►
From YouTube: 2021-07-29 Crossplane Community Meeting
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
All
right
recording
has
started,
and
this
is
the
july
29th
2021
cross,
plane
community
meeting
the
current
milestone
that
we're
working
on
now
is
1.4
and
we
are
sticking
to
our
typical
two-month
cadence,
where
the
next
expected
release
date
for
crossplane
will
be
the
end
of
august.
A
So
just
about
one
month
from
now
kind
of
halfway
through
the
milestone.
B
A
Cool-
let's
keep
rolling
here,
so
let's
go
ahead
and
bring
up
the
1.4
board
and
get
a
little
bit
of
status
on
some
of
the
recent
progress
here.
So
moaf.
Here's
a
official
informal
welcome
back
to
you
to
the
community.
We
missed
you
very
much.
It
is
now
publicly
known
that
we
all
missed
you,
but
you
you
already
got
started
back
up
on
like
the
cross
resource,
referencing
and
stuff,
like
that.
You
want
to
get
us
up
to
date
on
some
of
that
progress.
B
B
So
I
think
it
is
pretty
close
to
being
merged.
So
I'm
waiting
for
nick
to
take
another
look
which
will
resolve
both
issues
at
the
top
of
the
review,
progress,
column
and
the
the
last
two
are
related
to
ack
prs
that
I
had
and
they're
still
not
merged.
So
I
will
get
back
to
that
and,
like
you
know,
push
jay
to
take
another
look
and
possibly
merge
them.
A
Were
there
were
there
any
things
like
needed
from
our
side
on
those
pr's
while
you're
while
you're
out
or
is
it
like?
You
know
the
the
their
team
hasn't
gotten
a
chance
to
like
review
and
get
emerged.
B
A
Like
get
get
some
eyes
back
on
this
again
and
keep
those
moving
that'll
be
definitely
useful.
Let's
see
is
nick
here
nope,
I
don't.
I
don't
think
nick
is
here,
but
yeah
nick
has
has
a
design
doc
out
for
composition,
revisions
which
allow
you
to
update
compositions
with
new
versions
of
them
over
time
and
just.
A
Nick,
I
will
allow
you
to
speak
for
yourself
for
the
progress
on
the
composition.
Revisions
then.
D
Not
much
more
than
what
you
said,
there
is
a
design
doctor,
there's
also
a
partial
implementation,
but
I'm
not
going
to
go
any
further
until
I
get
some
review
and
we
we
ratified
the
design,
I
encourage
the
community
to
give
feedback
on
that
document.
Have
a
read
through.
I
did
elect
dan
and
move
off
as
the
approvers
for
that
doc.
So
they're,
the
ones
who
are
gonna,
have
the
final
say
on
on
whether
we
move
forward
or
not,
but
definitely
hear
what
the
community
thinks.
D
First,
as
jared
said,
if
I
I
recommend
I
recommend
having
a
read
through
the
pull
request,
it
is
related
to
a
very
long
issue.
That's
quite
old,
so
I
I
recommend,
reading
the
pull
request
rather
than
the
issue,
maybe
first
to
get
modern
context
for
this
problem
and
effectively
what
it
does
is
today.
If
you
update
a
composition,
every
composite
resource
that
uses
that
composition
reflects
those
updates
immediately.
D
A
Awesome
nick
and
I
will
go
ahead
and
add-
maybe
a
lick
here.
A
One
pager
for
a
little
bit
more
visibility
on
on
feedback
there.
A
Okay,
that's
awesome!
So
a
couple
good
things
really
good
things
in
progress
here
dan.
Do
you
want
to
kind
of
get
us
up
to
date
with
some
of
the
things
that
you
had
been
driving
as
well,
too?.
C
Yeah
so
the
introduce,
let's
see
what
was
the
official
name.
Let's
see,
exposing
poll
interval
and
external
client
kind
of
the
first
iteration
of
that
was
just
giving
a
flag
on
the
provider
in
general,
so
this
has
been
implemented
across
a
number
of
providers.
C
I
think
it's
still
missing
in
azure,
so
I
can
make
sure
to
get
that
updated,
but
essentially
it
allows
you
to
select
the
default
kind
of
happy
path
poll
interval
for
your
resources,
which
essentially
means
that
you
can,
you
know,
control
how
often
they're
being
checked
for
drift
detection,
which
I
believe
is
the
actual
description
of
the
flag
that
we
put
on
there.
But
you
can
run
like
poll
five
minutes
or
something
like
that
and
that'll
say
you
know.
C
I
check
my
infrastructure
every
five
minutes,
so
this
should
be
a
pretty
like
straightforward
thing
that
helps
folks
be
able
to
control
some
of
the
the
scale
issues
we've
seen
folks
running
into,
but
I
have
a
another
issue
and
proposed
that
will
likely
do
more,
for
that
which
I
don't
know
if
we're
in
in
that
part
of
the
discussion.
But
that
would
be
something
I'd
like
to
bring
up
today
as
well.
A
C
More
just
a
status
related
thing:
okay,
yeah.
I
think
everything
in
the
1.4.
C
Too
cool,
so
I
already
dropped
it
in
the
dock,
but
it's
and
I
think
I
brought
it
up-
maybe
last
meeting
as
well,
but
2411
consider
allowing
provider
partition
on
provider
config.
This
is
an
idea
for
basically
allowing
you
to
run
multiple
copies
of
providers
and
they're
sharded
based
on
the
the
provider
config.
C
So
how
that
would
work
right
is
the
package
manager
would
start
watching
for
provider
configs
for
each
provider
package
it
manages
and
every
time
one
was
created,
it'd
create
a
new
instance
of
the
controllers,
for
that
which
the
provider
config
is
convenient
because
every
every
managed
resource
points
at
a
provider
config.
C
So
if
you
wanted
to,
you
know,
say
like
all
of
our
dev
infrastructure
gets
reconciled
every
like
hour
or
something
like
that,
and
all
of
our
pride
infrastructure
gets
reconciled
every
minute
you
could
just
give
them
the
appropriate
provider
configs
that
were
deployed
with
those
settings.
There
has
been
a
little
bit
of
conversation
on
that.
I
think
movavic
and
I
had
a
little
back
and
forth
on
that
about
things.
C
It
would
fix
versus
things
that
wouldn't
but
I'd
love
for
folks
that
are
starting
to
hit
scale
issues
to
comment
on
there
and
say
you
know
this
would
be
useful
for
what
we're
trying
to
do,
or
this
wouldn't
or
you
know,
you're
dumb,
and
I
have
a
better
idea.
Those
are
all
helpful
things
to
hear.
Maybe
maybe
not
the
last
one,
because.
B
A
C
Yeah
there
is
some
like
interesting
stuff
around
how
this
would
impact
packages
and
that
sort
of
thing,
because
right
now
and
I'm
guessing
this
may
come
up
in
this
meeting
as
well.
There's
an
issue
I
was
talking
with
hassan
and
move
off
again
today
about
adding
something
else
to
the
controller
config.
C
The
controller
fig
is
kind
of
this,
like
large,
nebulous
object
that
does
things
sort
of
or
configures
your
controllers
is
what
it
does
right,
but
we're
kind
of
like
we've
used
it
as
kind
of
like
a
sync
to
put
the
things
it's
like
you
know.
I
want
to
be
able
to
do
this
thing.
It's
not
supported.
This
is
an
alpha
api.
We
can
just
drop
it
on
there
and
I'm
interested
to
see.
C
If
maybe
we
can
start
to
mature
that
a
little
bit
into
you
know
something
that
is
a
little
bit
more
of
an
elegant
design,
and
this
specifically
right,
if
we're
doing
things
like
a
different
irsa
for
each
provider,
config
that
that's
serving
as
a
partition
unit,
we
had
to
you
know,
figure
out
how
to
like,
assign
the
appropriate
service
account
to
each
of
those
and
that
sort
of
thing.
So
there
could
be
some
interesting
follow-on
effects.
C
The
nice
thing
about
this,
I
will
say,
and
something
that
I
think
we're
trying
to
keep
in
mind
for
all
the
designs
we're
doing
these
days
is
this
could
easily
be
put
behind
a
feature
flag,
and
so
we
could
say
you
know
this
isn't
on
by
default.
A
Makes
sense
stan,
did
you,
and
so
after
your
your
meet
up
presentation
last
night
for
for
atlanta
for
the
entire
city
of
it
did?
Did
you
get
some
feedback
about
scale
and
any
any
feedback
that
was
relevant
to
this
issue
too?.
C
So
I
definitely
there
were
a
few
folks
that
you
know
like
we're
asking
about
scale
and
that
sort
of
thing,
and
it
was
like
the
general
questions
you
get.
You
know
if
you
have
10
000
crds
is
this
gonna
fall
over
and
it's
like.
Well,
yes,
there
are,
like
you
know,
constraints
based
on
the
amount
of
memory
and
bandwidth
that
you
have
right
actually
at
what
I've
seen
more
commonly
among
folks
is
running
into
like
rate
limiting
and
that
sort.
C
Yep,
exactly
and
movavik
has
a
really
astute
comment
here
right.
This
doesn't
necessarily
solve
this.
It
depends
on
right
how
your
provider
does
rate
limiting
if
it's
by
credentials
like
the
account
that
you're
using
then
this
could
help
address
that.
But
if
it's
just
like
you
know
bulk
for
your
your
overall
account
or
something
like
that,
then
this
doesn't
address
that
and
and
to
some
extent
right
we're
we're
not
gonna
solve
cases
where
there's
an
external
rate
limit
imposed
right
that
there's
just
a
limitation.
C
You
know
if
enough
people
are
using
this
right,
we
may
be
able
to
put
some
pressure
on
folks,
but
more
likely.
What
we're
trying
to
do
is
give
you
all
the
knobs
to
tune
it
to
work
for
your
organization
right
so
kind
of,
like
my
response
to
that,
as
folks
can
read,
here
is
yeah.
We
may
not
be
able
to
solve
the
rate
limiting
issue,
but
we
can
do
things
like
you
know,
group
resources
together
to
allocate
the
the
total.
You
know
queries
you
can
make
in
a
given
time
interval.
C
You
know
like
used
to
the
greatest
benefit
of
your
organization,
so
anyway,
I
think
that's
probably
the
way
we're
going
to
get
around
those
sorts
of
scale
issues
there's
also
some
things
around.
Just
like
the
rate
limiting
we
do
in
controllers,
it
can
probably
be
better
tuned
and
that
sort
of
thing
as
well
yeah.
I.
D
I
definitely
want
to
caution
that
we
don't
look
at
this
issue
that
is
kind
of
leading
with
hey.
The
solution
is
shard
providers
as
the
one
place
that
we
gather
all
performance
things
right
like
I'm,
not
against
this
proposal,
like
especially
I
actually
like
it
for
the
you
know
the
iam
potential
like.
I
think
we
might
want
to
do
this,
but
my
hunch
is
still
that
there's
quite
a
lot.
D
We
can
do
even
to
allow
you
to
tune
individual
rate
limits
and
things
like
that
inside
of
one
controller
before
we
need
to
go
and
be
running
multiple
controllers
and
sharding,
and
things
like
that
so
yeah,
we
can
do
this
just
just
as
a
clarification
of
the
community,
though,
like
you
know,
we
may,
if
you
have
a
performance
issue,
don't
feel
like
you
have
to,
like.
C
Yeah,
I
think
that's
a
really
good
point
and
also
that
issue
that
we
were
talking
about
before
this
one
about
the
exposing
the
I
think
it
was
rate
limiting
on
the
the
external
client
there's.
Also
some
people
have
dropped.
I
think
there's
like
15
or
so
comments
on
there,
where
people
have
drops
like
yeah
I'd
like
to
be
able
to
say
the
provider
config.
You
know
how
often
this
is
reconciled
or
set
like
the
total
number
of
reconciles
that
this
controller
can
perform.
C
A
A
Sense,
okay,
so
then
other
other
1.4
stuff,
I
think
we're
you
know
we're
still
keeping
a
a
kind
of
a
handle
on
what
our
high
demand
community
issues
like
the
plugable
secret
stores
continues
to
come
up.
I
know
diego
put
in
a
vote
for
that
the
other
week
and
it's
kind
of
a
highly
highly
voted
for
issue.
So
that's
something!
That's
still
on
the
top
of
the
proposed
backlog
here
observe.
A
Way
up
there
as
well
too,
like
these
two
are
two
of
the
highest
ones.
I
think
that
across
the
community
has
been
asked
for
so
we're
keeping
those
at
the
top
of
the
back,
keep
an
eye
on
those
and
for
you
know,
when
we
either
someone
on
the
like
core
contributor
team
can
take
it
or
you
know
a
community
member
community.
Member
that
wants
to
take
on
some
of
these
issues
is
more
than
welcome.
A
We
could
be
supporting
folks
for
that
as
well
too,
but
they're
definitely
high
priority
stuff
that
that
have
been
demanded
or
been
requested.
Anything
else
in
1.4
that
folks
want
to
bring
up-
or
you
know,
folks
in
the
community
that
you
know,
issues
that
are
important
to
them.
That
haven't
been
represented
here,
necessarily.
A
All
right
so
we'll
keep
plugging
along
on
1.4
and
with
all
the
things
we're
focusing
on
here
that
we've
been
over.
I
think
the
board
is
fairly
up
to
date.
If
there's
other
updates
that
folks
need
to
make
to
the
board,
you
can
go
ahead
and
do
that
async
and
kind
of
make
sure
that
the
board
is
reflecting
the
reality
that
things
we're
working
on.
A
Okay,
let's
see
what
else
so
I
think
there
are.
We've
talked
over
all
of
these
links
here
in
the
1.4
milestone
planning.
So
I
think
that's
everything
for
1.4,
so
we
can
go
ahead
and
move
on
to
the
community
topics
section.
So,
first
off
a
whole
bunch
of
interesting
content
has
been
coming
up.
Just
yesterday,
dan
was
presented
presenting
to
the
city
of
atlanta.
On
some
it's
like
a
kind
of
interesting
topic.
You
know
kind
of
a
lot
of
content.
A
You
know
about
how
crossband
works,
but
also
getting
into
day
two
operations
sort
of
stuff
at
the
end
there.
So
that
was
that
was
super
interesting.
I
don't
think
they've
posted
it
to
the
youtube
channel
yet,
but
they
will
put
it
on
the
kubernetes
atlanta
meetup
the
youtube
channel
here
in
sometime
soon,
arguably
too
far
too
far
too
much
content,
I
don't
know
dan,
I
think
it
was.
A
It
was
pretty
useful
for
me
at
least,
but
it's
hard
to
see
that
to
a
lens
of
somebody
who
doesn't
already,
you
know,
understand
the
project
really
well,
but
I
think
it
was
pretty
good
and
then
there
was
a
sibo
organized
meetup.
Recently,
that's
you
know
about
get
ops
with
argo
and
crossplane.
There's
a
link
here
to
the
event
itself,
and
then
the
video
has
already
been
published
here,
as
well
too.
A
So
victor
was
was
part
of
that
and
anais
was
was
there
as
well
too,
so
it
was
really
cool
to
get
a
little
collaboration
there,
and
I
thought
that
was
that
was
really
good
and
then
a
couple
other
talks
recently.
I
think
this
one,
the
victor's
talk
about
how
to
use
you
know
how
to
approach
applications
as
well
too,
and
combine
cross
plane
for
infrastructure
cube
villa
for
for
apps,
and
then
argo
cds
for
the
get
out
portion
of
it.
A
I
think
that
was
released
like
the
day
of
the
last
cross
playing
community
meeting.
So
maybe
we
talked
about
that-
maybe
not,
but
it's
still
a
recent
one
as
well
too,
and
then
dan
gave
another
interesting
talk
about
how
to
use
packages
for
like
vendors,
releasing
software
and
distributing
software
as
well
too.
So
that
was
that
was
super
interesting,
any
high
level
observations,
thoughts,
etc
from
the
folks
that
were
have
been
delivering
this
content
in
victor
and
dan
and
others.
C
Just
let
us
know
if
there's
specific
things
that
folks
feel
like
we're
lacking
content
on.
I
think
that
composition,
especially,
is
something
that
has
been
getting
more
coverage
lately,
which
I
feel
like
anecdotally.
I
feel
like
there's
been
like
an
uptick
in
folks,
coming
to
crossplan
and
being
like
hey,
like
I
hear
that
you
can
build
abstractions
on
infrastructure,
as
opposed
to
saying
like
I,
I
hear
that
you
can.
C
You
know,
provision
your
infrastructure
using
kubernetes,
and
so
definitely
if
there
are
topics
that
people
feel
like
are
kind
of
like
undercovered
and
stuff
like
that
bring
those
up,
because
usually
the
meetups
especially
are
basically
like
you
can
talk
about
whatever
you
want,
as
evidenced
by
my
talk
last
night,
so
definitely
feel
free
to
offer
topics
that
y'all
want
to
hear
more
about.
A
Awesome
yeah,
I
love
the
all
the
all
the
content
that's
coming
out
now
and
the
meetups
and
the
live
streams
and
the
podcasts,
oh
yeah,
and
then
dan
and
I
recorded
another
podcast
on
monday
for
change
log
that
will
be
released
sometime
in
august.
So
we'll
put
a
link
to
that
as
well
too.
Once
that
podcast
has
been
released
so
a
whole
lot
of
activity.
A
It's
super
super
exciting
aaron
was
asking
in
the
chat
if
adding
the
the
gke
auth
provider
that
he
wrote
if
putting
that
into
controller
runtime,
so
that
other
repositories
and
providers
can
consume
it.
That
needs
to
be
tied
to
a
specific
release
and
then
my
my
response
to
that-
and
so
please
someone
correct
me
if
I'm
wrong
is
that
I
think
that
we
would
put
that
logic
into
cross
plane
run
time
right,
aaron,.
C
A
I
believe
it's
the
cross
plane
run
time.
We
wanted
to
move
it
into
since
anything
that
uses
a
cube.
Config
can
make
use
of
it
yeah.
So
so
my
understanding
is,
we
would
put
it
into
the
cross-plane
runtime
and
then
we
could
cut
a
release
of
crossband
runtime
anytime.
We
want,
I
believe
we
don't
have
to
wait
to
like
an
official.
A
You
know
1.4
or
1.5
of
crossplane
as
core,
so
we
could
do
a
cross-plane
runtime
release
anytime,
that
that
gets
in
there
and
then
downstream
providers
like
provider
gke
or
whoever
can
pick
that
up
as
they
need
as
well
too.
So
I
don't
think
it's
tied
to
any
specific
deadlines
or
cadence.
I
don't
think.
A
Yes,
sir
aaron
all
right
next
note
here
is
that
we
are
looking
for
some
feedback
on
our
docs
our
documentation.
This
this
link
has
been
circulated
for
maybe
a
week
or
two.
So
I
think,
though,
matthias
that
you're
still
looking
for
probably
like
collecting
more
feedback
to
come
in
there.
So
I
have
a
link
here
to
the
survey.
If
you
have
any
feedbacks
feedback
on
the
experience
around
the
cross
plan
and
crossplan.io
docs
getting
started
all
that
sort
of
stuff
so
feel
free
to
click
there
and
add
your
feedback.
A
So
for
cncf
incubation,
the
public
comment
period
was
opened
two
weeks
ago
by
my
calculations.
Today
is
the
last
day
of
the
public
comments
period.
So
then,
after
today
that
should
end
and
the
toc
should
be
able
to
vote
after
that.
A
I
had
previously
said
that,
oh
when
the
public
comment
period
gets
opened
up,
we'll
want
to
make
some
noise,
but
I
actually
went
back
on
that
and
decided
that
wasn't
the
case,
because
if
you
look
at
the
the
posting
there,
it's
really
just
kind
of
a
an
index
or
a
direction
to
the
artifacts
of
the
processor,
to
the
due
diligence
document
to
the
pull
request,
etc.
We've
had
a
lot
of
engagement
on
those,
so
I
don't
think
we
want
to.
A
You
know,
be
spamming:
the
that
public
comment
period
post
itself
with
like
plus
ones
and
stuff
like
that.
I
think
we're
we're
all
good
there.
The
vote
itself,
though,
when
the
vote
comes
out,
then
we'll
probably
want
to
make
some
noise
around
that
and
show
some
support
on
that
one.
A
But
for
this
public
comment
period
thing
you
know
if
you
want
to
add
comments
to
the
due
diligence
and
everything
that's
perfectly
reasonable,
but
we
don't
necessarily
need
to
be
showing
a
lot
of
support
or
on
on
this
post
here
at
least
so
fingers
crossed,
but
breath
not
held
I'd
say
for
for
progress
with
the
csf
incubation.
It's
been
a
long
time
coming,
but
hopefully
the
vote
is
coming
soon.
A
I
don't
think
there's
any
updates,
necessarily
on
performance
and
certification
stuff
when,
in
you
know,
incubation
is
approved,
we'll
be
able
to
make
progress
on
that
again
and
have
a
vote
on
that,
but
nothing
new.
A
I
don't
think
in
the
last
couple
couple
of
weeks
and
then
I
this
is
a
new
section
that
we
just
started
last
week
last
time
I
think,
and
just
around
you
know
interesting
things
going
on
in
the
community,
so
folks
can
add
more
links
there
if
they
want
to
of
you,
know
interesting
issues
or
discussions
happening
in
slack
or
anything
like
that.
A
One
thing
I
thought
was
interesting
is
that
I
saw
a
link
to
it
earlier
today
just
this
morning
and
I
did
not
get
a
chance
to
look
into
it
myself,
but
I'm
actually
kind
of
curious
for
someone's
take
on
it
if
they
got
a
chance
to
look
at
it.
Is
this
cross
plain
lips
on
it
repo
here
I'm
super
curious
as
to
what
that's
how
to
use
that.
A
What's
the
scenario
here,
what
does
you
know
help
me
understand
because
there's
not
a
ton
of
you
know
content
at
least
in
the
readme.
Does
anybody
have
a
take
on
this
lips
on
it
thing
for.
D
Crossplane,
sorry,
are
you
saying,
does
anyone
have
a
take
on
the
jsonnet.
A
D
I
haven't
looked
at
it
in
great
detail
yet,
but
I
appreciate
that
they
exist.
I
I,
like
I
like
jsonnet,
looks
good,
looks
like
it's
auto-generated
using
a
tool,
so
we
recommend
folks
check
it
out.
Json
is
a
great
way
to
write
configurations
if
you
don't
like
yaml.
B
There
is
also
a
link
in
in
the
about
section
of
the
repository
if
you
click
that
there
is
a
documentation
rendered
right
hand,
side.
A
A
A
So
this
is
kind
of
just
like
an
oh
wow,
so
this
is
like
a
you:
can
interact
with
cross
planes,
api
service
level,
surface
area,
using
this
like
generated
library.
D
Yeah
yeah
jay
started
renders
to
json.
So
so,
obviously
you
know
kubernetes
api
is
all
json
documents,
so
we
can
use
this
to
jsonp
is
before
jared,
but
gives
you
like
a
of
slightly
higher
level
programming
language
than
raw
json
with
like
loops
and
things
like
that
in
a
fairly
fairly
sort
of
elegant
way.
In
my
opinion,
oh
cool
interesting.
It's
like
like
a
mix
of
json
and
python
kind
of
to
use.
A
Yeah
I'd
heard
of
it,
but
I've
never
used
it
or
experienced
it
before.
So
that's
pretty
interesting,
and
so
so,
with
this
kind
of
thing
in
the
providers,
especially
too
like
this
would
be
maybe
a
way
to
generate
configurations
or
compositions
and
stuff.
Perhaps
with
some
of
these
functions
and
strong
types.
D
A
D
Yeah
the
commits
are
all
by
someone
from
grafana
looks
like
maybe
a
dutch
name
that
I
will
probably
butcher.
Jeroen
duologic
is
their
github
username
awesome.
A
Cool
nice,
okay,
cool
yeah,
so
well,
that's
available
and
if
folks
want
to
start
using
that
to
you
know,
generates
cross-plain
resources.
Then
that's
not
something
new,
a
new
way
to
do
it
and
that's
now
available.
That's
great
any
other
like
things
that,
like
people
have
been
like
debugging
or
experiences
within
the
community.
That
folks
want
to
bring
up
and
share
with
the
with
everybody,
on
the
call.
A
A
Then
anything
else
like
any
aprs
that
we
want
to
bring
up
for
discussion
or
get
some
eyes
on
or
anything
else.
I
think
that's
the
end
of
the
agenda
here,
so
anything
else
that
people
want
to
bring
up
or
get
some
attention
on.