►
From YouTube: Kubernetes Community Meeting 20180510
Description
See this page for more information! https://github.com/kubernetes/community/blob/master/events/community-meeting.md
A
Hello,
everybody
and
welcome
kubernetes
community
call,
I'm,
Tim,
pepper,
I'm,
an
engineer
at
VMware
and
I'm.
The
moderator
for
today,
as
is
normal
we've,
got
a
rotating
set
of
volunteers,
moderating
taking
notes
and
presenting
and
if
you'd
like
to
contribute,
I
suggest
you
reach
out
to
Jorge
Castro
and
volunteer.
A
Also,
as
is
the
norm
in
our
community,
this
call
is
being
recorded.
So
remember
what
you
do
and
say
here
is
going
to
be
on
the
record
on
the
internet
for
all
time.
Please
mute
your
microphone
if
you're,
not
speaking
and
just
be
cured,
yet
courteous
to
other
speakers
for
those
who
are
presenting
or
if
you
you
have
cause
to
ask
a
question
or
things
start
getting
interactive
I
just
ask
that
folks
be
mindful
of
the
overall
agenda
and
time
slots.
A
We
obviously
want
a
collaborative
meeting,
but
also
need
to
make
sure
we
cover
the
things
that
we're
intending
two
minutes
are
being
kept
in
a
Google
Doc.
We've
got
people
scribing
and
feel
free
to
collaborate
that
and
then
afterwards
the
recording
will
be
posted
to
the
YouTube
channel.
So
with
that,
I
would
like
to
hand
it
over
to
Richard
who's
going
to
be
doing
an
ambassador,
API
gateway
demo.
Today
and
Richard.
We
can
see
your
screen
now.
All.
B
Right
all
right,
thank
you,
Tim,
so,
and
thank
you,
everyone
for
joining
so
I'm
here
to
talk
about
the
Ambassador
API
gateway.
It's
an
open
source,
API
gateway,
Apache
Software,
license.
We've
built
this
for
kubernetes
only
it
doesn't
run
outside
of
kubernetes
and,
as
a
consequence,
its
architecture
is
very,
very
simple.
So
we
don't
have
a
database.
B
Compliance
is
that
ambassadors
actually
built
on
the
Envoy
proxy,
so
the
Envoy
proxy,
if
you're
not
familiar
with
it,
is
an
open-source
layer,
7
proxy,
originally
built
by
the
team
at
lyft,
and
it's
now
used
inside
ambassador
insight
is
do
lots
of
other
organizations
are
actually
using
envoy.
It's
got
a
really
amazing
upstream
community,
similar
to
kubernetes.
It's
very
open
lots
of
discussion,
lots
of
collaboration
and
I.
Think
for
the
purpose
of
this
conversation.
The
main
thing
I've
wanted
to
point
out
about
envoy
is
that
it's
really
designed
for
machine
configuration.
B
It
has
a
very
sophisticated
configuration
file
format.
A
lot
of
the
configuration
itself
is
also
designed
to
be
dynamic,
so
they
also
exposed
a
lot
of
that
functionality
through
gr,
pc-based
api's,
and
so,
if
you
really
want
to
deploy
and
manage
envoy
and
a
particular
configuration
you
actually
more
or
less
end
up
writing
some
code,
and
so
the
reason
why
we
wrote
ambassador
is
to
obviate
the
need
for
writing
some
of
that
code,
and
so
the
ambassador
architecture
is
very
simple.
You
we
basically
talk
to
the
kubernetes
api.
B
We
get
notified
when
their
configuration
changes
and
we
actually
operate
as
a
sidecar
to
envoy
and
so
every
time
there's
a
configuration
change.
We
compute
the
necessary
changes
to
envoy
we
configure
envoy
appropriately.
We
handle
things
like
connection
draining
and
restarting
envoy
if
necessary,
and
the
net
result
is
all
of
your
traffic
actually
flows
through
on
voice.
So
you
do
get
sort
of
on
raw
envoy
performance
and
sort
of
the
management
layer.
If
you
will
of
envoy
is,
is
done
by
ambassador.
B
So
with
that,
as
your
the
the
quick
overview,
I'm
gonna
actually
show
a
simple,
a
couple,
quick
examples.
So
right
here
you
see
the
yeah
mo
for
deploying
ambassador
I've,
already
deployed
ambassador
in
a
kubernetes
cluster,
and
you
can
see
this
is
just
a
standard
kubernetes
service.
We
create
roles,
cluster
role,
binding
and
then
the
part
of
it
is
just
the
deployment
where
you
can
specify
replicas,
and
we
give
you
a
docker
image
that
you
can
deploy
and
we
provide
liveness
and
readiness
probes
by
default.
B
So
what
I'm
going
to
do
is
I'm
gonna
actually
apply
a
route
to
the
HTTP
bin
service,
and
if
you
look
at
my
configuration,
it's
very
simple:
it's
just
a
curve
in
any
service.
It
doesn't
actually
go
anywhere.
Interesting,
kubernetes
and
I
create
a
mapping,
object
with
the
name,
a
prefix
and
have
it
go
to
a
particular
destination,
and
in
this
case
we
support
a
whole
bunch
of
additional
annotations.
B
On
top
of
your
core
mapping
like
host
rewrite,
where
we
override
your
default
HTTP
host
parameters
to
go
to
HT
bin
org,
and
so
now
I
can
test
this
out
by
doing
a
curl,
and
you
should
see
that
I
get
the
IP
address.
So
now
my
service
is
actually
talking
to
Stacey
bid
service.
Ambassador
includes
integrated
diagnostics,
so
we
show
a
UI
of
entire
routing
table.
You
can
see
here.
B
Here's
the
HTTP
bin
service
is
going,
HC
been
org,
and
we
do
this
so
that
it's
actually
easy
to
actually
troubleshoot,
and
so
we
show
all
the
configuration
that
we've
generated.
So
this
is
how
we
support
people
who
are
trying
to
troubleshoot
their
configuration
of
ambassador
I'm,
going
to
show
one
other
feature
of
Ambassador
here.
B
B
You'll
see
that
actually,
my
if
this
terminal
actually
opens
oh
here,
we
go
so
you'll
see
that
on
the
right
hand,
side
here,
I'm
actually
getting
live
traffic
and
I've
colorized
it.
So
they
all
say
anonymous
and
you'll
have
to
trust
me
here,
because
I'm
not
gonna,
just
let
the
scroll
by
but
you're,
basically
seeing
the
same
tweets
that
are
coming
in
on
production
as
they
are
on
the
shadow
service
except
the
shadow
services.
Reading
different
codes,
you're
actually
able
to
test
that
and
then
with
Ambassador.
B
We
also
expose
all
the
stats
through
stats
D,
and
so
you
can
actually
see
if
I
hit
refresh
here,
that
the
latency
of
the
shadow
service,
because
I
edit,
that
sleep
is
actually
quite
a
bit
higher
than
that
of
my
my
p99
latency,
is
actually
much
higher
than
that
of
my
production
stream
service.
So
so
this
is
sort
of
the
general
overview
of
Ambassador,
so
metrics,
monitoring
and
shadowing,
and
we
support
rate
limiting
and
authentication
and
sort
of
all
the
general
features.
B
A
Otherwise
definitely
appreciate
the
the
demo,
the
information
Richard
and
if
you
could
link
the
slides
that
you'd
present
to
their
into
the
the
meeting
minutes
that
would
be
really
appreciated
and
then
for
others.
Following
along
in
the
meeting
minutes,
there
are
also
links
off
to
the
project
pages
and
further
information
as
well.
A
C
So
this
is
week
six
I
believe
of
111
the,
and
at
this
point
we
are
approaching
two
important
deadlines.
One
is
beta
zero
and
the
other
two
days
after.
That
is
the
evaluation
point
where
we
see
whether
or
not
we
can
shorten
code
freeze
the
I
so
as
such
right
now,
what
we
are
focused
entirely
on
is
fixing
some
of
these
test,
fails
on
into
n
tests
and
and
upgrade
tests
and
scalability
tests.
So
I'm
going
to
turn
this
over
to
ice.
E
D
So,
to
give
an
overview
early,
when
we
started
this
week,
we
can
kinda
we're
in
a
good
space
where
the
master
blocking
we
have.
We
are
monitoring
like
two
Suites
right
now.
One
is
the
master
blocking
and
the
master
upgrade.
As
Josh
was
saying
to
begin
the
week
we
were
like
pretty
stable,
and
then
we
had
a
very
busy
week,
we're
all
JK
I'm
fired
so
and
then
we
were
also
like
saying
a
bunch
of
upgrade
issues.
D
So
during
the
week
we've
been
actively
working
with
the
different
sakes
and
thanks
to
the
different
various
members
in
the
release
group
itself,
wavelike
divided
the
responsibilities
and
we're
focusing
on
each
of
us
are
focusing
on
different
six
to
make
sure
that
you
know
we
have
this
test
of
test
failures,
investigated
and
hopefully
fix
before
beta,
which
is
on
the
15th.
The
good
news
is
now
things
are
again
stabilizing,
as
you
can
see,
master
blocking
is
kind
of
almost
green.
D
We
have
a
few
storage
issues
there,
for
which
the
fixes
are
in
flight,
and
we
have
a
couple
of
I
mean
with
one
scalability
issue,
for
which
again
both
Shawn
and
Jordan.
They
are
working
on
a
PR
to
address
this,
so
I
did
see
a
couple
of
cube
admin
failures
crop
up
this
morning
and
yet
to
investigate
that
and
follow
up
on
that.
But
overall
this
looks
kind
of
stable
and
I'm
hopeful
that
we
can
hold
this
up
until
meter.
Upgrade
is
a
little
bit
more.
D
D
Okay,
I
have
this
pretty
I'll
share
this
link
to
this
tracker
out
in
the
meeting
notes.
So
currently,
these
are
the
top
fail,
fail
failures
that
we
are
tracking
and
I
also
have
status
on
where
they
are
and
who
is
the
point
of
contact
and
if
we
have
PRS
and
what's
going
on
so
good
news
is
all
of
this
on
progress.
People
are
working
on
PRS
I'm
hopeful
that
it
could
land
by
the
end
of
the
week,
or
at
least
by
Tuesday.
D
There
are
a
couple
here
that
we
haven't
seen
any
traction
on
so
I'll
be
following
with
up
with
them
closely
today.
As
for
the
upgrade
tests,
there
is
a
we'll
issue,
umbrella
issue
that
Tim
and
Tim's,
and
all
of
them
are
actively
following
upon.
So
we
are
hopeful
that
we
can
get
to
that
greatness
also
during
this
week.
So
if
anybody
is
interested,
please
follow
along
in
this
tracker,
so
yeah.
So
that's
the
main
update
josh
am
I
missing
anything.
C
D
It's
been
really,
it's
been
looking
up
since
yesterday,
so
we
are
getting
and
also
following
up
on
the
issues
there.
They
are
there's
more
frequent
status,
updates
and
also
links
to
PRS
that
might
resolve
the
issues.
The
ones
that
I've
marked
in
red
are
the
ones
that
we'll
have
to
follow
up
little
closely,
which
I
plan
to
do
today.
What
is
the
Signet
work
specifically
on
the
networking?
Granular
checks,
almost
all
state
tests
there
seem
to
be
failing,
so
I'll
have
to
follow
up
on
that.
C
C
Well,
thank
you.
So
I
will
update
everyone
on
the
kubernetes
development
mailing
list
on
Monday,
with
what
our
statuses
are
knows
and
whether
or
not
beta0
is
going
to
be
going
out
on
time,
because
we're
not
going
to
be
putting
out
a
beta
if,
if
that
status
doesn't
improve,
but
it
looks
like
it
is
going
to
improve.
C
A
C
D
C
D
A
A
F
That
says
the
describes
our
current
proposal
process
or
our
pre-existing
proposal
process,
so
we're
just
trying
to
formalize
that
provide
more
structure
and
make
it
more
obvious
to
people
what
they
should
be
doing
like
what
should
have
a
formal
proposal
which
is
important
to
include
in
a
proposal
how
you,
importantly,
how
do
you
actually
bring
a
proposal
to
closure?
So
you
know
you
can
move
ahead
with
the
implementation
things
like
that.
F
We
also
have
been
discussing
an
API,
a
review
process
formalizing
that
we've
had
about
five
people
who
have
traditionally
done
a
large
fraction
of
all
the
API
reviews.
So
it's
been
informal
in
terms
of
how
to
do
review,
what
to
look
for
things
like
that,
so
we
want
to
formalize
that
we
do
that
have
office
hours.
So,
if
you
have
questions
about,
is
this
the
right
way
to
do
something?
Or
does
this
follow
our
API
convention?
Or
you
know,
I,
have
this
new
kind
of
extension
mechanism
I
want
to
add
the
data?
F
Okay,
definitely
sign
up
for
the
Sudanese
fig
architecture
and
office
hours
there
on
the
meeting
any
office
hours
on
Thursday
mornings
at
8:30
Pacific.
You
can
go
to
the
community
repo
to
find
our
information
about
the
meeting
times.
Calendar
zoom
agenda
dock
and
all
those
things
definitely
make
sure
you're
in
the
agenda
Docs,
because
we
do
use
it
and
we
have
a
slack
channel
if
you
have
questions
Anna
mailing
lists.
F
Of
course,
the
lately
we've
been
working
on
our
Charter
trying
to
outline
these
areas
that
we
work
on
and
to
kind
of
more
formal
convert
them
to
more
formal
sub
projects.
Today
we
had
a
big
discussion
on
improving
the
conformance
tests
and
other
major
things
happening
in
the
you
know.
Major
changes
to
kubernetes
or
big
new
architectural
changes.
Tick
architecture
is
the
place
where
those
get
discussed.
So
we
have
a
small
crew
that
attends
regularly.
But
if
there's
something
that
is
important,
that
you
want
to
learn
about,
definitely
feel
free
to
come.
That's
about
it.
E
G
Everyone
I
did
in
the
agenda
linked
to
the
slide
deck
that
we
created
last
week
for
the
contributor
experience
update.
That
will
have
excuse
me
that
will
have
much
more
detailed
information
as
to
things
like
our
full
sub
project
list,
and
things
like
that.
First
thing:
I
wanted
to
say
that
we
do
have
a
new
contributor
site
underway.
This
will
be
something
along
the
lines
of
contributor
to
expert
any
thought.
I
of
something
like
that
built
with
Hugo.
The
link
to
the
cap
is
in
the
agenda.
G
If
you
have
anything
that
you'd
like
to
add
or
share
or
opinions
there,
please
feel
free
to
add
that
development
is
underway.
Right
now
we
have
just
gone
up
an
instance
of
discourse,
which
is
a
forum,
that's
100%
open-source
and
we're
ready
to
test
it.
It's
discussed
Kerber,
Nettie's
I/o,
so
please
post
any
content
that
you
have
their
announcements,
need
up
reminders
or
just
introduce
yourself
to
see.
G
If
this
will
work
her
for
our
community,
we
are
looking
for
mentors
now
we
have
roughly
six
different
mentoring
programs
all
built
with
time
in
mind
and
now
we're
looking
for
mentors.
One
of
those
is
our
meet
our
contributors,
monthly
YouTube
series
link
in
the
agenda
to
that.
So
you
can
see
what
we're
talking
about,
but
this
is
roughly
four
to
six
contributor
mentor
types
who
come
on
and
ask
her
the
answer,
questions
that
have
to
do
with
things
like.
G
Why
is
my
why's
my
test,
making
all
the
way
to
why
they
got
into
open
source
or
kubernetes
how
to
be
an
approver?
Just
very
general
questions
like
that
about
someone's
growth
here
within
within
kubernetes.
As
a
contributor,
we
are
going
to
be
doing
a
contributor
experience
survey,
that's
going
to
go
out
in
June.
G
This
will
cover
everything
from
our
communication
platforms
and
your
experience
with
them
all
the
way
to
just
how
you
like
the
workflow
in
the
automation
within
github,
and
how
we
can
possibly
help
and
tailor
that
from
our
deep
dive
at
cube
con
Copenhagen,
we
discuss
the
developer
guide,
and
now
we
are
officially
underway
with
a
new
developer
guide.
The
developer
guide.
G
Right
now,
we
do
have
a
very
lengthy
and
robust
Abell
repo
on
the
community,
repo,
not
free
from
sorry
folder
on
the
community
repo
and
we're
now
just
trying
to
surface
that
and
make
all
the
docs
in
there
up
to
date
and
have
additional
information
that's
relevant
to
now
with
the
contributor
site.
We
are
really
hoping
that
a
lot
of
deep
discoverability
lows
will
be
taken
care
of
there.
G
A
All
right,
Thank,
You
Paris.
We
did
have
one
question
in
the
chat,
folks
interested
and
contributing
to
the
new
contributor
site
and
the
where
to
go,
and
the
answer
is
reach
out
to
the
contributor
experience
sig
on
slack
and
there
I
there's
a.
Is
there
an
issue
associated
with
a
kept
Paris
issue?
Number
yeah
yep.
G
A
G
A
All
right,
thank
you,
Chris
hand
up
or
adjust
the
headphones.
Chris
short
headphones.
Okay,
just
checking
so
next
on
the
agenda
of
in
is
announcements.
So
we
have
our
usual
shout
outs.
There
is
a
channel
and
slack
hash,
shout
outs!
If
there's
anybody
that
you
would
ever
like
to
give
a
shout
out
to
that's
the
place
to
do
it,
and
then
we
roll
them
up
in
the
community
meeting
here.
A
So
recent
ones
a
has
sent
a
thanks
out
to
is
it
Justin
Magnus,
and
that
was
on
his
work
as
release
shadow
for
the
features
with
a
and
as
mentioned
earlier,
Josh
Burke.
Is
this
big
shout
out
to
ice
for
all
the
work
she's
doing
on
CI
signal
lead
for
the
111
release
team
as
well,
one
for
me,
Aaron,
Creek
and
Berger,
who
just
recently
announced
he's
gonna,
take
a
little
bit
of
time
off
and
get
some
R&R
and
rest
and
relaxation
well
deserved.
A
So
just
a
shout
out
and
thank
you
for
all
the
leadership
you've
been
giving
to
the
project
lately.
Thank
You
Chuck
has
a
shout
out
to
the
dock
team
notes
that
working
on
the
website
is
such
a
good
experience.
Now
that
it's
on
Hyuga
pay
tree
rebuild
time
went
from
about
20
seconds
to
60
milliseconds
and
a
heart,
emoji
and
Jason.
The
timer
would
like
to
think
Lika
pili
aka
at
stealthy
box
on
slack
and
or
github
for
the
hard
working
long
hours,
helping
to
fix,
cube
ATM
upgrade
issues.
A
A
second
shout
out
in
a
row
this
week
for
Leigh
and
I
would
echo
that
shout
out
I
think
maybe
on
behalf
of
I,
should
probably
Tim,
st.
Claire
and
others.
This
this
upgrade
set
of
issues
has
been
a
persistent
problem
just
lately,
and
it's
really
great
to
see
some
progress
on
that
positive
progress
shout
out
from
Jorge
Castro
and
Parrish
Pittman
to
Vanessa
Herrick
and
the
rest
of
the
cnc
of
linux
foundation,
personnel
that
helped
pull
off
a
great
contributor
summit
and
cube
con.
A
I
I
know
I
can
speak
from
my
experience
there
that
it
was
a
great
week.
So,
thank
you.
There
shoutouts
to
our
top
staff
overflow
users
in
the
last
little
while
we
have
Anton
cos
stinko
Nikola,
Ben,
Marv,
Tuen,
Jonah
Benton.
So
thank
you
all
they're
very
much.
That's
a
great
way
to
support
the
community
helping
answer
questions.
A
A
A
E
Have
quite
a
backlog,
there's
a
little
bit
of
chatter
in
cig
docs,
the
cigars
channel
right
now
about
the
fact
that
you
know
everybody's
been
conferencing.
This
week
has
been
right:
the
docs
and
most
of
the
docs
maintainer
zorak
right
the
docs,
and
then
we
were
at
a
Doc's
summit
yesterday.
So
we
are
getting
back
on
triage
and
rolling
through
the
PRS
and
issues
as
quickly
as
we
can.
But
there
will
be
a
little
bit
of
lag
here.
A
A
H
H
The
meetings
earlier
and
being
more
efficient
of
may
be
either
alternating
in
a
section
with
with
the
graph
of
the
week
where
we
would
discuss
a
kept
of
the
week
or
if
there
was
a
way
that
we
could
service
more
of
the
things
that
are
happening
in
cups
of
the
community
meeting
to
kind
of
foster
communication.
So
if
you
feel
strongly
about
that
plus
one
minus
one
in
the
sidebar
or
if
anyone
has
any
comments
with
regard
to
that,
especially
since
some
of
these
cups
might
be
touching
across.
F
So
one
thing
I
would
recommend
is
if
you
create
a
cup
or
you
recognize
a
cup
that
you
think
is
important
email,
the
regular
irrelevant
mailing
lists
and
possibly
cigars
pictures
as
well.
Every
cap
is
supposed
to
list
sig
that
owns
it
and
participating
SIG's.
So
if
you
believe
those
are
the
right,
things
send
a
link
to
the
pr2.
Those
mailing
lists
give
notifications
for
people
who
probably
should
be
looking
at
them
are
not
as
useful.
F
A
C
Think
it
would
be
worth
trying
I
know
there
are
some
who've
been
tuning
these
community
meetings
for
years
and
years
and
talk
about
how
back
in
the
old
day
old
days,
it
was
like?
We
were
actually
talking
about
real
heavy,
concrete
design
and
stuff,
and
so
this
could
be
a
throwback
to
that
sort
of
thing.
C
My
my
only
like
caveat
is
that
this
seems
like
a
symptom
that
this
is
a
really
high
visibility
place
and
people
are
trying
to
get
their
work
to
be
shown
in
a
high
visibility,
place
and
I'm
not
entirely
sure
what
the
motivation
is.
Some
thought
that
maybe,
if
you
don't
have
a
high
enough
visibility
place
for
caps,
which
I
thought
was
sort
of
part
of
the
regions
were
trying
to
create
more
of
a
contributor
site
where
we
could
have
like
a
database
of
caps
or
something
more
crowd
liked
to
give
them
more
attention.
C
So
I
would
view
having
a
regular
rotation
of
talking
about
caps
more
as
a
bootstrap
process
until
we
find
sort
of
the
longer
term
sustainable
way
and
a
cap
is
a
kubernetes
enhancement
proposal,
so
I'd,
like
I
literally
just
had
somebody
talked
to
me
one-on-one
this
morning
about
like
hey.
What's
the
like,
can
I
map
out
the
lifecycle
of
a
design
proposal
and
that's
literally,
what
I
kept
this
supposed
to
be
and
we're
in
this
weird,
the
in-between
state
I
think
where
people
don't
necessarily
have
to
use
caps.
C
So
we
can
try
and
keep
track
of
this
to
avoid
the
problem
we've
had
over
the
past
couple
of
years,
where
there
are
lots
of
design
proposals
in
the
design
proposals
directory
and
it's
really
unclear
which
of
those
have
actually
been
implemented
and
to
what
degree
they've
been
implemented
or
if
it's
still
just
a
catch-all
or
like
good
ideas,
so
I
would
yeah.
Thank
you
so
would
expend
the
awesome.
It
still
has
well
alright.
F
H
A
look-
oh
I
could
do
that.
That's
that's
an
easy
one
and
then
kept
five
is
the
cap
that
will
allow
us
to
index
and
search
caps,
and
things
like
that
because,
right
now,
if
you
try
to
search
for
kept
something
it's
impossible,
you
almost
have
to
go
into
github
and
know
the
name
of
the
cat,
so
we're
working
on
making
those
index
of
all
you
should
be
able
to
see
which
ones
are
open,
which
ones
have
been
implemented.
All
that
good
stuff,
so
I
will
okay.
A
I
I
It
took
folks
five,
ten
minutes,
depending
on
how
many
things
they're
using
this
was
all
about
running
applications
within
kubernetes,
looking
at
tools
that
the
kubernetes
project
owns
and
what
people
think
of
them,
the
parts
they
use
of
them
and
how
that
works
along
with
API,
such
as
the
workload
controllers
and
things
like
that,
and
so
the
raw
results
are
available
for
everybody,
and
the
link
is
in
the
meeting
minutes
today.
In
addition
to
that,
because
it
can
be
a
little
bit
hard
to
navigate,
we've
also
started
there's
a
blog
post
on
this
as
well.
I
I
We're
actually
folks
view
on
dealing
with
things
like
cluster
management,
in
addition
to
application
management,
which
is
slightly
different
yeah-
and
there
was
a
neat
thing
here-
that
77%
of
people
who
said
they
develop
applications
that
interact
with
the
kubernetes
api
also
are
operating
applications
in
kubernetes.
I
don't
know
if
that's
because
we've
got
a
lot
of
the
tool
developers
who
happen
to
take
the
survey
or
a
lot
of
people
just
have
to
build
their
own
workflows.
There's
some
more
insight
on
that
earlier.
I
But
we
got
insight
into
things
such
as
when
it
comes
to
core
controllers.
How
are
people
using
things
like
there's,
24
percent
of
folks
who
use
this,
are
actually
supplementing
the
core
controllers
with
their
own
controllers
and
that
I
think
that's
a
good
justification
for
some
of
the
work
that
folks
are
doing
with
meta
controllers
and
some
of
the
other
things
to
make
creating
custom
controllers
a
little
bit
easier.
I
But
there
are
a
lot
of
graphs
in
here
which
make
things
easy
to
use
like
what
are
folks
doing
with
dashboard
or
if
we
go
into
helm.
There's
a
lot
of
detail
on
helm
on
how
folks
are
using
it
or
the
different
charts
on
I'll
scroll
down
to
some
of
this.
We
asked
questions
in
there,
such
as
how
do
you
want
to
do
things
always
are
usually
sometimes
never
and
not
applicable,
and
so
do
you
want
to
be
able
to
visualize
the
objects
that
make
up
an
application.
Folks
usually
want
to
do
that
right.
I
I
want
to
be
able
to
visualize
the
kubernetes
parts
that
sometimes
make
up
an
application
right
or
there's
this
one.
This
one
got
the
most
always
I
want
to
be
able
to
find
the
version
of
an
application
I'm
running.
We
got
details
on
this.
That
is
meant
to
help
inform
us
as
we
craft
some
of
the
experience
and
the
details,
but
that's
all
available
in
here.
We
also
looked
at
the
kinds
of
tools
folks
we're
using.
So
we
asked
what
publicly
available
tools
and
we
listed
a
bunch
of
tools
that
people
are
using.
I
I
I
Go
poke
at
this
because
this
gives
insight
into
what
people
think
and
how
they're
using
it,
and
especially
into
some
of
the
free
text
spots
if
it's
a
project
you're
interested
in,
because
people
took
a
lot
of
time
to
give
us
feedback
and
there's
a
lot
of
good
little
nuggets
of
information
that
you
can't
just
put
in
a
chart.
But
if
you
just
sit
down
and
maybe
copy
a
column
and
paste
it
into
a
text,
editor
just
start
reading
the
responses
people
gave
there's
some
real
good
nuggets
of
information
in
there
with
that
I'll.
I
A
I
A
Alright,
thank
you
very
much.
Interesting
information
and
I
definitely
encourage
folks
to
go.
Look
at
that.
Read
it
one
of
the
things
that
really
makes
for
a
thriving
community
on
a
large
project
like
this
is
is
listing
to
your
stakeholders
and
users
across
various
persona.
So
getting
this
level
of
feedback
is
really
informative.
A
G
Okay,
miss
Paris
I,
put
that
on
there
for
them,
I,
don't
think
they
can
make
that.
But
yes,
if
anyone
is
interested
or
knows
folks
who
are
interested
that
have
JavaScript
and
or
go
experience,
they
are
looking
for
new
contributors
as
well
as
possible.
Maintainer
so
feel
free.
It's
either
dig
into
an
issue
or
to
reach
out
to
one
of
the
folks
listed
on
the
agenda.
As
they
said,
they
will
gladly
onboard
new
contributors.
C
Especially
in
the
context
of
the
survey
that
Matt
just
presented,
we
can
see
that,
like
30
percent
of
the
people,
31
percent
of
the
people
who
responded
said
they
used
kubernetes
dashboard.
So
it
can
certainly
use
some
love
and,
conversely,
they're,
like
I,
already
percent
of
the
people
who
don't
use
a
UI
of
any
sort.
So
the
majority
of
people
responding
to
that
survey
at
least
would
appreciate
a
functional
UI,
out-of-the-box
yep.
G
F
I
So
I
could
this
is
Matt
again
I'll
jump
in,
because
I
was
with
Parris
in
the
last
siggy
wine
meeting.
They,
the
response
was
that
they
are
trying
to
move
to
a
newer
version
of
angular
if
I
remember
it-it's,
on
angular,
1
and
they'd
like
to
get
that
updated,
and
so
that
is
blocking
a
little
bit
of
the
work,
because
they're
only
looking
to
bring
critical
fixes
in,
but
the
whole
idea
is
right.
I
Now
they
need
contributors
to
keep
this
thing
going
because
sig
UI
many
of
the
people
who
were
involved
in
developing
and
even
working
on
that
have
been
tasked
with
doing
other
things,
and
so
now
the
call
is.
Can
we
find
people
to
come
in
and
wrangle
this
and
get
this
moving,
and
that's
kind
of
the
hope
at
the
moment.
J
A
Anybody
else
interested
in
engaging
the
the
sig
has
meetings
mailing
lists,
jump
in
and
get
involved
alright,
unless
anybody
has
anything
else
that
they
would
like
to
attack
on
to
the
end
of
the
agenda
or
a
question
or
comment
to
bring
up.
We
have
reached
the
end
of
today's
agenda
and
we
could
go
ahead
and
give
all
of
you
a
little
bit
of
time
back
in
your
day.