►
From YouTube: Kubernetes Public Steering Committee Meeting 20190213
Description
SC Repo https://github.com/kubernetes/steering
Agenda and Notes http://bit.ly/k8s-steering-wd
A
Run
through
the
steering
board
projects,
it's
in
the
project
number
one
in
the
steering
repo
I've
been
connected
to
zoom
unit,
so
I
can't
post
the
chat
and
we
go
through
the
name,
lady
in
progress
column
and
the
reason
we
thought
would
be
a
good
idea
to
invite
more
focuses.
So
more
people
can
help
work
on
some
of
these
issues
and
participate
more
directly
than
just
watching
the
videos.
After
the
fact
I
don't
know.
If
Erin
is
on
a
lot
of
times,
he
runs
through
the
board.
A
I,
don't
know,
there's
nothing
else
is
in
the
agenda
right
now.
I
did
want
to
make
people
aware
that
we
just
recently
added
a
concept
of
user
groups.
Brendon
documented
that
and
I
reviewed
the
PR
so
that
got
merged,
so
that
adds
a
new
kind
of
group.
We've
been
discussing
it
for
quite
a
while.
There
are
previous
discussions
of
birds
of
a
feather
groups
or
informal
groups,
groups
that
are
not
the
same
as
working
groups
necessarily
focus
on
a
specific
topic.
A
Cluster
ops
was
sort
of
a
working
group
and
that
sig
we're
kind
of
cleaning
up
the
organizations,
so
we've
archived
some
SIG's.
Recently
so
things
for
the
folks
have
been
helping
with
that
officially
in
the
community
repo.
If
you
take
a
look-
and
this
is
sort
of
part
of
that
cleanup
so
right
now
there
are
discussions
about
converting
the
Big
Data
sig
to
a
user
group,
because
it
is
not
really
working
on
code
and
the
kubernetes
code
base.
So
we'll
continue
to
look
at
how
we
can
more
effectively
organize
the
project.
A
The
other
main
thing
we've
been
pushing
on
in
that
regard,
and
there
was
a
big
push
before
key
Khan
is
charter.
So
we
have
this
charters,
meta
issue
and
most
of
the
charters
at
least
initial
versions
are
in
I.
Think
sick
docks
was
one
that
was
still
active
Phil.
Do
you
know
anything
about
that?
Yeah
I
mean
pull
that
up.
It
was.
C
A
A
The
next
in-progress
issue
is
document.
What
working
groups
are
and
how
to
form
and
disband
I
think
that
was
a
lot
of
that
was
done
so
I'm
looking
at
what
is
left
so
notes
from
Aaron
says
document
the
working
groups
may
encode
and
associated
repos.
That's
been
closed
and
add
a
field
to
60ml
to
list
SIG's
that
are
working
group
spans.
So
I
think
that's
why
this
issue
is
still
open.
A
E
D
A
D
A
A
F
A
A
G
G
A
A
A
D
C
D
Knowing
all
I'll
share
I'll
share
the
doc
with
everyone
right
now
that
I'm
working
on
it's
it's
roughly
I'd,
say
six
dependencies
that
I'm
that
I'm
combining!
So
that's
why
I
did
it
in
a
doc
versus
a
PR,
because
I
just
wanted
to
list
out
the
tasks
and
things
like
that,
but
I'll
share
what
the
I'll
share.
Everybody
with
that
excuse
me
I
will
share
with
everyone
on
the
line
right
now
on
the
agenda
with
that
link,
cool.
H
A
A
B
A
A
The
work
that
remains
to
be
done.
There
is
also
on
the
automation
front,
similar
to
the
working
group
stuff.
We
have
owners
files
listed
in
60ml,
but
there's
still
work
that
people
want
to
do
to
consolidate
owners
files
because
we
have
like
$5,000
files
and
to
make
the
lists
of
people
in
the
with
the
different
roles,
approvers
reviewers
and
so
on.
Four
sub
projects,
more
centrally
managed
I
think
there
was
some
work
in
contributor
experience
being
done
on
that
related
to
the
work
to
synchronize
or
to
program
github
teams
using
checked
in
files.
Yes,.
A
A
A
Really
it
should
be
sub
projects,
so
we've
been
reaching
out
to
those
working
groups
to
figure
out
the
status
is
Derek
on
and
Derek
and
mention
that
meet
the
resource
management
working
group,
which
is
existed
for
a
long
time,
might
be
able
might
be
ready
to
wind
down
and
they
could
write
retrospective
of
about
the
work.
What
the
working
group
accomplished
and
things
like
that,
so
that
seems
like
that-
would
be
good.
A
F
The
latest
update
is
the
Contra
backs,
has
a
sub-project
a
few
of
us
Nikita
myself,
Steven
Augustus
and
Steven
Winslow.
The
four
offers
are
participating
in
that
right
now.
The
idea
is
to
try
to
see
if
we
can
use
automated
tools
or
manual
tools
to
identify
which
licenses
we
use
and
whether
they
are
compatible,
whether
they
are
in
the
list
of
the
approved,
whitelist
or
not,
and
send
the
list
of
the
ones
that
are
not
in
the
approved
list
to
the
cnc,
if
a
governing
board
essentially
Steve
and
takes
care
of
that.
F
G
A
F
A
F
H
H
F
So
one
one
major
thing
that
we
need
to
do
here
is
at
this
point
the
main
KK
repositories.
A
repository
is
good.
You
know.
The
only
thing
that
were
that
was
spending
war
was
will
be,
you
know,
approved
in
the
cnc
of
counting
body,
but
the
problem
is
all
the
other
repositories
that
vendor
code
from
KK
still
have
really
old
licenses.
So
we
have
to
somehow
figure
out
how
to
update
the
rest
of
the
repositories
from
the
main
KK
repositories
over
a
period
of
time,
so
that
everybody
else
is
cleaned
as
well.
H
So
one
of
the
notes
there
is
that
so
far,
so
the
scanning
tool
that
we're
currently
using
has
the
ability
to
our
configure
policies
for
the
project.
So
now
that
we
have
the
list
of
whitelist
whitelisted
licenses
for
the
scene
CF,
we
can
build
a
policy
around
those
and
then
rerun
scans
across
the
kubernetes
org
repos,
and
get
a
report
on
that.
Okay,.
F
And
yeah,
the
only
other
thing
I
need
to
mention
here
is
right:
now
it
will
the
license
scanning.
The
manual
process
is
done
on
ad
hoc
basis.
We
are
trying
to
see
if
we
can
synchronize
it
with
the
cig
release.
Can
you
and
try
to
do
it?
You
know
when
we
try
to
lock
down
the
code
and
the
features
are
you
know
we
can
be
logged
on
the
features.
I
H
F
Right
I
mean
they'll,
always
be
a
nightly
one.
That
will
be
part
that
we'll
be
publishing,
but
this
one
is
the
more
formal
one
that
that
Stephen
Winslow
does
by
hand
at
least
right
now,
and
he
will
be
continuing
to
do
that.
The
one
that
he's
doing
by
hand
until
we
are
okay
with
the
tool
and
what
it
provides.
You
know
the
faucet
to.
A
B
G
A
A
I
can
speak
to
that
a
little
bit.
I
don't
know
anybody
else
is
on
the
call
who
may
have
been
more
involved
with
that,
but
I
did
actually
review
multiple
versions
of
that
proposal.
So
yeah
for
the
folks
who
are
not
aware
the
CN
CF
TOC
has
been
trying
to
discussing
how
to
scale
the
effort
over
the
past
six
to
nine
months.
The
demand
from
projects
to
get
reviewed
by
the
TOC
has
been
fairly
steadily
growing
and
other.
A
Other
activities
have
also
been
growing,
like
project
reviews
for
graduation,
from
sandbox
to
incubation
or
incubation
to
graduated,
and
things
like
that.
Also,
we
had
maybe
a
year
and
a
half
ago
created
a
list
of
the
areas
where
we
thought
there
were
gaps
in
the
cloud
needed
ecosystem
where
we
needed
for
projects
and
pretty
much.
Those
gaps
got
filled,
and
now
there
are
probably
is
the
next
set
of
gaps
we
need
to
be
looking
for,
but
the
TOC
doesn't
necessarily
have
expertise
in
all
areas.
A
So
this
idea
of
SIG's
loosely
modeled
after
community
SIG's,
a
number
of
folks
from
the
community
like
Quinta
cool
they're,
employed
works
on
that
proposal,
along
with
other
folks
to
create
kind
of
bodies
of
expertise
that
could
do
help
the
TOC
with
some
of
that
legwork
and
apply
domain
expertise
to
multiple
different
areas.
So
a
concrete
set
of
SIG's
initial
set
of
six
has
been
proposed.
I,
don't
necessarily
think
it's
cast
in
stone,
but
you
know
it
would
be
idea.
A
A
In
complementary
areas
you
know,
networking
storage
are
still
gonna,
be
there.
They're
still
gonna
exist,
they're,
looking
at
adding
additional
areas,
for
example,
moving
up
the
stack
into
more
application
deployment
areas
that
are
kind
of
pretty
clearly
out
of
scope,
kubernetes
I
think
in
terms
of
there
had
been
confusion
in
the
past
in
some
overlapping
areas,
particularly
storage,
but
also
networking.
A
Hopefully,
that
is
mostly
been
addressed.
As
you
know,
in
CSI
evolved.
Other
storage
related
efforts
are
clearly
out
of
scope
for
things
of
interest
to
kubernetes
direct
interest
anyway,
so
I'm
hoping
it
will
turn
out
to
be
fairly
complimentary.
Did
you
have
specific
thoughts
in
mind
dims
about
areas
where
you
thought
there
would
be
more
connections
right.
F
So
some
of
the
questions
I
had
was
like
who,
from
this
group
or
the
overall
cube
Redis
community
will
be,
you
know,
will
need
to
go
show
up
for
those
meetings.
Is
there
kind
of
like
a
reporting
structure
where
we
have
to
send
some
information
back
to
them
or
is
there
avenues
where
we
can
ask
for
help
from
CN
CF
or
the
TOC
in
things
that
we
are
hitting
against
that
kind
of
questions?
So.
C
C
With
respect
to
ask
for
help
there
are
you
know
there
are.
You
know,
multiple
examples
of
where
the
same
CF
has
funded
stuff
for
the
project.
I
think
we
continue
to
explore
where,
where
that
makes
sense,
and
in
to
continue,
you
don't
go
down
that
path
now
in
terms
of
help
from
the
TOC
I'm,
not
sure
you
know
what
you
were
thinking
about.
Their
games,
I'm.
G
C
The
SIG's
are
largely
sort
of
under
the
TOC,
not
under
the
governing
board,
and
there
is
no
explicit
connection
between
kubernetes
and
representation
into
the
TLC,
but
Brian
and
I
both
sit
on
both
the
steering
committee
and
the
TOC.
Well,
you
know
there
is
a
de
facto
connection,
yeah
and
Brendan.
Oh
and
Brendan
also
yeah
I.
Guess
we
have
a
representation,
but.
G
A
E
E
Sure
so,
sig
apps
on
the
kubernetes
side
has
long
done
both
the
ecosystem
around
applications,
as
well
as
things
like
the
workloads
controllers
and
other
efforts
there,
the
saiga
application
or
whatever
it
ends
up
getting
naming
they
haven't
figured
out.
The
final
names
for
them
is
kind
of
going
to
deal
with
that
ecosystem
area.
Looking
at
things
like
where
the
gaps
are,
what
are
all
the
projects
across
the
system
what's
going
on
and
we
haven't
really
figured
out
how
we're
going
to
interface?
E
Maybe
there's
things
we
don't
do
in
kubernetes
see
gaps
because
it's
happening
in
the
other
place,
but
it
is
one
of
those
things
we
put
on
the
roadmap
for
it
and
I
do
realize.
This
is
in
part,
my
fault,
because
I
helped
write
up
this
doc
and
the
application
area
was
one
of
those
things
that
Alexis
and
some
others
had
in
mind
right
off
the
get-go,
but
we're
going
to
try
and
figure
out
what
are
the
new
roles
and
responsibilities.
E
C
A
A
So
you
know,
because
there
are,
the
landscape
is
really
huge.
I
think
we
prioritize
filling
out
the
filling
the
gaps
before
you
know,
rather
than
focusing
on
a
lot
of
projects
in
a
specific
area,
but
yeah.
Definitely
it
just
a
single
we're,
not
looking
for
just
a
single
project
in
each
area,
necessarily
if
other
projects
may
be
promising
or
may
make
sense
right.
A
General
topic,
this
is
being
discussed,
is
right
now,
not
only
is
there
not
that
much
interaction
between
projects,
but
there
are
there's
not
much
interaction
between
different
parts
of
the
scene,
CF
right
between
the
GB
and
that
you
see,
or
the
TOC
and
user
community
and
part
of
that
was
just
by
design,
but
it's
kind
of
gone
overboard,
so
we're
trying
to
course-correct
that
a
little
bit
and
see
if
we
can
get
more
cross
foundational
interaction
like
the
projects
interact,
almost
not
at
all
with
TOC,
is
just
yet
another
example.
I.
C
Do
think
it's
important
to
recognize
that
you
know
the
TOC
can
provide
services
into
the
projects
the
TOC
doesn't
like
can't
arm
twist
the
projects
to
do
anything,
and
so
you
know
part
of
this
sort
of
cross-pollination
across
projects
is
really
one
of
creating
opportunities
versus
norms
and
I.
Think
the
assumption
was
that
a
lot
of
this
stuff
would
happen
organically,
but
I
think
there
might
be
opportunities
for
us
to
sort
of
Brian
make
the
magic
happen.
C
A
Top-Down
governance,
I
would
say
there
are
a
few
exceptions
to
that,
but
yeah.
So
there's
the
principles.
Doc,
there's
also
the
graduation
criteria,
doc
somewhat
relevant,
but
yeah
I
mean
it's
still
pretty
nascent
like
I,
don't
think
any
of
the
there
any
SIG's
have
been
officially
formed
or
had
meetings
or
anything
yet
so
we'll
keep
an
eye
on
that
and
some
of
us
who
you
know
I,
see
four
of
us
on
the
call
to
cross
on
the
TOC.
A
C
C
There
are
places
where,
where
there
is
that
level
of
support,
I'm,
wondering
and
I,
you
know
if
we
should
maybe
have
like
a
regular,
cadence,
monthly
or
quarterly,
where
we
review
those
things
actually
figure
out
and
actually
look
at
the
scope
of
this
and
figure
out
whether
whether
we
think
that
we're
using
those
resources
effectively
or
whether
there's
opportunities
to
to
double
down
or
expand,
where
we're
actually
engaging
with
the
CNC
F
for
help.
What
do
what
do
you
all
think
about
that?
Yeah.
A
I
A
A
But
if
something's
not
on
the
list-
and
we
ask
for
it-
and
you
know
they're-
probably
added
to
the
list,
so
the
other
projects
know
that
that's
a
thing
and
then
other
projects
also
have
regular
meetings
with
CNCs
staff
to
discuss
how
things
are
going
and
what
seems
you
have
my
views
so
far.
We
have
not
taken
advantage
of
that.
I
didn't
even
know
about
it
for
more
than
a
year
that
it
was
going
on,
there
is
actually
a
calendar
with
these
meetings
in
it.
A
D
I
It'd
be
nice
if
we
had
a
backlog
that
describes
some
of
our
asks
from
the
communities
project,
the
CN
CF
and
timeliness
like,
for
example,
the
performance
effort
is
one
the
what
we're
gonna
do
with
the
case
in
four
working
group.
Ideally,
we've
talked
about
handing
this
off
to
the
CN
CF,
but
we've
never
actually
had
any
fun.
So
I
think
it's
kind
of
on
our
shoulders
to
actually
make
sure
that
we
have
some
level
of
tracking
and.
I
D
G
D
C
A
C
G
G
A
C
A
A
Aaron
is
not
on
the
call
he's
kind
of
been
shepherding.
Those
contractors
in
terms
of
handing
off
I,
don't
think
we're
other
than
an
enforcement.
We're
not
really
handing
off
anything
to
anybody.
So
I'm
not
really
sure
what
that's
about,
but
for
sure.
If
the
contractors
are
not
adding
value,
we
should
not
continue
with
that,
but
since
Aaron's
not
on
I
suggest
we
don't
discuss
that
specific
issue.
Yeah,
that's
not.
F
We
have
a
GK
cluster
where
we
are
running
a
few
things
now,
then
we
are
trying
to
set
up
staging
repositories
for
artifacts,
then
a
few
other
we
already
work
with
DNS
stuff
has
moved
over
so
its
DNS
GCR
yeah.
Those
are
the
three
major
things
that
we
are
working
on
right
now,
but
over
a
period
of
time
we
want
to
make
sure
that
people
like
Club
providers
and
cluster
API
providers,
they
will
be
able
to
publish
their
artifacts
and
things
like
that.
F
I
Just
worried
about
like
for
me,
the
CCF
is
spending
money
on
the
separate
project
that
there
is
a
responsible
party
at
the
end
to
make
sure
that
there's
success.
I
know
that
there
are
folks
inside
the
communities
community
that
I
highly
trust
that
will
do
a
good
job
at
the
other
end,
who
are
the
actual
owners
that
are
actually
going
to
be
owning
this
process
on
the
other
side
and
are
they
spun
up?
I
F
F
Right
them,
so
the
the
main
person
that
we
work
with
today
is
a
a
comes
to
all
the
informations
and
helps
with
the
billing
stuff.
Looking
at
the
billing
reports
and
things
like
that,
so
he's
our
main
contact.
So
on
the
AWS
side,
it
is
currently
stuck
with.
We
we
we
I
choked
the
throat
of
evil,
but
then
it's
actually
with
the
AWS
folks.
At
this
point,
the
ball
is
in
their
court.
Yes,
it
is
a
sad
situation.
I
I
think
my
primary
reason
of
bringing
it
up
is
that
there
needs
to
be
an
accountable
person
that
actually
is
going
to
own
the
maintenance
of
this
infrastructure
and
possibly
report
back
to
this
group
periodically
or
to
report
back
to
the
other
testing
SIG's
on
the
state
of
things,
and
that
visibility
is
zero.
To
me
right.
G
Could
also
ask
to
have
a
project
meeting
with
them,
as
they
do
with
other
groups
with
the
steering
committee,
and
then
it
could
be
either
a
portion
of
one
of
these
meetings
or
an
initial
meeting.
I
think
that's
I
think
that's
the
right
way
to
start,
and
then
we
go
from
there
to
make
to
hold
them
accountable.
Just
as
we
hold
ourselves
accountable,
I.
G
Want
to
make
sure
that
we
leave
time
for
the
community
questions
that
are
in
here,
because
there's
at
least
one
on
the
slack
update
and
then
I
drop,
something
that
is
not
pressing
but
a
return
discussion
from
last
time
on
the
calendar
on
that
agenda.
But
you
have
to
go
first,
no
I'm
going
to
do
the
community
stuff.
Okay,.
D
So
I
just
wanted
to
give
everybody
an
update
on
slack
and
quick
context
for
a
lot
of
folks
on
the
call
that
may
not
have
context
not
this
past
Sunday,
but
the
Sunday
before
that.
Excuse
me
if
I
am
I'm
already
four
my
dates
around
8:00
8:30
p.m.
Pacific
time
Sunday
evening
we
had
several
bad
actors.
Join
slack,
potentially
use
a
vulnerability
that
slack
has
against
us
to
display
over
75
porn
images
and
harass
many
folks
take
on
their
personalities
and
trust
them
in
direct
messages.
D
D
Do
we
have
roughly
60,000
members
on
slack
if
we,
if
we
know
that
all
26,000
contributors
to
kubernetes
are
on
slack,
that's
still
1/3
of
that
population
is
contributors,
so
we
have
tons
of
consumers
that
take
advantage
of
of
slack
the
slack
administrators
are
working
diligently
around-the-clock
to
attempt
to
achieve
quorum
around
what
to
do.
Next.
We
are
taking
the
ultra-conservative
approach,
which
also
some
admins
disagree
with
to
keep
it
down
until
we
have
heard
from
slack
that
the
potential
vulnerability
is
not
a
vulnerability.
D
D
For
that
folder
I
can
tell
you
that
I
feel
like
we
are
officially
out
of
scope.
I
think
that
we
should
take
this
to
the
TOC
and
the
end
user
committee
to
have
a
broader
conversation
of
whose
role
is
this
and
things
along
those
lines
again.
I
personally
feel
that
our
are
being
the
moderators
and
contributors
that
we
have
that
are
doing.
This
role
could
be,
could
have
their
time
spent
in
other,
more
pressing
community
matters.
So
I
feel
like
at
this
point
in
time.
A
D
D
Worries.
Thank
you.
Thank
you.
Tim
there
was
ton,
there's
tons
of
people
that
are
working
on
this
right
now,
trying
to
figure
it
out.
So
many
thanks
to
everybody,
that's
banging
their
heads
against
wolves,
we
are
I
mean
I
can
definitely
publicly
make
the
comment
that
slack
does
not
support
open
communities
that
is
not
open,
source,
open
communities,
meaning
that
there
is
no
manual
addition
of
members
that
you
have
a
slack
inviter.
That
is
the
I
think
misconception
that
people
have
with
slack
and
open
source.
They
only
support.
D
D
A
G
So
there's
a
number
of
challenges
in
in
that
one
it
doesn't
fix
the
moderation
problem,
it
doesn't
fix
the
exploit
problem.
It
does
mean
that
we
have
to
forklift
60,000
people
of
whom
we
will
lose
a
number
of
them,
because
slack
is
becoming
more
and
more
more
and
more
part
of
people's
daily.
You
know
work
environment,
so
we
have
talked
about
this
in
the
past
when
it
was
a
much
smaller
community
yeah
on
slack
and
it's
we
come
to
the
conclusion
that
wasn't
really
viable
in
that
way,
it
didn't
fix
the
problems
we
wanted
to.
G
D
Have
reaction
of
a
planning
doctor
we
plan
to
share
in
the
next
I'd,
say
24
to
48
hours
that
lists
all
our
options
and
things
along
those
lines.
I
mean
I,
think
are
perfect
or
perfect
case
scenario
at
this
point
is
force
lock
to
come
back
and
say:
hey
everything's
great,
and
then
we
increase
our
moderation
team
and
and
then
involve
the
consumer
community
as
well.
I
feel
like
that's
our
perfect
scenario,.
D
C
C
Moving
something
like
to
the
discourse
that
contribute
access
set
up
a
while
ago
might
be
a
good
good
way
to
go.
That
is
much
more
tailored
towards
more
open
communities
with
with
the
right
tooling
around
that
it
also
is
probably
more
searchable
more
threaded
more
on
topic
for
that
type
of
discussion,
Stack
Overflow
also
so
I
think
we
can
shunt
some
of
the
activity
happening
in
slack
into
other
other
venues.
I
do
want
to
actually
say
that,
like
like
George
yeah,
so
George
posted
this
tweet
I'm.
D
C
F
D
D
D
I
mean
they,
the
OSS
solutions.
Do
they
don't
solve
for
everything,
but
they
do
like
the
discourse
auction.
It
solves
her
blocking
at
IP
level,
where
slack
only
allows
for
blocking
at
the
user
level,
which
means
that
they
could
respond
with
another
email
address
or
something
along
those
lines.
But
I
think
discourse
is
regularly
fighting
security
in
moderation
battles,
whereas
slack
does
not,
because
that's
not
their
use
case.
I,
don't
know,
maybe
if
it
is
now
considering,
10
plus
slack
workgroups
have
been
hit
with
the
same
thing.
So
you
know
TBD.
G
A
C
D
C
Those
auditing
the
class
there
was
some
concern
of,
like
you,
know,
cuz,
because
usually
we're
pretty
informal
about
the
steering
committee
meetings.
What
would
it
be
like
with
a
larger
audience,
and
so
there
was
a
little
bit
of
trepidation
going
in,
but
we're
like
you
know
how
bad
could
it
be
and
it
turns
out
things
turned
out
well,
so
yeah.