so
last
week,
week
and
a
half
ago,
where
I
detailed,
when
you
should
and
shouldn't
submit
a
cap
and
what
is
required
for
certain
people
to
be
doing
right
now.
E
So
now
that
we've
passed
enhancements
freeze
as
in
as
Erin
mentioned,
you
need
to
file
an
exception
and
also
make
sure
that
your
caps
are
approved
and
set
to
implementable
right.
So
the
difference
here
is
that
caps
that
are
targeting
urban
attics
Kerber.
Nowadays
we
require
blue
enhancements
that
are
targeting
kubernetes
kubernetes.
We
require
a
cap
for,
and
it
require
an
enhancement
tracking
issue
for
incur
Bonetti
as
enhancements
for
people
who
are
working
on
out
of
tree
code.
E
The
release
team
doesn't
really
not
not
that
they
don't
care,
but
they
don't
need
to
track
out
of
tree
code.
So
if
you
have
an
enhancement
issue
there
that's
great,
but
it's
not
required
for
the
release
team
to
move
forward,
and
they
won't
be.
You
know
the
upside
of
that
is.
They
won't
be
bugging
you
every
every
two
days
about
about
updating
your
test
plans
or
updating
different
things
like
that
all
right,
so
that's
a
little
added
benefit
there.
E
So
part
of
part
of
the
so
there's
been
a
little
bit
of
confusion
and
part
of
that
confusion
is
that
we
needed
to
update
the
cap
cap
template
right.
So
there
is
a
PR
up
for
the
cap.
Template
update.
Anyone
who
is
interested
in
submitting
feedback
on
that
definitely
definitely
go
go
and
do
that
I
will
probably
time
box
it's
a
few
more
days
just
so
that
we
have
an
updated
template
in
place
and
then
we'll
work
on
bringing
the
document
the
documentation
in
the
repo
forward
to
kind
of
match
against
that
template.
E
So,
as
I
mentioned,
the
the
cap
template
is
there
in
the
repo.
So
that's
that's
a
kept
zero.
The
kept
process
is
detailed
in
kept
one
and
the
cap
implementation.
It's
kind
of
all
the
stuff
that
we're
working
on
right
now
is
detailed
and
cap
zero
one
a
there
is
also
discussion
around
a
contributor
site
and
how
to
essentially
bring
the
in
improve
the
presentation
of
caps
right
so
bringing
them
to
something
like
the
contributor
site
so
that
it's
easy.
It's
got
a
view.
Instead
of
you
know
kind
of
jogging
through
a
repo.
E
So
some
things
that
we
were
kind
of
discussing
and
wondering
if
they're
a
good
idea
is
definitely
open
to
feedback
here.
One
I
think
this
should
absolutely
happen
if
you
are
already
a
kept
reviewer
or
you're.
A
sig
lead
chair,
feel
free
to
add
people
in
your
sig
that
that
you
believe
are
good,
kept
reviewers
to
your
owners
files
within
the
the
kubernetes
enhancement.
Arriba.
E
Aaron
I
believe
Aaron
came
up
with
this
idea
for
kubernetes
for
kept
office
hours
so
having
and
an
opportunity
to
discuss
the
kept
process
if
you're
being
blocked
on
certain
things.
You
need
to
talk
about
certain
things
about
the
structure
and
having
those
happen
alternate
in
alternating
time
slots
for
the
sig
PM
meeting.
So
we
have
two
again
one
meeting
one
week
and
then
they
kept
office
hours
the
next
week
and
moving
forward
like
that.
E
We
think
that
that
will
give
a
good
opportunity
to
kind
of
like
break
down
this
process
and
and
have
a
tighter
feedback
loop
here,
and
then
we
also
want
to
migrate
the
design
proposals
from
kubernetes
from
kubernetes
community
into
kubernetes
enhancements,
not
not
a
forklift
but
split
this
up
by
cigs
and
have
them
migrate
into
caps
right,
so
more
plans
for
the
upcoming
cycles
and
I'll
kind
of
speed
through
these.
Here
we
want
to.
We
want
to
enhance
the
sig
PM
charter.
The
Signum
charter
was
kind
of
created
before
charters
were
solidified.
E
So
now
that
that
has
happened,
we
want
to
come
back
and
redefine
our
efforts
for
the
project
cleaning
up
the
process,
documentation
really
getting
to
the
point
where
we
understand
exactly
who
owns
what
you
know
there
have
been
discussions
about
who
owns
the
process
around
caps,
which
is
sig
jam.
We've
recently
decided
that
and
who
owns
the
process
around
reviewing
caps,
which
is
SiC
architecture,
building
that
glossary
of
terms.
So
what's
a
cap
what's
Enhancement?
E
You
know
we
want
again
an
easy
presentation
mechanism
for
the
four
keps
so
considering
a
separate
page
on
the
contributor
site
as
well
as
we
want
an
easy
way
for
people
to
move
through
the
different
graduation
stages
of
a
cap
right,
so
bringing
a
programmatic
way
of
doing
that
via
a
CLI
tool
that
that
Kayla
miles
has
spent
a
quite
a
bit
of
time
on.
So
we
want
to
figure
out
the
best
way
to
onboard
that
and
bring
that
into
the
project
and
then
kind
of
once
we
have.
E
You
know
we're
starting
to
build
on
this
people,
this
people
process
and
then
tools,
situation
and,
and
once
that
happens,
I
think
we
have
a
good
opportunity
to
start
introducing
things
like
project
wide
roadmaps
and
release
planning
events
right.
That
happened
crossing
right.
So
those
are,
those
are
essentially
the
goals
for
2019,
first
APM,
so
we're
to
find
us.
You
know
who
we
are:
we've
got
Aparna
Igor,
caleb,
chase
myself
and
Dustin
are
the
chairs
of
the
sub
projects
and
chairs
of
sig
p.m.
E
A
E
So
those
are
basically
organization
around
handling
handling
different
things
like
deciding
the
enhancements
leaves
an
enhancement
process,
so
that's
sort
of
under
product
management,
the
program,
management
and
and
release
or
more
project
program,
wide
efforts.
So
we're
looking
at
definitely
like
within
the
within
the
sig
PM
charter.
The
revamp
Charter
will
we'll
look
to
better
define
that
I'll
be
honest.
It's
not
well
defined
right
now,
so
that's
as
best
as
an
answer.
I
could
give
sure
I.
A
A
E
There
is
that
there's
a
cap
CLI
tool
that
is,
it's
essentially
a
library
that
Caleb
built,
so
that
will
that
will
work
to
kind
of
reduce
the
burden
on
people
who
are
submitting
caps
right,
ensuring
that
the
metadata
is
all
up-to-date
and
stuff
like
that,
and
then
also
people
who
are
reviewing
and
approving
caps.
So
that's
a
start,
but
in
the
future,
I
also
want
to
look
at
means
of
bringing
the
feature
triage
process
into
place.
E
So
when
someone
drops
say
this,
when
someone
drops
like
hey
I,
want
a
new
future
in
kubernetes
on
group
nineties
kubernetes,
how
do
we
process
those?
How
do
we
bring
those
into
place
where
a
cig
has
visibility
on
them
and
we
can
and
we
can
actually
bring
them
into
the
project
so
that
that
I
created
a
cap
at
some
point?
That's
hidden
in
a
Google
Doc
that
I
can
I
can
bring
forth,
but
we're
not
there.
E
A
One
other
knit
I
guess
I
have
froze
it
in
the
question
are:
are
you
sure,
sig
architecture
owns
the
kept
review
process?
I
was
under
the
impression,
as
I,
shopped
caps
around
between
architecture
p.m.
and
release
that
we
had
agreed
that
sega
architecture.
Members
are
really
useful
people
to
have
listed
as
reviewers
or
approvers
for
caps,
and
we
would
definitely
solicit
their
feedback
on
a
consistent
set
of
criteria,
for
example
that
maybe
defines
what
alpha
and
beta
and
stable
mean,
but
we
don't
expect
them
actually
in
the
process
per
se
right.
E
F
F
One
thing
that
I
want
to
talk
to
Steven
about
introducing
into
the
cap
process
is
this
idea
of
C
specific
caps,
which
are
really
just
sort
of
a
way
for
SIG's
to
organize
and
communicate
versus
cross-cutting
caps
that
are
going
to
be
sort
of
architectural
cap
swords?
You
know,
project
wide
caps
and
I
think
there's
going
to
be
different
processes
for
each
of
those
for
the
cross-cutting
stuff.
I.
Imagine
that
cig
architecture
is
going
to
be
intimately
involved,
but
the
process
for
all
this
stuff
right
now
is
it's
with
sig
p.m.
right.
E
So
so
part
of
some
part
of
that
delineation
is
its
breaking,
so
we've
already
started
to
do
that.
Break
the
the
individual
caps
into
separate
sub
folder
sub
sub
directories
within
k
enhancements.
So
each
of
those
sub
directories
are
owned
by
the
the
individual
SIG's.
But
yes,
we
do
need
to
introduce
more
of
a
defined
process
around
kubernetes,
wide
caps.
A
F
A
These
do
I,
don't
know
how
many
of
you
have
had
the
pleasure
of
interacting
with
Tennessee
Tim
st.
Clair
and
where
he
talks
about
how
we
have
this
wonderfully
consistent
process
that
we
use
in
the
cluster
lifecycle
projects.
Everybody
loves
it,
it
works,
we
get
work
done.
Everybody
knows
what
work
is
supposed
to
be
getting
done.
A
Everybody
knows
what
the
right
priority
for
stuff
is
and
to
me
this
sounded
like
a
great
solution
for
SIG's
that
say
that
people
show
up
and
they
have
no
idea
what
to
work
on,
and
so
it
turns
out,
though,
that
when
Tim
says
we
have
this
process
that
it
wasn't
actually
documented.
It's
just
something.
A
G
A
Announcement
combo
shout
out
to
Tim
for
that.
Thank
you
and
then
someone
else
was
going
to
say
a
little
something
about
what
happened
on
Sunday
and
the
follow
up
from
them.
I
was
at
Paris.
Are
you
gonna
do
this?
I
was
somebody
else
gonna
do
this?
I
can
speak
to
this
if
Paris
isn't
surrounded,
okay,
so
and
Josh?
A
Maybe
you
can
correct
me
if
I've
missed
the
tails
here
so
loosely
speaking,
we
had
an
incident
on
Sunday,
where
some
malicious
individuals
showed
up
on
slack
and
potentially
abused
privileges
and
posted
a
bunch
of
code
of
conduct
violating
content
everywhere.
In
response,
the
slack
admin
team
moved
to
delete
those
messages,
banned
those
users
and
then
also
to
just
lock
us
down.
We
shut
down
the
invite
page,
we
disabled
pretty
much
every
single
apps
access
to
slack
in
case
that
was
how
they
were
getting
in
and
we
are
effectively
in
lockdown
from
so.
A
A
Right
right
now,
the
process
would
be
go
to
CID
contributor
experience
or
now
actually
go
to
slack
Edmunds,
where
we
have
people
of
the
rights
to
shoe
personal
invitations.
I'm
I'm
wary
of
suggesting
even
that
I
don't
know
if
that's
what
we
hacked
actually
had
agreed
upon
within
the
retro,
for
example,
one
an
option
that
had
been
discussed
is
like
what,
if
we
open
up,
slack
invites
but
try
and
dump
it
into
some
kind
of
moderation
queue,
and
the
issue
is
that
we
receive
on
the
order
of
75
to
100
invite
requests
per
day.
A
G
I'm
one
of
the
slack
admins
one
thing:
one
thing
we
we're
putting
on
the
page:
there
is
if,
if
it's
like
an
important
thing-
and
you
really
need
to
slack
to
somebody
sure,
but
we
don't
want
to
get
into
the
into
the
having
to
process
all
of
the
SIG's
invites
and
stuff
there's
just
way
too
many
for
us,
so
case-by-case
I
guess,
while
we
sort
while
we
sort
out
this
the
the
the
situation.
Sorry
I.
A
Think
another
another
rah-rah
thing
I'll
just
say
real
quick,
is
I
feel
like
this
was
a
great
learning
opportunity
for
everybody
involved.
Kubernetes
slack
instance
was
by
far
not
the
only
slack
instance
hit
by
this
particular
kind
of
attack,
and
so
we
are
reaching
out
to
other
open
communities
and
trying
to
figure
out
a
way
to
kind
of
work
on
establishing
best
practices
and
tools
to
help
us
deal
with
this
in
the
future.
A
So
if
you
know
anybody
who
runs
a
large
slack
instance,
like
we've,
been
in
contact
with
cloud,
foundry
hang
off
to
go
line,
and
things
like
that.
Please
reach
out
to
the
slack
admins
channel,
then
we'll
get
that
going
and.
A
F
A
That's
both.
F
Me
and
Aaron,
but
like
yeah
I,
think
you
know,
I
I
didn't
find
out
about
this
until
Monday
morning
and
I
was
like
blown
away
with
folks
dropping
everything
to
really
attack
this
deal
with
some
real
unpleasantness
I,
you
know
figure
out
the
right
way
to
mitigate
it
under
pressure,
so
amazing
work
with
the
with
a
set
of
people
that
really
stepped
up
to
do.
This
I
think
you
know
Aaron
was
it
was
one
of
those
folks
Paris
and
Tim
Hawken
and
I'm
sure
I'm
forgetting
some
names
here.
George
do
you
know
who
code.
G
F
A
The
the
shout
outs
that
I
have
from
for
for
various
people
who
helped
with
the
effort
were
Noah
Paras,
Jase,
Aaron,
Ben
elder
and
Tim
Hawkins,
all
of
whom
jumped
in
at
various
parts
of
this
crisis.
In
order
to
deal
with
with
deal
with
it,
including
dealing
with
figuring
out
who
had
the
permissions
on
certain
things,
like
the
invitation
page
turned
out
to
be
something
we
didn't
have
written
down.
So.
A
One
other
person
add
to
that
list
would
be
Sarah
Novotny,
who
I
managed
to
page
in
has
owner
privileges
to
our
slack
instance,
I
caught
her
like
sometime
after
midnight,
her
time
zone
so
really
appreciate
it.
Okay,
so
I'd
love
to
have
a
quick
round
of
applause
for
this
group
of
people
for
being
up
till
what
was,
in
some
cases,
4:00
in
the
morning
people's
local
time
dealing
with
us.
So
thank
you
very
much
for
for
doing
that,
and
keeping
slack
useful
for
the
contributors.
A
G
A
And
next
week
is
the
14th,
I,
guess
yeah.
So
next
week
our
six
presenting
our
cig,
GCP,
cig
storage
and
cig
docks.
So
if
you
are
a
member
of
one
of
those
SIG's,
please
talk
in
your
cig
meeting
about
getting
an
update
presentation
together.
Our
demo
next
week
is
to
do
cube
service
exporter
hi
from
Guinevere
so
tune
in
next
week,
or
you
know,
watch
this
on
youtube
and
see
you
all.
Then
thanks
everyone
happy
Thursday.