►
From YouTube: 2022-06-30 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
let's
see
what
is
it
june.
30Th
2022
cross
playing
community
meeting
agenda
doc
is
still
open
now
and
will
remain
open.
So
if
you
have
a
topic
that
you
want
to
discuss,
feel
free
to
add
it
to
the
agenda
doc
as
we're
going
on.
So
let's
jump
into
milestones
and
releases
as
we
typically
do
to
start
the
meeting
so
the
for
core
cross
plane
and
cross
plane
runtime.
A
I
did
a
quick
check
and
the
last
two
weeks
there
weren't
any
specific
releases
that
we
want
to
go
over
there.
That
does
not
surprise
me,
since
we
are
ramping
up
for
the
1.9
release
in
just
a
couple
of
weeks
here
I
think
that'll
be
yeah.
I
think
that
would
be
right
before
the
next
community
meeting.
We
will
have
the
1.9
release
out
on
july.
12Th
is
the
planned
our
attention
or
scheduled
date
for
that
it's
not
sensitive.
A
That
is
the
scheduled
date,
so
yeah,
so
we're
making
progress
on
1.9
and
getting
towards
the
end
of
that.
So,
let's
jump
on
into
the
project
board
and
see
if
there's
any
things
that
we
want
to
address
or
bring
up
or
talk
through
any
any
potential
blockers,
etc.
So,
let's
go
to
the
1.9
board.
Here
we've
got
a
few
things
that
have
been
merged
with
fixes
into
into
the
done
column.
Now,
there's
this
one.
A
Actually,
this
one
here
is
this:
let
me
see
this
current
status
on
this
one:
okay,
yeah.
I
was
concerned
about
this
one
because
it
looked
like
you
know,
there's
a
panic
that
was
happening
when
someone
when
someone
from
the
community
was
wanting
to
turn
on
the
alpha
feature
for
composition,
revisions,
and
I
was
concerned
if
that
was
a
panic
that
was,
you
know,
causing
a
crash
and
the
crossbar
provider.
Sorry
core
crossblade
pod.
B
A
Got
it
well,
if
it's
a
problem
for
christopher,
then
it's
definitely
a
problem
for
me,
so
we'll
keep
going
on
that.
Okay,
so
okay,
so
I
see
okay.
So
nick
nick
closed
this
issue,
because
it's
a
it
is
a
it
is
a
panic.
There's!
No
we're
not
gonna
deny
that
that
looks
like
a
definite
crash,
but
just
a
different
issue
with
root
cause
being
tracked
there.
A
So
this
is
the
issue
that
we
do
care
about,
and
so,
when
this
is
in
1.9,
this
is
something
that
we
are
addressing
and
there's
a
pull
request
open
for
it
now
so,
okay,
so
we
are.
This
is
fine
to
be
closed
because
we
are
following
up
on
it
with
the
the
now
root
cause.
That's
actually
happening
now
and
that's
yeah,
that's
in
review
right
now.
A
I
would
we
can
have
a
discussion
about
when
that
fix
gets
merged,
because
it
is
a
panic
crashing
bug
to
consider
doing
a
patch
release
for
it.
If
it
affects
1.8
and
1.7
as
well,
then
I
would
consider
sorry
1.8
is
the
current
released?
Yes,
so
1.8
1.7
and
then
before
1.9.
A
The
last
three
releases
would
also
include
1.6,
so
if
those
versions,
if
it
is
affected,
I
would
I
would
want
us
to
consider
a
a
patch
release
for
that.
I
don't
know
if
anybody
else
has
strong
opinions
that
they
want
to
share
on
that,
but
we
can
take
that
up
with
nick
in
context
of
the
fix
that
he's
working
on
for
it.
A
All
right
and
then
the
other
issues
here
that
we
have
pull
requests
open
on
as
well.
A
I
don't
know
if
they're
directly
connected,
but
nick
did
mention
that
first
fixing
the
the
panic
that
we
just
saw
here
and
then
taking
the
fixes
for
these
as
well
after
that,
so
I
don't
know
if
they
were
related
in
terms
of
a
dependency
thing
or
if
it
was
a
prioritization
thing
of
fix
the
panic
crash
first
and
then
do
you
know,
accept
or
get
these
other
fixes
merged
in,
but
these
will
come
in.
A
I
think
behind
the
fix
for
for
the
for
the
the
panic
that
we're
seeing
here
but
they're
all
in
progress,
and
I
believe
they're
intended
to
be
accepted
and
and
and
included
in
the
1.9
release.
A
Excuse
me
so
support
patching
from
common
data
sources.
You
know
there.
There
is
a
lot
of
velocity
still
going
on
on
that.
So
max
has
been
doing
a
lot
of
great
work
to
continue.
You
know
flushing
on
trying
to
finalize
the
api
that
we
that
we
want
here
for
this.
So
I
think
there's
it's
I
that
is
getting
much
closer
to
a
finishing.
The
finish
line
than
it
was
previously.
A
I
don't
think
I
don't
know
if
this
will
be
a
part
of
1.9,
though,
because
there's
still
some
discussion
outstanding
discussion
on
what
the
api
shape
and
experience
is
like,
and
you
know
I
think,
we're
we're
heading
into
the
feature
freeze
now
as
well,
so
I
probably
wouldn't
want
to
rush
this
into
1.9,
so
I
would.
A
I
would
bet
that
this
would
not
be
included
in
1.9,
but
there's
a
lot
of
velocity
on
it
and
a
lot
of
progress
to
getting
this
to
the
finish
line
sometime
in
the
near
future.
Here
this
is
a
priority.
Multiple
people
are
collaborating
on
it,
so
this
is
something
that's
going
on.
A
Let's
see,
I
don't
think
nick
or
al
pair
are
on
the
call
today,
but
so
those
two
those
two
folks
have
been.
You
know
focusing
a
lot
on
scaling
issues
and
performance
issues
that
come
from.
You
know
installing
thousands
of
crds,
like
crossplane,
has
a
tendency
to
do
so.
Just
a
quick
update
on
that
to
share
with
folks
in
the
in
the
community
here
or
is
this
the
right?
A
There
may
be
a
better
issue
that
captures
the
latest
latest
updates
there,
but
one
thing
that
nick
found
here
is
that
he
was
doing
another
test
pass
to
with
all
of
the
fixes
that
we
had
put
into
upstream
kubernetes,
both
on
the
server
side
and
the
api
server
and
then
within
the
client
side,
so
cube,
ctl,
cube,
control
nick
was
doing
a
test
pass
with
those
in
on
a
linux
client.
A
So
from
his
linux
desktop,
he
was
seeing
that
you
know
with
a
a
cold
cash
and
empty
cash
and
doing
a
full
disc
discovery
from
the
client
side
with
thousands
of
crds
installed.
It
was
taking
just
about
one
second,
so
that's
obviously
a
drastically
improved
situation
from
where
we
were
previously-
and
you
know,
that's
a
much
much
better
experience,
the
control
plane
and
the
client
side
doing
discovery
is
now
better
prepared
to
handle.
A
You
know
a
lot
of
crds
being
installed,
but
then
he
was
also
seeing
on
a
mac
using
a
mac
client,
a
cube
ctl
from
a
you
know,
mac
laptop
that
was
taking
like
double
digits
seconds:
12
13,
14
15.
You
know
almost
20
seconds,
sometimes
and
so
nick
looked
into
that
and
found
what
we
believe
to
be
potentially
root
cause
for
that,
and
so
nick
has
a
pr
open
on
upstream.
A
You
know,
cube
ctl
that
will
potentially
address
that
for
max
as
well,
so
that
the
client
side
experience
is,
you
know
down
to
order
magnitude
like
one
second
for
doing
a
full
cold
cash
discovery.
So
that
would
be
very,
I
think,
a
really
good
outcome
here.
I
think,
there's
probably
still
some
like
the
manage
cloud
provider
managed.
Kubernetes
services
can
still
have
some
issues
with
scaling
the
control
plane
in
memory
usage
and
going
through
a
couple
rough
patches.
A
As
the
crds
are
being
installed,
so
I
think
there's
a
number
of
things
to
improve
there
still,
but
overall,
the
experience
here
has
drastically
improved
over
the
last
in
in
a
number
of
months.
While
you
know
we've
been
doing
client-side
and
server-side
fixes
and
putting
those
into
webstream
kubernetes.
So
a
lot
of
really
good
and
impressive
effort
from
here
and
it'll
be
interesting
to
see.
You
know
with
everything
that
we
get
into.
A
You
know
upstream
kubernetes
of
what
it
looks
like
after
all
that,
but
I
think
things
are
much
better
now
and
mafik
do
you.
I
thought
this
was
the
right
issue
that
had
like
a
you
know,
a
summary
nick
wrote
a
summary.
I
swear
in
the
past
24
hours
that
was
like
you
know,
pointing
to
the
upstream
contributions
that
we're
putting
in
to
like
improve
the
mac
client,
and
you
know,
like
have
even
like
a
link
to
hey
here's
a
you
know
a
locally
built,
cube
ctl.
A
If
people
want
to
use
that
now
as
well,
do
you
know
where
that
issue
would
be
dude
or
where
that
was.
A
Sure
feel
free
to
if
you
find
that
feel
free
to
drop
a
link
to
it
in
the
agenda
doc,
because
I
think
folks
that'll
be
interesting
for
folks
as
well.
I
thought
it
was
this
issue,
but
this
one
has
been
updated
in
a
week,
so
I
might
have
been
wrong
about
that.
A
Okay,
other
questions
or
comments
on
1.9
release,
milestone
time
frame.
C
A
So
we
will
keep
going
along
in
the
1.9
release
cycle
here
and
we
will
be
working
towards
getting
these
outstanding
bug,
fixes
and
changes
into
the
main
branch,
so
they'll
be
included
in
the
1.9
release
that
is
coming
in
just
under
two
weeks,
all
right,
let's
hop
into
providers
as
well,
so
christopher
or
muafik.
Do
you
all
want
to
talk
about
this
patch
release
that
was
in
went
into
aws?
They
fixed
the
regression
since
the
last
community
meeting.
C
D
Yeah,
I
can
do
a
quick
update.
There
was
a
bug
in
iam
user
crd
that
added
tags
infinitely
and
that
caused
some
problems
with
it
with
kubernetes
cluster,
where
xcd
started
to
not
respond.
So
it
was
pretty
critical
bug
and
we
pulled
back
the
0.28.0
to
pre-release
and
suggest
installing
the
patch
version.
D
A
It's
pretty
good
font
size
for
this.
That's
very
noticeable
like
excellent
work
on
making
that
visible,
nice,
okay
and
then
christopher.
Do
you
want
to
roll
us
right
on
into
the
0.29
release
that
will
be
coming
out
soon
for
for
provider,
aws.
B
I
think
I
updated
the
the
release
things
because
we
have
one
open
pr.
We
want
to
bring
in
with
ecs,
but
the
problem
at
the
moment
is
that
the
ci
is
failing,
but
this
is
only
failing
in
ci,
not
locally.
I
think
I
talked
today
with
more
also-
and
I
think
and
his
machine
is
also
working
so
yeah.
I
will
have
a
look
afterwards.
A
C
A
So
everybody
like
running
locally,
like
make
build,
make
tests
make
reviewable
whatever
that
all
works,
fine
for
everybody
on
their
local
machines,
but
in
the
ci
system,
with
github
actions,
etc.
That's
what
we're
seeing
failures,
yeah.
B
We're
seeing
that
that
in
the
ci
the
crds
are
modified
and
one
of
the
files,
I
don't
know
why
we.
A
Dang,
so
we
don't
have
like
a
diff
of
of
the
like
what
is
changing
in
those
files
that
could
give
us
a
clue
to
shoot.
Okay,
yeah.
That
is
weird.
A
Okay
got
it,
so
that's
yeah,
so
you've
tracked
that
here
as
oh
three
minutes
ago,
nice
very
recent
so
that
we
yeah.
So
this
is
blocking
the
getting
the
release
out
so
fixing
the
issue
with
the
ci,
then
we'll
work
on
getting
the
0.29
release
out.
If
that
makes.
B
A
Christopher
thanks
for
your
efforts
on
on
you
know,
getting
things
merged
and
getting
features
in
and
and
cutting
the
release
and
all
that
sort
of
stuff
man.
It's
definitely
appreciated
anything
else.
You
all
want
to
add
for
a
provider
for
provider
aws
specifically.
B
Oh,
I
have,
I
have
later
one
pr
which
is
closed
in
the
past
and
we
would
talk
about
this,
but
we
can
do
this
later.
I
think.
A
Okay,
you
got
it,
you
added
it
to
the
agenda
doc
later
on.
At
the
end,
okay,
sweet.
We
will
do
that
then,
okay
and
then
yuri.
I
think
you
added
this
note
here
about
official
edition
of
a
new
maintainer.
C
Yep
so
private
terraform,
I
got
a
new
maintainer
thanks
bob
for
all
your
contributions
and
accepting
their
all.
So
we
have
a
nice
and
faster
review
process
now
and
we
managed
to
get
in
right
after
the
bob
assignment
sample,
request,
merged
and
yeah
looking
forward
to
co-maintaining
provider
terraform
together
thanks
a
lot
bob.
A
Yeah
right
on
bobby
that's
fantastic.
It's
definitely
happy
to
hear
for
your
commitment
to
that
project
there
and
the
you
know
the
the
benefits
that
you're
bringing
to
it
as
well,
all
right,
sweet
okay.
So
I
wanted
to
have
a
note
here
about
your.
We
continue
to
have
more
providers
that
the
community
is
building
to,
for
you
know
new
scenarios,
or
you
know
new
environments,
etc
that
are
coming
into
the
crossman
ecosystem,
which
is
amazing,
and
I
love
every
single
one
of
those
that's
happening.
A
So
I
wanted
to
call
attention
that
we
haven't
really
had
like
the
policy
or
instructions
or
anything
well
laid
out
yet
for
that.
So
I
in
the
crossplay
and
slash
org
repo
there.
I
opened
a
quick
issue
just
to
kind
of
outline
what
we
do
today.
We
want
to.
You,
know
further
document
this
and
have
a
streamlined
process
for
it,
but
this
is
kind
of
a
quick,
real
quick.
A
This
is
the
broad
approach
that
we
take
for
folks,
that
want
to
bring
their
repository
into
the
crossband
contrib
organization
to
have
a
you
know,
vendor
neutral
and,
and
you
know,
governance
and
work
under
the
cross,
plane
project
governance
there
so
yep
we
have.
So.
If
folks,
you
know
want
to
do
that,
you
know
you
can
just
send
them
to
me.
I
think
christopher's
done
that
a
couple
times
he's
pointed
folks
over
to
me
to
help
facilitate
getting
those
providers
in
the
provider
the
palette
provider
was
one.
A
That
is
a
good
example
of
that.
That
was
added
recently
that
that
came
from
from
those
folks
there
that's
now
available.
I
think.
A
The
ci
and
such
in
there
we've
got
the
first
like
builds
out
of
the
main
branch
working
right.
Now
I
don't
think
an
official
release
has
been
done
yet,
but
that
provider's,
bootstrapped
and
already
ready
to
images
are
ready
to
be
to
be
pulled
and
used
now
and
then
there's
a
cops
provider
as
well
that
the
door
reached
out
to
us
and
we'll
go
through
that
process.
A
Start
it
today
to
get
the
brand
get
a
new
repo
open
for
it
and
accept
as
the
first
pr
all
the
content
from
that
repository.
A
So
more
providers
are
coming
along
from
the
ecosystem
and
we're
you
know,
have
a
bit
of
a
sketched
out
early
process
for
it,
but
we'll
flush
it
out
further
and
make
it
more
official
as
well,
but
thanks
to
everybody
in
the
community
who's
bringing
new
providers
into
the
ecosystem.
That
is
always
pretty
pretty
awesome.
A
B
I
I
added
one
thing:
it
would
be
really
great
if
we
can
create
a
new
release
and
provide
a
jet
aws,
because
I
think
the
last
release
in
jet
aws
is
from
march,
and
we
have
a
lot
of
pr
sits
there
and
a
lot
of
folks
running
between
the
versions
and
the
master
branch
and
yeah.
It
would
be
great
if
we
can
see
their
release.
A
B
A
Yep
that
that
seems
reasonable,
yep
good
thanks
for
calling
that
out,
okay
cool,
then
we
can
move
ahead
onto
the
community
section.
Then,
if
there's
nothing
else
on
providers
so
yeah.
So
you
know
I
like
this
section
here
for
every
every
community
meeting,
just
calling
out
any
sort
of
blog
posts
and
videos
etc.
From
the
last
couple
weeks,
I
only
quickly
in
a
quick
search.
A
I
only
found
this
one
here
that
that
somebody
had
published
for
the
in
the
community
there
about
their
experience
to
cross
in,
and
you
know
their
thoughts
on
on
where
it's
going
so
that's
available
here.
If
you
want
to
read
this
blog
post
about
crossplane
being
the
next
big
thing,
I
like
that
title
I
like,
where
the
thinking
on
that
one
and
it's
available
here
for
folks
to
be
able
to
read
any
other.
You
know
videos,
tutorials
posts,
etc.
A
That
folks
want
to
add
to
this
list
feel
free
to
do
so
because
it
definitely
serves
as
a
nice
like
aggregation
point,
for
you
know,
latest
news
and
stuff
about
cosplaying
all
right,
let's,
let's
jump
then
into
the
lfx
mentorship
program
updates.
So
I
think
both
of
our
mentees
are
on
the
call
today
and
so
I'll
give
an
opportunity
to
both
of
them
to
speak
about
the
latest
status
on
their
project
and
what
they'll
be
focusing
on
next
so
rika.
C
Right
so
hello,
everyone,
I
hope,
you're
all
doing
great
I'll
start
by
introducing
myself.
C
Actually,
I
wasn't
able
to
do
it
in
the
last
community
meeting,
so
I'm
ruini
from
india
and
I'm
currently
studying
bachelors
of
engineering
and
information
technology,
and
this
one
itself
I
got
selected
for
lfx
mentorship
for
crossplane
and
I'm
really
glad
to
be
a
part
of
crossplane,
and
so
currently
I'm
working
on
implementing
a
resource
and
a
controller
for
gcp
dns
policy
resource
and
for
the
same
I
actually
had
a
demo
session
with
mr
owners,
where
he
showed
me
the
implementation
step
by
step,
and
my
next
steps
would
actually
include
testing
out
crds
and
seeing
what
actually
causes
a
breaking
change
and,
as
this
would
be
directly
linked
to
my
lfx
mentorship
project.
C
A
Yeah
right
on
right,
good
yep,
that's
we
love
the
progress
so
far
and
we
love
that
you're.
You
know
what
you're
working
on
there
and
so
we're
looking
forward
to
continuing
to
you
know,
work
on
the
contributions
together
and
get
things
merged
so
yeah,
so
the
dns
policy
resource
and
controller
for
provider
gcp
will
be
hopefully
landing
sometime
soon
and
then
we
can
move
on
into.
You
know
the
meat
of
the
project
here
with
all
of
that
foundational
knowledge
to
work
on.
A
You
know,
detecting
breaking
changes
in
crds,
so
super
excited
about
that
and
welcome
to
the
community
erica
and
glad
to
have
you
here.
Awesome
then
parole.
So
I
think
you
are
on
the
call
as
well
today.
A
So
if
you
want
to
also,
you
know,
introduce
yourself
and
talk
about
the
latest
progress
in
your
project
as
well.
C
Sure
so,
hey
everyone,
I'm
carl,
and
I
am
a
bachelor
of
technology,
undergrad,
personal
media,
science
and
engineering
and
similar
to
week.
I
was
also
selected
for
lfx
mentorship
program
or
to
cross
plane,
and
my
project
basically
involves
adding
tests
for
pulling
packages
from
private
registries
on
and
different
oci
images
for
which,
until
now
I
have
been
working
around
testing
crosby
and
how
it
works
and
how
to
work
with
different
providers
and
stuff.
C
Then
we
had
a
bit
of
a
demo
session
with
daniel
and
jared,
where
they
a
sort
of
code
walk
through
through
the
pre-existing
tests
that
are
in
place
for
provider
updates
and
stuff
and
taking
inspiration
from
there.
I
would
be
working
around
some
tests
for
verifying
the
pulling
of
private
packages,
and
that
would
be
what
I
would
be
working
on
this
summer
and
yeah
it's
great
to
be
a
part
of
the
community
and
learn
about
different
stuff
that
was
happening
around
and
how
crosstalk
is
helping
bridge
different
gaps.
A
Right
on
parole,
yeah,
you
know
also
really
grateful
to
have
this
second
project
as
well,
and
both
of
these
I
think,
gonna
benefit
the
ecosystem
and
the
in
the
community
as
a
whole,
so
really
happy
to
have
you
there
and
the
progress
that
you're
making
on
getting
to
a
place
where
we
have
some
confidence
in
our
private
polling
packages
from
private
registries.
A
So
great
progress,
but
to
both
you
all
and
thank
you
very
much
for
both
of
it
all
right
so
adam.
I
think
you
are
here
to
talk
about
some
of
the
effort
that
the
cncf
is
initiating
to
do
some
fuzz
testing
and
some
security
focus
there.
Do
you
wanna
kind
of
give
us
a
little
bit
of
an
introduction
to
that
project
and
then
also
any
sort
of
context
about
how
we
can
help
make
that
successful
too
would
be
appreciated.
Also.
E
E
Okay-
okay,
that's
great!
So
yes,
hello,
everyone!
I
must
apologize,
I'm
in
the
middle
of
a
power
outage,
so
I'm
on
a
dial
in
the
line.
Here.
E
Please
interrupt
me
if,
if
the
connection
is
awful
anyways
to
yes,
the
context
of
forcing
is
that
the
cncf
is
investing
in
introducing
fussing
to
the
cncf
landscape
and
has
done
so
over
the
last
approximately
two
years
where
it
has
been
working
with
us
at
adalogics,
on
bearing
fusing
to
critical
projects
like
grenades,
helm,
envoy,
fluent
bit,
etc,
and
now
the
cncf
has
reached
out
to
us
to
hear
if
and
and
you
as
well
and
chris
anishtik
sent
out
an
email
last
week
as
if
I
remember
correctly
that
whether
you
were
that
would
be
something
that
you
were
interested
in
so
yeah.
E
So
this
this
is
kind
of
an
introduction
or
introductionary
to
talk
about
what
what
can
be
done
in
this
regard,
so
how
these
engagements
usually
work,
is
in
a
way
where
the
the
the
fundamental
infrastructure
set
up
around
fusing
a
given
cncf
project,
after
which
a
bunch
of
forces
are
written
to
bring
up
up
coverage
of
of
the
given
project.
E
And,
finally,
a
report
is
written
about
the
efforts
and
the
results
that
came
up
came
out
of
the
engagement
and
usually
it
takes
between
one
and
two
months,
and
the
maintainer
involvement
is
can
be
from
zero
to
basically
as
much
as
you
have
time
for
and
are
interested
in,
we
are.
We
are
able
to
carry
out
the
work
completely
without
any
any
anything
from
from
you.
E
But
but
if
you
have
any
wishes
and
and
thoughts
about
good
parts
of
the
code
base
that
that
would
benefit
from
fuzzing,
then
it
it
is
only
a
help
to
to
discuss
it
with
us
and
for
those
that
are
not
familiar
with
fussing.
It's
a
way
of
testing
software,
whereby
pseudorandom
data
is
passed
on
to
a
target
application
and
the
the
application
is
observed
for
any
crashes
and
bugs
some
of
which
can
be
can
affect
the
reliability
of
the
program
or
just
can
affect
the
security
of
the
the
program.
E
And
yes,
I,
yes,
we
have
had
a
look
at
the
cross
plane
initially
to
make
an
initial
assessment,
and
our
observations
are
that
crossplane
is
not
a
very
a
project
that
would
benefit
as
much
from
fuzzing
as
other
projects,
and
this.
E
This
is
just
an
initial
assessment
in
the
sense
that
we
haven't
found
a
lot
of
data
processing
and
marshalling
and
passing
in
the
in
the
code
base
that
there
are
a
few
apis
that
that
do
this,
but
in
in
general
the
code
is
not
super
complex
and,
and-
and
this
is
something
that
I
would
be
very
interested
in
hearing
your
your
input
from.
A
Yeah
right
on
adam,
that's
all
really
really
useful
and
thank
you
for
the
introduction
to
you
know
the
goals
there
and
some
of
the
approach
to
what
one
of
the
big
questions
I
had
about
to
understand
the
effort
a
little
bit
more
was
like
what
what
level
do
you
typically
do
the
fuzz
testing
at?
E
Yes,
so
that's
a
good
question,
and
that
depends
a
little
bit
on
you
not
not
on
you,
but
oh
that
that
you
have
an
influence
here
as
well
in
terms
of
what
you
think
makes
sense,
and-
and
here
here
the
the
thread
model
is
definitely
a.
E
Denominator
project:
for
years
you
have
to
have
some
opinions
on
that
and
in
terms
of
the
specifics
we
have
first
reconciles
before
where
we
were
simply
asked.
You
know
only
only
first,
the
exposed
and
exported
apis,
because
that's
that's
the
only
thing
that
makes
sense,
and
if
you
do
first,
the
internal
parts
of
a
given
application.
A
Got
it
got
it
yeah?
That's
I
would
I
have
a
couple
ideas.
I
could
share
real
quick
and
then
we
can
like
well
one
thing
process
wise
is
that
you
know
in
order
to
so.
This
is
a
really
good
introduction
in
the
community
meeting
here
and
we
have
a
better
understanding
of
you
know
the
effort
that's
going
on
here
and
then
folks
that
are
excited
about
it
as
well.
Can
you
know
get
a
chance
now
with
this
awareness
to
participate
in
it
as
well?
A
The
so
like
a
high
level
thing
that
might
be
useful
is:
are
you
all
taking
our
approach
of
potentially
writing
down,
like
you
know,
like
a
one
pager
or
something
like
that,
that
captures
some
of
the
scenarios
that
will
you'll
be
focusing
on
and
then
like
we'll
have
a
chance
to
weigh
in
and
have
feedback
and
discussion
on.
That.
Is
that
something
reasonable
that
you
all
normally
do
or
do
you
know
me
you
know?
Is
it
more
of
a
meet
together
and
collaboratively?
Do
these
things
what
works
best
for
you.
E
All
right,
okay,
yeah,
so
we
are
definitely
open
to
writing
an
initial
assessment
of
of
what
would
make
sense
for
crossplane
that
that
would
also
be
a
great
idea
to
get
our
heads
together,
but
we
don't
so
usually
like.
I
said
the
engagement
takes
usually
between
one
and
two
months,
and
it
is
very
iterative
in
the
sense
that
we
will
write
a
bunch
of
fosters,
observe
the
feedback
from
those
write,
some
more
that
do.
E
Do
it
better
or
differently
and
and
and
kind
of
repeat,
repeat
that
process
and
we
we
usually
participate
in
the
community
meetings
to
to
you
know,
get
feedback
from
from
the
community
and
but
but
at
the
same
time
there
will.
There
will
be
also
be
chances
to.
You
know,
communicate
asynchronously
through
slack
or
through
any
bug
report
that
may
be
reported,
and
I
should
have
mentioned
that
it
is
a
goal
for
this
in
the
the.
E
A
Got
it
got
it
got
it
yeah,
so
that
makes
total
sense,
and
you
know
my
opinion
typically
here
is
that
you
know
you've
done
this.
Like
you
and
your
group,
there
has
done
this
fuzz
testing
for
other
cncf
landscape
projects
before
and
so
you
have
a
very
good
and
you
know,
sense
and
domain
expertise
on
how
to
approach
these
things.
So
you
know
taking
an
iterative
approach.
A
If
you
write
some
fuzzers
you,
you
know,
if
you
could
share
those
thing
with
the
community
of
like
hey
here's,
what
we've
written
you
know
is
there
feedback?
So
if
we
get
it
a
bit
like
it
doesn't
it's
my
point,
is
it
doesn't
have
to
be
in
the
form
of
like
an
upfront
one
pager,
I'm
totally
fine
with
that.
But
you
know
the
sharing
the
initial
work
and
results
that
you're
finding
so
that
we
can
iterate
together
and
folks
from
the
community
can
contribute
their
ideas.
A
That
sounds
like
a
great
approach,
so
yeah
that
sounds
that
sounds
wonderful.
E
Great,
I
I
should
add
here
so
so,
for
example,
this
box,
that
was
in
the
we
this
box,
that
was
discussed
in
the
in
the
start,
beginning
of
this
meeting,
that
that
is
something
that
a
foster
would
be
able
to
find.
I
think
I
think
it
was
it's
a
simple
new
reference
symbol,
as
in
isn't
isn't
the
other
reference
and
that
that's
something
that
foster
will
be
able
to
find.
So
if
you
care
about
such
bugs
and
panics,
then
then
I
believe
fussing
will
be
able
to
to
help
help
the
project.
A
Awesome,
yep
yeah.
I
have
no
question
that
you
know
there
will
be
some
value
that
comes
out
of
the
findings
here
and
you
know
improving
like
the
reliability,
resiliency
and
security
of
the
project
is,
I
I
see
only
good
things
coming
out
of
this,
so
I'm
excited
for
this
effort.
E
Perfect,
okay
yeah,
so
I
suggest
that
we
so
so
there
is
a
pr
open
on
oss
first
from
last
year.
I
believe
it
has
stalled
a
bit.
I
think
it
would
make
sense
to
get
that
one
merged.
E
I
can
attack
tag
relevant
people
or
make
a
note
about
that
one,
and
and
after
that
we
we
have
kind.
We
have
kind
of
had
the
freedom
to
to
to
improve
on
and
write
more
forces
to
to
the
to
the
setup
on
on
those.
So
would
that
be
possible
from
maintainers?
I
think
nick
and
someone
else
was
involved
in
in
that
pr.
E
I
think
two
other
two
other
maintainers.
So
if
we
could
get
that
merged,
then
we
are
ready
to
go
from
our.
A
Yeah,
that's
great
yeah,
I
kind
of
you
know,
fell
off
my
radar.
I
guess
because
it
was
it's.
You
know
in
a
different
org
there,
so
yeah.
A
So
if
you
would
like
tag
the
people
that
are,
you
know
at
mention
the
folks
that
are,
you
know
contributing
to
that
pr
already
so
we'll
get
another
notification
about
it
and
then
feel
free
to
drop
a
a
link
to
existing
pr
in
oss
fuzz
here
in
the
agenda
document
as
well,
and
then
we
can
follow
up
on
that
and
get
that
you
know
drive
that
to
get
merged
in
and
then
we
can
iterate
after
that.
Yeah!
That's
perfect!
Thank
you
very
much.
A
A
All
right,
all
right
so
next
and
then
also
thanks
for
persevering
through
a
power
outage
to
talk
to
us
about
it
too
man.
A
B
Yeah,
that's
correct.
Let
me
check
if
I
have
the
other
one
give
me
a
second.
B
B
Yes,
for
us
so
forth
for
for
tagging,
resource
in
ec2,
there's
an
api
available
for
tagging,
resources
which
are
not
managed,
for
example,
from
crossplan
or
terraform,
so
that
we
can
bring
tax
on
resources
in
ec2,
and
we
have
one
or
two
needs
for
this
internally.
So
one
thing
is,
we
will
roll
out
carpenter
in
our
setup,
so
the
new
I
call
it
autoscaler
from
aws-
and
the
thing
is
aws-
creates
this
eks
security
group
from
a
cluster
perspective
and
we
need
to
add
attack
there.
B
Otherwise,
carpenter
is
not
working
and
the
other
thing
is.
We
are
creating
a
lot
of
transit
gateway
attachments
and
if
we
create
transit
gateway
attachments
in
the
transit
gateway
account,
the
attachments
will
come
without
any
text
and
we
will
re-tag
and
we
need
to
re-tag
the
things
there
and,
from
the
other
account
perspective,
we're
not
managing
the
the
transit
gateway
attachments.
We
only
need
to
manage
the
text,
and
I
saw
that
in
the
jet
providers,
because
of
terraform
the
the
tax
in
ec2
is
available
as
a
resource.
B
That's
why
I
have
the
feeling
that
we
can
reopen
the
pr
here
in
the
provider.
Aws.
Also,
and
can
sort
a
little
bit
if
this
is
good
and
with
a
few
informations.
That
is
only
for
resources,
we're
not
actively
managing.
Because
of
then
we
have
problems
between
both
reconcilers
that
that
we
will
overwrite
the
text
or
remove
tags,
but
we
definitely
need
this
for
resources.
We
are
not
managing
with
crossplane
directly.
A
This
is
a
christopher.
Does
that
also
your
last
statement
there
does
that
also
imply
that,
like
there's
a
need
for
observe,
only
resources
to
be
implemented
too,
or
does
this
not
depend
on
that?
I'm
not
sure
if
I'm
fully
I'm
kind
of
coming
up
to
speed
here.
So
I'm
not
sure
if
I
fully
understand.
B
A
Okay,
I
think
I
think
I
understood
that
cool.
So
so,
then,
are
you
so
this
is
you're
you're,
making
a
proposal
here
that,
like
this
functionality
that
was
closed,
like
is
now
needed,
yeah
again
for
the
proprietary
us
and
the
resources
there,
okay,
cool,
do
you
do
you
want
like
some
feedback
on
that?
Or
do
you
want
to
like
go
ahead
and.
B
I
think
it
would
be
great,
I
think
I
think
nick
was
on
this
and
also
I
think
carl
was
on
this
with
ideas
or
things
in
the
pa
yeah.
Oh.
B
A
Got
it
yeah,
I
think
yeah,
since
this
was
a
pr
that
was
was
closed
like
oh?
Well,
I
don't
I
don't
know.
Maybe
maybe
a
good
way
to
do
this
is
to
reopen
it,
but
maybe,
like
another
good
way,
is
to
capture
an
issue.
If
this,
the
issue
that
this
is
for
doesn't
really
capture
like
the
core
of
this
idea,
it
might
be
a
good
idea
to
open
up
a
new
issue
and
then
like
tag
carl
and
nick
on
it,
so
they
can
have
some
share
their
opinions
on
it
and
moffat.
A
Do
you
already
have
an
opinion
on
this
that
you'd
like
to
share
at
this
time,
or
are
you
maybe
not
up
to
speed
on
it
just
yet
to
have
an
opinion.
D
The
problem
is
that,
like
you
know,
for
example,
we
have
this
high
fidelity
convention
where
we
either
have
an
inline
field
or
the
crd,
and
we
prefer
having
it
as
separate
crd,
for
example,
if
you
have
any
cluster
resource
instead
of
and
and
the
api
also
supports
node
group,
even
if
the
api
cluster
exposes
node
groups,
we
don't
add
that
to
the
cid
so
that
it's
a
different
crt
for
both
node
group
and
cluster,
and
you
can
manage
either
one
of
them
through
the
other.
D
So
I
think
this
is
a
little
bit
similar
to
that.
The
fact
that,
like
you
know
a
lot
of
resources,
have
the
tagging
and
like
that
was
kind
of
like
an
exception
that
we
wanted
to
make
like
you
know.
Hey
tagging
is
like
in
a
very
so
specific.
D
So
in
this
case
I
think,
like
you
know,
having
tags
on
individual
resources
is
kind
of
like
the
pattern
that
we're
following
and
introducing
a
separate
attack.
Controller
might
result
in,
like
you
know,
racing
conditions
that
we
wanted
to
prevent
with
that
high
fidelity
convention.
D
C
A
Yeah
that
sounds
like
that
might
be
a
good
idea
christopher.
If
you
want
to
like
capture
this
as
like
its
own,
like
a
distinct
issue,
that's
open
and
then
tag
you
know
yeah.
I
tag
carl
and
nick
and
muafik
on
it
to
get
some
discussion
on.
That
sounds
like
a
good
next
step.
B
Yeah
cool,
I
I
will
open
the
issue
and
yeah
add
a
few
things
there.
So
from
my
two
comments,
and
then
we
can
have
a
look
yeah
awesome.
That
sounds
good.
A
B
A
A
Got
it
got
it
got
it
yeah,
that's!
That
is
a
possibility
thanks,
yeah
cool
all
right.
That
sounds
good
christopher.
You
can
follow
up
on
that
thanks
for
making
that
up
man,
okay
and
then
yes,
and
then
we
had
like
back
to
the
czech
div.
A
Failing
thing
you
know
in
the
I
don't
know
exactly
what's
going
on
here,
but
in
the
past
you
know
I
have
approached
like
build
ci
failures
that
you
know
only
happen
in
the
build
ci
environment,
with
like
adding
some
extra
like
output
statements
or
something
like
to
the
build
sub
module.
And
then
you
know
opening
up
a
pr
that
has
you
know
that
updated
build
sub
module,
so
we
can
get
ci
to
run
it
and
get
that
extra
output
there,
since
it
only
happens
in
ci
environments.
A
Cool
yeah,
I'm
super
curious
about
what
the
what
is
causing
this
it's
a
mystery,
okay,
cool.
So
I
think
that
was
everything
that
was
on
the
agenda
doc
here.
Any
other
topics
that
folks
want
to
bring
up
before
we
adjourn.
B
So
I
have
only
one
thing
we
had
also
in
our
own
environment-
and
I
also
see
this
from
days
to
days
in
the
community.
B
So
if
you
update,
for
example,
the
provider
aws
or
other
one
so
changing
the
version
number,
then
it
then
it
occurs
an
error
that
the
crd
is
currently
managed
from
older
uid.
So
from
the
old
one,
and
the
only
thing
as
to
to
solve
the
issue
is
to
add
more
cpu
in
cross
plane
core
and
at
as
argument
max
reconciliation
rate
to
100
or
more.
B
A
B
Is
something
like
that
that
roll
out
the
new
provider
version
you
get
only
an
issue
in
the
in
the
pk
reference
that
the
following
crds
are
currently
managed
managed
from
other
uid
and
then
you
are
lost,
then
nothing
happens
anymore.
A
More
resources
or
like
more
concurrent
reconciliations,
fixes
that
or
addresses
it.
A
A
Yeah
yeah,
it
feels
like
something
that's
kind
of
that.
I
I
have
heard
of,
and
I,
and
I
have
a
recollection
of
it
is
was
there
is
there
is
a
distinct
tracking
issue
for
for
this.
That
was
open
already.
A
Let
me
check
I
yeah.
That
might
be
good
to
add
to
the
add
to
the
agenda
doc
here,
so
we
can
maybe
follow
up
with
with
other
attention
on
that.
B
C
A
Well,
something
timing
out,
at
least,
can
I
establish
control
of
objects,
connect
static
control,
yeah.
B
If
you
scroll
a
little
bit
down,
I
have
also
from
the
package
revision
level
the
the
messages
you
only
see
as
an
event
is.
A
Which
one
this
this
was
yours
right,
where's,
christopher
yeah,.
B
Yeah,
you
can
see
here
the
events
and
the
last
event
is
the
only
thing
you
get
there
and
then
it's
completely
blocked.
If
you're
not
doing
anything,
you
see
blah
blah
is
controlled
by
the
following
uid
and
then
everything
stops
and
the
only
thing
you
can
do
at
this
point
is
you
need
to
modify
the
crd
mentioned
there,
remove
the
old
uid
and
then
give
crossplane
core
more
cpus
at
max
reconciliation
right
as
argument,
and
then
in
the
next
loop.
Everything
is
working
as
expected
and.
A
Yeah
yeah
yeah,
okay,
I'm
glad
you
brought
this
up,
then
christopher,
because,
like
it,
so
it
looks
like
the
the
error
message
might
be
a
little
bit
different
in
terms
of
like
a
a
resource
already
being
having
like
ownership
from
another
object,
but
like
the
the
the
core
thing
is
or
the
the
bigger
issue
here
is
that
you
know
the
the
it's
timing
out:
it's
it's
not
finishing
its
operations
in
in
the
given
time
that
it
has.
A
So
this
feels
very
related
to
this
issue
here
that
we
already
have
it's
in
1.9.
I
don't
know
if,
if
that's
yeah
like
what
nick
is
talking
here
like
it
seems
totally
related.
So
maybe
this
is
something
we
can
we
can
get
in.
It
looks
like
something
we
had
been
thinking
about
for
1.9,
so
at
the
very
least
like
I
agree
with
you
that
you
know
this
issue,
you
brought
up
in
that
it's
very
related
to
this
issue.
A
That
is
a
priority
that,
like
yeah,
this
sounds
like
something
that
we
should.
Let's
create
a
link
between
these
two,
oh
crossdressing,
two,
five,
six,
four!
Is
that
oh,
that's
this
one!
Okay!
So
they
are
already
linked
together:
okay,
yeah,
okay,
so
yeah,
then
maybe
we
could
like
for
easier
tracking.
We
can
add
this
to
1.9
also,
so
it
doesn't
fall
off
the
radar
but
yeah.
Okay.
I
think
this
is.
This
is
getting
a
little
bit
of
traction.
Then.
C
A
Cool
awesome
all
right.
Well,
if
there's
no
other
agenda
items
here,
then
we
can
go
ahead
and
adjourn
for
the
the
week
and
then
we
will
we're
intending
to
have
1.9
out
before
the
next
time
we
meet
in
two
weeks.