►
From YouTube: 2020-05-11 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
A
The
release
notes
are
linked
here
in
the
edge
into
document,
Thank,
You
Nik
for
add
and
for
running
that
release
and
some
great
release
notes
here
about
some
of
the
features
that
made
it
in.
We
had
talked
about
a
lot
of
these
I
think
in
the
last
community
meeting
because
we're
pretty
close
to
doing
the
0.10
release,
but
this
is
a
great
write-up
if
folks,
hadn't
seen
it
for
everything
that
was
included.
A
In
that
release
side
note,
I
was
trying
to
figure
out
when
the
release
date
was-
and
you
know,
github
just
as
12
days
ago
and
I
had
to
ask
Google
I,
said
Google
what
was
12
days
ago.
Maybe
maybe
there's
a
better
way
to
do
that,
but
that
is
what
I
did
so.
It
was
April
29th,
so
V
0.11.
We
have
traditionally
had
a
mid
monthly
cadence
of
our
releases
where
we
release
every
every
month.
A
Around
the
middle
of
the
month,
we
pushed
back
the
release
a
bit
for
0.10
to
get
some
extra
features
and
polish
the
completion
into
some
of
the
new
features
like
composition,
and
so
we
will
be
getting
back.
Excuse
me
to
our
mid
monthly
release
at
this
time
again
for
0.11
in
the
proposal
is
to
to
be
to
release
on
Monday
May
18th.
So
that's
one
week
from
today
and
I,
don't
think.
A
There's
gonna
be
a
huge
investment
in
new
features
and
a
lot
of
deeper
changes,
but
there
will
be
polishing
bug
fixes
and
we
have
started
some
deeper
conversations
and
a
bigger
investment
around
documentation
to
continue
to
figure
out
what
the
right
approach
is.
What
are
the
you
know,
the
design
of
it
and
the
key
scenarios
that
we
want
to
be
sharing
and
have
a
nice
experience
for
new
users,
the
project
to
be
able
to
be
productive
and
functional
when
they
come
to
the
site
and
start
trying
to
use
cross
plan.
A
B
A
C
A
C
I
mean
you
can
see
you
just
pulled
up
here.
We
added
cluster
API
on
here
this
morning,
so
the
folks
over
there
we're
interested
and
having
it
supported
so
went
ahead
and
put
that
on
they're
still
doing
manual
additions
right
now,
not
just
parsing
any
repo,
just
to
kind
of
make
sure
we
get
all
the
edge
cases
ironed
out,
but
hopefully
eventually
it's
just
gonna
be
like
go
doc
where
it
sports
any
repo
anywhere
cool.
A
A
D
F
Public
cloud
provider,
another
thing
that
we
were
talking
about
this
is
kind
of
the
first
time
in
a
long
time
that
we've
we've
added
a
completely
new
provider,
so
it
was
sort
of
talking
about
what
would
make
it
release
ready
and
another
thing
that
would
help
a
lot
is
adding
some
tests
to
this
repository.
I.
Think
yeah.
B
F
A
A
C
C
We
also
demonstrate
some
of
the
work
they're
doing
with
secret
binding,
so
we
provision
by
referencing
a
secret
directly
in
the
service
and
also
then
used
their
sequel
binding
to
inject
the
secret
into
the
file
system
of
the
container
of
the
service
function,
so
that
was
pretty
cool
to
see
and
they
are
doing
some
pretty
innovative
stuff
with
their
controllers
over
there.
So
we
had
kind
of
a
more
general
architecture,
discussion
as
well.
That
episode
ran
a
little
long,
but
there's
a
lot
of
good
information.
C
There
valera
was
also
great
because
we
recently
kind
of
made
a
push
in
cross
plane
to
support
the
functionality
that
Blair
provides.
So
it
was
definitely
something
that
we
are
familiar
with
and
really
excited
about
being
able
to
support
that.
It
also
was
a
little
bit
different
than
a
lot
of
their
use
cases,
because
Valero
for
those
who
aren't
familiar
kind
of
back,
helps
you
backup
and
migrate
to
Renee's
resources,
because
our
resources
represent
external
cloud
infrastructure.
C
You
can
do
some
pretty
creative
things
like
moving
your
control
plane
from
one
cluster
to
another
and
automatically
reassuming
control
of
existing
cloud
M
structures.
So
we
showed
some
things
like
original
WordPress
and
then
destroying
the
cluster
and
starting
a
new
one
and
restoring
their
I'm
not
having
any
downtime
between.
So
there's
some
really
interesting
stuff.
There
I
think
there's
gonna
be
some
further
collaboration.
They
actually
already
addressed
an
issue.
We
opened
about
how
they're
restoring
some
CR
DS,
so
a
big
shout
out
to
the
team
over
there
one
this.
F
F
It's
usually
just
being
kind
of
the
cross
plain
core
team
sort
of
rolling
through
and
supporting
these
things,
but
Ryan
pointed
out
that
we
just
sort
of
went
to
did
that.
So
we
may
need
a
better
process
in
future
to
sort
of
communicate
changes
like
that,
where
we,
we
probably
need
to
make
them
to
all
repos
and
those
repos
now
have
distributed
ownership
sort
of
thing.
You're
gonna
have
different
maintainer
teams
and
things
like
that.
Don't
that
looks
like
a
cross,
plane,
maintainers
mailing
list
or
something
eventually.
A
Yeah,
that's
a
good
point
Nick
and
thanks
to
the
feedback
to
Ryan
on
that
as
well.
You
know
that
being
able
to
understand
the
needs
of
all
consumers
and
contributors
to
the
project
and
making
sure
that
you
know
who
are
the
key
stakeholders
and
do
they
have
a
say
in
the
changes
and
then
also
downstream
impacts
through
their
consumers,
and
you
know
how
they
may
be
affected
by
changes.
Broad
changes
as
well,
so
yeah,
that's
a
good
reminder.
No.
E
I'm,
actually
that
the
more
general
question
is
this,
so
it's
it's
a
dependency
across
repro
I.
Don't
know
how
to
check
that,
because
the
CI
wouldn't
pick
up
that
change.
I
only
found
out
when
I
changed,
helm
in-store
to
the
latest,
then
I
see
the
see
the
difference
and
in
that
breaks
the
code.
What's
the
general
way
to
do
that
so.
E
C
I
think
one
of
the
ways
that
we're
doing
that
right
now
is
having
these
libraries
that
are
shared
across,
and
you
know,
having
releases
and
sharing
release,
notes
and
mentioning
breaking
changes
in
those.
So
in
this
case
we're
in
sort
of
a
weird
situation
where
the
application
configuration
controller
was
in
core
crossplane
and
then
code
was
moved
to
OEM
kubernetes
runtime.
So
it's
kind
of
a
unique
situation
in
general
I
think
we're
set
up
pretty
well
to
kind
of
have
the
traditional
kind
of
downstream
consumption
of
libraries.
So
I.
C
Imagine
that's
how
it
will
be
supported
in
the
future,
and
those
kind
of
things
will
be
more
apparent.
It
is
still
a
little
weird
with
omq
renée's
runtime,
because
it's
basically
just
controllers
that
cross
pane
is,
you
know,
doing
the
setup
function,
for
it
add
add
to
the
controller
manager.
But
if
those
changes
are
happening
you
know
am
crew,
nays,
runtime
and
then
crossplane
updates.
You
know
that
should
be
part
of
its
release
process
to
know
that
it's
dependencies
have
changed
and
that
should
be
apparent
in
in
notes
and
that
sort
of
thing.
E
Let
me
see,
because,
even
even
in
this
current
setting
right,
if
you
change
the
OEM
Cooper
native
runtime
and
it
breaks
one
of
the
repo
like
the
remote
and
the
local,
that
that
checking
will
not
be
blocked
by
any
CI.
That's
what
I
was
wondering,
because,
let's
see
I
will
pass.
The
only
thing
we
will
break
is
when
the
next
checking
on
the
other
repos,
then
we
will
see
the
CI
fail.
That's
actually
what
I
would
I
see
I
found
out,
because
I
changed
the
CI
setting,
then
the
CI
failed.
How
does
that
work?
E
E
B
F
Yeah
I
think
this
there's
two
parts.
This
switches,
which
is
that
test,
will
be
the
validation
but,
as
dad
pointed
out,
we'll
be
able
to
do
a
much
better
job
of
surfacing
this
once
we're
semantically
versioning
the
OEM
runtime
library,
so
that
you'll
at
least
have
to
actively
choose
to
get
the
new
version
and
then
you'll
be
able
to
see
the
release,
notes
that
have
a
reason
to
expect
that
something
might
break.
You
know
they'll,
hopefully
be
a
big
breaking
change.
This
thing
goes
away
in
there,
so.
F
Yeah
exactly
I
think
they
would
both
be
good
and
the
reason
that
pointing
it
out
mostly,
is
because
building
those
integration
tests
is
typically
something
that
I
see
being
really
nice
to
do.
But
it's
quite
hard
to
do.
It
takes
a
lot
to
get
done,
so
the
version
in
the
OEM
runtime
library
is
probably
gonna.
Come
first
a
hand,
be
you
know,
get
us
30
percent
of
the
way
there
and
then
seventy
percent
of
the
way
we'll
be
watching
those
tests.
A
I
think
I
think
all
that
makes
sense,
and
it's
probably
yet
like
a
phase
sort
of
thing
where,
like
the
Nick
was
referring
to
of
you,
know
throughout
most
of
the
most
all
of
the
projects
within
the
Crossman
organization,
they
are
semantically
versioned
and
you
know,
have
official
releases
and
such
as
well.
So
you
know
that
shouldn't
be
too
hard
to
incorporate
into
our
process
for
this
repo
as
well
move
off
on
the
call
night,
I
think
he's
not
actually
Nick.
F
We
pushed
to
get
an
MVP
of
this
functionality
into
0.4
infrastructure,
so
I
think
almost
every
one
of
these
calls,
probably
don't
me
talk
about
it
before
so
I
want
to
kind
of
assume
that
you
know
what
composition
is,
but
roughly
it'sit's
pretty
much
gonna
replace
cross-play
and
reasonless
classes
of
claims.
So
our
goal
at
the
moment
is
to
deprecate
reasonable
classes
of
claims.
They'll
still
be
there,
but
they'll
be
mock
deprecated
in
the
0.0
Levin
and
to
document
using
a
composition
instead.
So
part
of
the
reason
that
we
want
to
do.
F
That
is
because
it
opens
up
a
lot
of
use
cases
where
you
can
compose
infrastructure
of
multiple
different
things
in
your
API
or
tips
of
you
know.
You
can
compose
the
database
in
the
file
rule
or
a
database
to
the
user
in
so
they
say
they're
there
and
you
kind
of
resolve
called
my
great
database
or
something
like
that
and
provide
those
as
an
infrastructure
operator
to
your
users.
So.
F
As
to
what
functionality
is
there,
maybe
the
easiest
way
to
think
about
it.
If
you
do
want
to
go
and
sort
of
read
a
little
bit
about
this
afterwards
is
the
design
doc
talks
about
infrastructure
and
application
composition,
we've
only
added
support
for
infrastructure
composition.
At
this
point,
we
are
missing
a
few
things.
For
example,
if
you
edit
a
requirement
for
a
resource
which
is
kind
of
resource
claim
concept,
we
would
like
it
to
actually
go
and
make
those
changes
flow
untold.
F
It
could
post
resources,
but
it
is
not
at
the
moment,
so
that
is,
that
is
one
place
that
needs
to
be
worked
on
in
future.
We
also
would
like
to
be
able
to
force
a
composition.
A
composition
is
kind
of
like
a
resource
class
under
the
new
design
or
set
one
as
a
default.
I
believe
neither
of
those
are
supporters.
At
the
moment
you
can
just
select
them
using
labels
and
going
into
open
eleven
I.
Don't
think
we're
going
to
change
a
lot
of
functionality,
really
we're
mostly
shoring
up
testing.
F
F
Need
to
open
eleven
really
of
any
significance,
closely
conferences,
I
believe
there
is
also
I'm,
not
sure
if
it
comes
up
in
that
issue.
Query,
but
another
thing
that
would
be
good
to
get
feedback
on.
Oh
yeah,
it
was
the
top
thing
in
that
query.
Is
the
composition,
revision
proposal,
which
proposes
we
have
a
similar
problem
with
compositions
that
we
had
with
OM
application
configurations
where
you
have
one
sort
of
thing:
that's
a
template
that
shared
by
many
resources.
F
If
someone
goes
and
edits,
the
template
than
all
those
potentially
would
get
updated,
we're
thinking
about
taking
a
similar
approach
to
what
was
proposed
with
OEM,
by
adding
a
revision
type
similar
to
the
controller
revision
in
kubernetes.
That
would
allow
you
to
explicitly
move
back
and
forward
versions
of
the
composition,
so
the
feedback
once
it
on
that.
A
F
A
D
I
actually
left
a
few
comments
in
the
composition,
ter
and
I'm,
actually
not
very
sure
that
Halloween
handled
the
application
layer
combination,
my
personal
feeling,
Lee.
That
combination
is
not
a
very
common
user
case,
for
you
know
to
compose
several
pieces
of
applications
into
the
same
application
definition
and
I'm
submitting
that
partying.
The
proposal
so
I'm
not
sure.
What's
up
m4
in
the
application,
they
makin
part.
F
It's
just
a
different
patent,
different
language
for
doing
that
thing,
but
I
think
we
we
saw
the
feedback
and
we
would
like
to
sync
up
at
some
point
and
see
if
we
can
find
some
alignment
there,
I,
don't
there's
no
plans
to
implement
application
called
position
by
0.0
live
in
any
way.
So
it's
not
it's
not
really
a
priority
for
this
week
or
anything
like
that,
but
but
we're
happy
to
set
up
a
chat
to
talk
about
how
it
interfaces
with
OEM
before
we
starting
and
earnest
on
that
functionality.
Yeah.
D
B
C
F
Yep
yep
we're
not
working
on
application.
Did
the
application,
definitional
composition
for
the
Oh
point,
11
timeframe
cool
you
did
say
it
is,
there's
still
something
that
we
do
want
to
the
shoes.
So
after
after
11,
we
definitely
should
follow
up
on
that.
I
think
we've
spent
months
in
lots
and
lots
and
months
working
on
this
design
and
we're
I.
Don't
think
we
agree
that
it's
that
it's
inferior
or
too
complicated,
I'm,
biased.
A
Well
sounds
great,
okay
and
then
I
have
a
number
of
updates
around
support.
Specifically
then
I
asked
Dan
to
include
a
couple
of
these
on
here,
so
I
think
Dan
has
added
I.
Think
all
all
of
these
issues
here.
So
do
you
want
to
speak
real,
quick
about
those
or
whoever's
on
the
call
for
these
two?
They
can
they
can
speak
for
themselves.
Dan.
Do
you
wanna
start.
C
Yeah
for
sure
so,
I
just
added
in
those
other
two,
so
when
I
get
to
them,
Ryan
feel
free
to
jump
in
or
hung
ciao
I,
don't
know
if
he's
on
the
call,
but
this
first
one
is
just
adding
integration
tests
I
just
dropped
it
because
it
was
an
update
and
something
we
had
talked
about.
So
we
have
some
some
integration
tests
on
the
OEM
clear
days
runtime.
C
Now
these
are
kind
of
a
model
for
how
we
might
do
integration
testing
more
in
crossplane,
which
is
a
sorely
needed
area
to
improve
upon
so
excited
to
actually
have
some
of
those
running.
One
thing
I
will
mention
here
is
this
was
super
easy
with
github
actions,
it
was
really
easy
to
get
it
going.
Just
cuz
it'll,
there's
a
basically
one
for
a
kind
cluster
on
there,
so
it
might
be
something
that's
worth
taking
a
look
at
for
that.
C
Another
thing
well
I'm,
not
sure,
there's
actually
any
more
on
that
I'll
go
ahead
and
you
a
quick
summary
of
this
kind
of
bring
your
own
trait,
which
is
my
chosen
name.
That's
not
what
the
PR
is
called,
but
it's
just
kind
of
a
summary.
Essentially,
we
want
to
enable
being
able
to
use
existing
kubernetes
resource
types
either.
C
You
know
native
Karuna
sites
or
just
once
implemented
by
other
projects
as
traits,
and
this
PR
is
a
design
that
Ryan
has
put
together
for
enabling
that-
and
there
is
quite
a
few
comments
on
it
at
this
point,
so
feel
free
to
weigh
in
there
and
Ryan.
If
you
want
to
add
any
more
I,
think
I
saw
you
on
the
call
Yap
you're
still
here,
you're
welcome
to
jump
in
here.
Yes,.
E
So
I
think
the
only
last
thing
that
we
are
still
not
decided
on
is
actually
two
things.
One
thing
is
I
think
if
you
scroll
down,
if,
if
you
scroll
down
to
the
bottom
of
it,
I
have
to
I,
had
a
comment
at
the
end.
I
think,
and
the
only
thing
we're
not
agree
on
anyway,
agree
on
is
one
is:
will
we
allow
user
to
bringing
their
own
CR
DS?
If
we
do
that,
like
then
use
a
different
name,
but
original
proposal
doesn't
have
that
and
so
seems
like
we're?
F
Think
I
think
you
might
be
talking
about
me
in
that
case
said:
I'm.
Sorry,
if
I
gave
that
impression,
I'm
bi
on
board
with
bringing
your
own
CR
I'm-
okay,
oh
yeah,
I
I,
have
no
issue
with
with
Brittany
or
NCR.
The
main
thing
that
I've
been
going
back
and
forth
and
sort
of
a
little
bit
of
a
dog
with
a
bone
here
was
the
the
design
really
makes
it
sound
like.
E
F
E
F
Let
me
let
me
try
and
let
me
try
and
explain
my
thinking
here.
So
it
seems
like
what
we're
saying.
So
what
was
it
further
for
the
others
word
or
on
the
call
who
might
not
be
familiar
with
this
specific
thing
traits
have
a
reference
to
the
workload
that
they
apply
to
and
currently
in
the
one
trait
that
we've
implemented
the
manual
scalar
trade,
that
is
a
spec
field.
F
Workload
because
I
have
a
reference
to
it
and
now
what
we're
saying
is
well,
that's
potentially
limiting,
because
that
means
it
assumes
that
every
OEM
trait
has
this
specter
workload.
Ref
and
Harry
and
Ryan
have
bought
some
examples
of
things
that
you
might
want
to
bring
to
kubernetes
CR
and
use
it
as
a
trait
that-
and
maybe
this
could
be
an
indicia
doesn't
know
it's
a
trait.
Maybe
it
doesn't
know
anything
about
om.
So
an
example
was
the
EDD
backup
CR,
which
I
think
comes
from
the
sed
operator.
You
could
use
that
as
a
trait.
F
So
then
the
thinking
is
okay.
Well,
if
that's
gonna
be
a
trait
and
it
needs
a
reference
to
its
workload,
we
can't
put
assumptions
on
its
schemas,
so
we're
saying
maybe
we'll
move
that
into
annotations,
because
we
can't
say
the
entity
backup
which
we
have
no
control
over
at
CR
will
have
a
specter
workload
breath.
So
my
point
is
putting
annotations
on
that.
As
far
as
I
can
tell,
unless
I'm
missing
something
this
design
is,
is
effectively
pointless
because
the
entity
backup
controller,
doesn't
know
those
annotations,
it
won't
use
it
it
has.
F
It
has
a
spec
field,
called
it's
in
the
endpoints
and
that's
how
it
figures
out
what
it
backs
up
it
will.
It
will
just
always
ignore
and
see
gauge
stuff,
sorry
always
ignore
OEM
stuff.
So
in
my
mind,
I
categorize,
these
into
sort
of
two
different
patterns
this
this
traits
that
we
wrote
intending
to
be
OEM
tricks
and
those
can
use
this
metadata
about
a
workload
we
could.
E
F
F
D
E
D
E
D
F
We
actually
talked
about
this
a
while
back
that
issue
about
can
runtimes,
add
fields
that
are
pop
respec
I.
Think
I
agree
that
it
should
be
a
recommendation
rather
than
pod
the
spec,
because
maybe
other
runtimes
don't
need
this.
Do
something
else
and
I
think
realistically
either
way
whether
you're
saying
it
can
optionally
go
in
the
spec
or
optionally
go
in
the
annotation.
F
You
kind
of
it's
you'd
be
putting
constraints
on
the
runtime
anyway,
or
putting
requirements
on
the
runtime,
so
I
I
would
put
it
in
this
back
and
say
it's
optional
and
just
document
that
hey,
if
you're
a
trade,
then
you're.
Writing
if
you're
writing
your
controller
to
be
a
trade
specifically,
we
strongly
recommend
you
add
this
field
and
cross-claim
will
automatically
take
care
of
it.
If
you
do
support
this
bill,.
F
No
yeah
this
we
talked
there
was
this
there's
kind
of
a
there's,
two
or
three
things
going
on
in
this
design.
One
of
the
things
that
we
talked
about
was
how
a
workload
would
figure
out
what
resource
kinds
it
was
on.
So,
for
instance,
a
containerized
workload
can
declare
that
it
might
produce
a
deployment
in
a
server,
so
this
rather
at
that
I
think
right
proposed
putting
it
into
a
into
the
metadata
of
the
workload
definition,
which
I
think
is
a
really
good
idea
and
in
general,
if
that
was
possible.
F
D
Don't
think
it's
part
of
the
trace
destination,
a
stagnation,
more
like
the
index
to
the
chiles,
the
Rd,
you
check,
values,
and
that
means
extra
room
feels
like
applies
to
Alicia.
It's
generally
a
feature
of
the
street,
but
this
bill
is
not
a
feature
option.
It's
more
like
an
what
we
need
for
the
implementation
to
work,
so
we
don't
need
to
enforce
it
at
all
and
it
should
not
be
part
of
the
home
staying
at
all
all
right.
This
is
just
internal
mechanism,
yeah
totally
general.
E
D
F
E
D
F
I,
just
just
to
fall
on
from
for
the
trait
workload,
interaction
for
request,
if
the
only
things
that
I
think
stopping
me
from
approving
down
at
the
moment
is
if
we
could
encode
what
we
just
just
aligned
on.
There
say
that
will
use
a
respect
field
and
also,
if
you
could
put
Harry's
example
of
the
entity
back
up
in
there.
It
would
be
really
handy
because
that
design
dog
does
talk
about
this
use
case.
A
C
I
can
just
give
a
brief
overview,
basically
we're
just
trying
to
figure
out
the
best
way
to
install
controllers,
because
there's
both
the
local
and
remote
scenario.
This
specific
issue
is
talking
about
packaging
them
into
the
cross
plane
chart
which
is
one
option,
but
I
have
a
feeling
that,
with
some
of
the
coming
refactoring
and
that
sort
of
thing
that
this
might
kind
of
become
more
apparent
and
sort
itself
out
on
its
own.
C
But
this
is
probably
the
issue
where
we'll
track
that
sort
of
thing,
but
basically
we
want
to
make
the
experience
hold
that
our
because
right
now
it
is
a
bit
of
a
pain
to
install
cross
plane
and
then
add
these
controllers
that
are
doing
most
of
the
heavy
lifting
here.
So
I
just
want
to
service
that
in
the
in
the
meeting
today,.
E
There's
a
higher-level
question,
I
think
is
I'm
not
encouraged
on
Tom
mentioned
that
in
that
proposal,
OPR
is
is:
are
we
going
to
do
a
helm,
carting
store
or
I
from
my
current
understanding?
Is
the
at
least
a
remote
around?
You
would
do
helm
child
Institute
or
the
cross
flane,
and
then
there's
a
stack
manager
that
managed
a
Don.
Are
we
going
to
keep
that
thing
or
music
and
do
just
one
helm
chart
with
whatever
sub
sub
chart
or
whatever
to
in
stop
yeah.
C
I
think
that's
the
the
core
question
here
and
like
I
said
right
now,
the
the
stack
install
is
a
pretty
good
experience,
because
it's
a
simple
thing
for
the
end
user
to
install.
That
being
said,
like
I,
said,
I
think
some
some
changes,
you're
gonna
be
made
to
that
process
and
I
think
it's
part
of
that.
It
might
become
apparent
how
we,
how
we
want
to
do
these
OEM
controllers,
specifically,
okay,.
B
C
F
Think
something
that
dad
was
alluding
to
there
that
I
won't
make
any
promises
on,
because
it's
still
really
early
days,
but
we're
we're
figuring
out
some
potential
changes
or
improvements
to
sort
of
packages
and
stacks,
and
things
like
that.
So
we
might
probably
the
order
of
weeks
have
a
design
talk
that
that
sort
of
clarify
is
where
that
sort
of
part
of
cosplaying
is
going.
A
Yeah,
that's
that's
gonna,
I,
think
that
you
know
making
sure
that
the
idea,
the
concept
of
stacks
of
packaging
in
etc
and
the
crossplane
ecosystem
is
something
that
is
functional
and
useful
and
helps
adoption.
So
they
have
further
discussions
about
that'll,
be
really
useful.
Hashed
in
last
one
here
service
record,
yep.
C
C
It
would
be
nice
to
also
show
how
we
could,
instead
of
using
stacks-
or
you
know
just
manually,
creating
these
manage
resources
if
we
can
basically
create
all
the
manage
resources
required
for
service
tracker
in
one
go
by
using
composition,
so,
for
instance,
on
Azure
not
having
to
manually,
create
that
v-net
rule
and
things
like
that,
so
should
be
an
improved
user
experience.
It
shouldn't
be
a
huge
heavy
lift,
because
the
great
thing
about
composition
is
that
you
can
define
all
this
in
llamo.
C
H
C
D
D
That's
really
cool
they
people
complain
about
that
complain.
A
lot
about.
Ok,
even
even
today,
you
have
your
class
to
bootstrap
your
or
in
the
chapter,
but
you
don't
have
to
do
a
lot
of
manual
configuration
or
cause
different
for
our
school
to
clone
I.
Think
yeah.
This
table
will
show
them.
Ok,
and
this
is
really
the
power
of
the
dignity,
application
management.
A
H
Yes,
so
I
think
the
big
update
right
now
is
that
we
are
working
on
figuring
out
the
back
end
of
that
whole
tool,
how
it
will
go
from
the
intermediate
format,
the
manage
resource
team,
ax
format
to
boilerplate
for
controllers
and
there's
also
some
conversation
on
a
github
issue
for
using
a
cube
form.
So
I
think
there's
some
questions
about
like.
If
you
know
the
terraform
front
end
option.
A
Yeah
I
think
that
sounds
like
a
good
investment
to
understand
and
at
least
have
a
well
inform
stance
on
them.
If
you
can
accelerate
the
project
or
if
it's
you
know
the
right
architecture,
decision,
etc,
great,
already,
okay,
so
that
was
everything
that
is
in
the
community
topics
and
questions
section
here
before
we
head
on
to
the
PR
section.
Is
there
anything
else,
agenda
topics
here
for.
D
About
a
new
flooring
and
the
issuing
of
the
support
and
actually
spoke
in
the
application
scope,
implementation
in
close,
then
I
make
sure
the
last
part
for
the
role
of
support
in
the
project.
I'm,
not
sure
if
alter
has
any
updates
on
that
part
actually
add
in
the
release
for
the
neighbor
point,
11.
D
B
F
A
Yeah
we
stick
pretty
strongly
to
you,
know
monthly
release
cadence
as
well,
so
you
know
some
things
more
complicated
it
doesn't
it
all
fit
into
0.11
and
when
we
turn
out
that
you
know,
and
essentially
a
month
from
now
we'll
have
another
release
near
not
12
and
mid-june.
So
that
would
be
another
boat
or
train
to
catch.
A
A
Okay
and
so
he'll
sent
me
a
message
this
morning
that
there's
a
new
PR
opened
about
rocket
three
supports
in
AWS.
So
this
is
a
brand
new
PR
with
support
for
some
of
those
resources
like
hosted
zones
and
resource
record
sets,
so
we
know
we'll
be
able
to
take
a
look
and
get
feedback
on
this
one.
Were
there
any
other
items
of
note
for
that?
So
you
know.
E
A
A
F
F
F
G
A
G
Am
policy
is
good
to
go
the
network
resources
I
should
update
it
in
some
time
and
that
you
know
and
cache
cluster,
since
we
had
made
it
lower
on
the
priority.
I
haven't
checked
it
in
last
couple
of
days,
but
yeah
I'll
take
a
look
and
see
if
it's
you
know
if
it
needs
a
reduced
from
the
master
or
anything.
Okay.