►
From YouTube: 2022-06-02 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
I'm
on
view
the
recording
has
started-
and
this
is
the
june
2nd
2022
cross
playing
community
meeting
folks
feel
free
to
continue
adding
yourself
to
the
attendees
list
and
also
to
agenda
items
into
the
agenda
here.
If
you
have
any
that,
you
want
to
discuss
I'll,
go
ahead
and
drop
into
the
chat.
Now
the
zoom
chat
here
a
direct
link
to
the
agenda
document.
So
if
you're
not
in
there,
then
you
can
hop
on
in
now
and
add
add
any
suggestions
that
you
want.
A
So
we'll
go
ahead
and
start
here
as
usual
with
recent
milestones
and
releases.
So
we
did
during
kubecon
yeah
right.
It
was
like
the
tuesday
of
kubecon.
In
valencia
we
did
the
1.8
release,
so
that
was
after
the
last
community
meeting.
So
that
is.
That
is
a
new
thing.
So
a
couple
of
fixes
and
improvements
that
were
in
from
the
general
community
are
included
here.
A
You
can
read
about
them
in
the
in
the
release,
notes
and-
and
I
also
wanted
to
call
out
some
of
the
the
first
time
contributors
to
this
repo
bob.
I
know
you've
been
contributing
a
bunch
of
different
repos,
but
it
looks
like
this
was
your
first
contribution
to
core
crossplane
as
well,
so
thanks
for
that
definitely
appreciate
that
right
on
cool,
so
yeah.
So
the
release
notes
can
be
read
about
here,
so
that
was
the
the
minor
release,
the
1.8.
A
We
also
did
a
set
of
patches
as
well,
so
the
the
versions
of
cross
plane
that
are
under
active
maintenance
right
now
or
the
last
three.
That's
the
typical
policy,
so
1.8,
1.7
and
1.6
all
got
a
series
of
patch
releases
to
fix
a
panic
issue
that
was
happening
with
a
logger
that
had
some
invalid
arguments.
So
that's
that
got
fixed
and
merged
and
then
released
as
these
three
patch
releases
right
here.
A
So
we've
done
the
1.8
release
and
then
a
series
of
three
patches
to
the
three
most
recent
crosstalk
versions
since
the
last
community
meeting.
So
we're
kind
of
early
since
we
just
did
1.8
we're
kind
of
early
in
the
cycle
for
for
1.9,
with
the
typical
release
cadence
of
every
eight
weeks
that
puts
the
next
release.
It's
the
july
12th
time
frame.
A
I
don't
think
we've
done
a
whole
lot
of
thinking
right
now
about
what
would
be
included
and
what
the
main
focuses
are
so
we're
kind
of
early
on
that,
but
we
can
kind
of
listen
to
some
some
priorities
today,
while
we're
talking
about
it.
So
we
got
a
couple
couple.
I
wrote
a
weekend
or
so
so
we've
got
you
know
a
number
of
weeks
left
in
it.
They
do
the
I
started
just
today,
a
project
board
and
carried
over
some
things
from
1.8
that,
oh
that's,
odd.
B
C
A
Let
me
see
if
I
can
fix
that
real
quick
in
with
the
login
account.
A
C
About
that
delay
here
I
am
able
to
see
it,
but
it
seems
like
it's
private.
A
B
C
A
A
There
we
go
cool
so
yeah.
So
just
I
just
added
a
couple
of
the
issues
on
here
that
are
ongoing
priorities
and
that
there
are,
you
know,
there's
outstanding
work
on,
so
you
know
there's
a
couple
of
of
things
that
are
being
invested
in
in
the
crossland
runtime,
the
the
core
there
that
powers
of
all
the
rest
of
the
functionality
in
crossplane.
So
I
know
this
has
been
a
frequently
requested
issue
from
another
number.
A
Folks
and
there's
an
ongoing
pull
request
for
that
sergeant
is
still
driving,
so
I
think
this
would
be
something
that
should
be
included
into
the
1.9
time
frame.
Custom
compositions
is
an
ongoing
effort
as
well.
A
You
know
surrogate
and
nick
have
been
investing
in
that
over
a
number
of
a
number
of
weeks
now
so
that'll
still
be
a
general
focus
for
continuing
to
drive
that
functionality
from
the
poc
stage
to
something
that's
more
ready
for
general
consumption
and
then
I
think,
there's
an
ongoing
pull
request
as
well
for
general,
like
patching
from
common
data
sources.
That's
a
highly
highly
requested
feature
in
in
general
for
across
planes.
A
So
I
think
all
those
will
be
getting
some
we'll
be
getting
some
some
attention
in
the
1.9
time
frame.
Are
there
other
things
that
are
like
priorities,
or
you
know,
desires
from
the
community
like
that?
We
should
be
thinking
about
for
the
1.9
timeframe
as
we're
getting
getting
off
the
ground
and
getting
moving
with
it.
Now.
D
A
Yeah-
and
you
know
the
composition-
revisions
have
been
in
alpha
so
far
with
their
initial
functionality
and
it's
a
bob.
There
are
specific
functionality
or
features
within
the
composition,
revisions
that
you're
looking
to
see
as
they
continue
to
get
matured.
D
I'm
I'm
real
early
in
my
in
my
study
of
the
of
the
revisions.
You
know
we're
just
getting
we're
just
at
the
point
where
we're
starting
to
get
things
deployed
and
we're
not
quite
yet
to
the
point
where
people
say:
okay,
that's
great!
Now
what
can
you
do
with
it?
But
I
know
that's
coming
so
you
know
the
we
need
to.
We
personally
need
to
start
doing
some
thinking
around.
A
Yeah,
it's
some
really
good
questions
about
the
life
cycle
of
compositions
and
how
they
evolve
and
iterate
on
them
over
time
and
so
bob.
It
wouldn't
surprise
me
at
all,
as
you're
going
down
that
path
that
you're
going
to
have
some
good
feedback
on.
You
know
either
missing
functionality,
or
you
know
the
way
that
things
should
be
approached
there
for
environments
such
as
yours.
So
if
you
run
into
things
you
know
with
what
is
out
there
right
now
for
composition,
revisions
that
you
know
you
want
to
provide
some
feedback
on.
A
D
A
Awesome:
that's
great
bob
cool
cool
cool,
so
yeah
early
in
1.9
time
frame,
but
yeah
we've
got
the
the
release
dates.
You
know
set
there
for
the
typical
cadence
and
we
can
keep
the.
C
A
Board
up
to
date
as
we're
moving
along
through
the
cycle
here
cool,
so
we
can
move
along,
I
think,
to
providers.
Then
I
wanted
to
definitely
call
out
that
the
per
all
the
providers
have
been
moved
from
the
core
cross
plane
organization
to
the
to
the
crossblink
and
trip
org.
A
So
that's
as
part
of
the
recent
charter
updates,
you
know
providers,
we
have
a
neutral
home
for
collaboration,
and
you
know
folks
working
together
on
these
community
providers,
so
the
details
about
that
can
be
found
in
this
issue
here
and
that
has
been
executed
so
that
these
providers
have
been
moved
to
their
new
homes
in
the
cross,
plain
contrib,
repo,
all
the
ci
and
you
know
publishing
the
builds
and
everything
has
been
updated.
A
So
I
believe
all
of
that
should
result
in
in
you
know,
usable
builds
and
everything
should
be
working
still
if
there's
any
issues
with
folks
that
folks
have
with
with
consuming
or
contributing
to
or
anything
with
these
providers.
Now
that
they've
moved
to
their
new
home
in
the
cross,
blind,
contrib
organization,
do
let
us
know
so
we
can
get
that
fixed
and
make
sure
that
there's
not
a
lapse
in
in
these
providers.
That
folks
are
depending
on.
A
Okay
and
then
for
recent
releases
from
from
providers,
the
one
I
was
thinking
of
immediately
is
there's
a
0.19
release
for
provider
azure
here
which
we
can
see
the
release
notes
on
that
is
since,
since
the
last
community
meeting.
So
I
just
left
a
link
to
the
release
notes
here,
so
folks
can
dive
in
and
dig
into
that
there's
a
few
new
resources
that
were
added
and
a
couple
of
updates
and
fixes
for
some
some
key
scenarios
there
were
there.
A
Other
recent
updates
for
the
folks
are
sorry
recent
releases
for
other
providers
in
the
community.
That
folks
want
to
call
out.
You
know:
there's
a
number
of
of
of
providers
out
there
and
wanted
to
make
sure
that
there's
recent
releases
that
folks
have
done
that
they
get
called
out
here
to
get
a
little
noise
on
them.
B
B
A
Sounds
good
any
other
any
other
releases
of
other
providers.
Recently,
the
folks
want
to
call
out.
A
We'll
well
we'll
we'll
main,
we'll
continue
to
have
this
space
in
the
community
meeting
for
folks
to
talk
about
developments
of
providers-
and
you
know
new
releases
and-
and
things
like
that,
so
the
space
will
be
continue
to
be
open
in
previous
or
sorry
subsequent
community
meetings-
okay,
sweet,
so
I
guess
we
can
jump
into
community
topics
now
also,
so
there
was
like
typically
in
the
content
section
here
I
like
to
call
out
you
know
all
sorts
of
blog
posts
and
videos-
and
you
know
all
sorts
of
stuff
that
is
happening
within
the
crossplane
ecosystem
and
you
know
folks
adopting
it
and
telling
their
stories
and
what
they're
building
etc.
A
A
ton
of
those
recent
ones
are
all
kubecon
stuff,
so
they're
a
little
bit
lighter
this
time
on,
like
blog
posts
and
in
other
videos
and
such
because
you
know,
basically,
everything
in
the
last
couple
weeks
has
been
all
about
kubecon.
So
I
think
this
will
be
a
good
section,
maybe
to
to
directly
link
to
the
kubecon
talks
that
focused
on
crossplane
in
valencia.
A
I
think
there
were
like
something
like
eight
or
nine
or
so
talks
that
we're
talking
about
crossplane
and
the
various
things
that
folks
are
building
with
it
and
some
of
the
internal
details
as
well.
So
I
think,
there's
a
whole
bunch
of
super
interesting
content
to
dive
into
now
that
those
talks
have
been
released
and
published
by
the
cncf.
A
So
we'll
add
some
links
to
this
section
of
the
agenda
here
to
that's
specifically
linked
to
all
the
of
the
talks
and
feel
free
folks
to
if
you
had
a
talk
there
to
go
ahead
and
add
a
link
directly
to
the
video
in
the
agenda
doc,
we
can
crowdsource
that
effort
a
little
bit.
A
All
right
during
the
coupon
time
frame,
I
think
there
is
not
a
whole
lot
to
report
on
with
custom
compositions.
You
know
the
design
updates
are
still
being
updated
in
this
pull
request
here,
with
ongoing
thoughts,
results
of
pocs.
You
know
lessons,
we're
learning,
etc.
A
It's
pretty
complex
feature
and
a
lot
of
moving
parts
in
it.
So
you
can
follow
along
with
some
of
the
updates
on
that
people
request
that
nick
and
sturgeon
are
keeping
up
to
date.
I
don't
think
either
of
them
are
on
the
call
today
to
get
more
details,
but
yeah
you
can.
You
can
subscribe
to
this
year
and
dig
in
and
catch
up
with
some
of
the
updates
there.
A
So
a
lot
of
folks
were
in
valencia.
It
was
really
cool
to
meet
some
of
you
all
and
spend
some
time
together
in
person.
I
wanted
to
hear
from
some
folks
that
were
there
participating
in
neither
the
talks
or
being
sorry
being
at
the
excuse
being
at
the
cross,
plain
booth
or
you
know,
just
being
around
in
general
in
the
ecosystem.
Did
anybody
want
to
share
some
of
their
observations
or
experiences
from
for
being
at
kubecon?
Anything
to
share.
C
So
that's,
like
you
know,
a
refreshing
talk
about
how
crds
work
and
how
upstreams,
how
upstream
handles
it
and
yeah.
If
you
wanna,
if
you
want
to
build
the
crosstalk
provider
iron
hassan,
we
have
a.
We
have
a
talk,
editorial
format,
90
minutes,
and
then
there
is
the
maintainer
session,
with
yuri
stephen
and
and
increase
the
fur.
A
Yeah,
I
think,
there's
all
all
different
levels
of
content
for
for
folks
to
be
checking
out
right
because,
like
just
in
the
maintainer
track,
talk,
there's
like
an
introduction
to
crossplane
for
people
that
aren't
even
familiar
with
the
project
and
then
there's
a
bit
more
of
a
deep
dive
into
some
of
the
internals
and
specifics
as
well.
Just
within
that
talk
and
then
yeah
you've
got
talks
like
out
there
of
you
know
how
we
how
we
essentially
broke
the
api
server
with
like
scaling.
A
You
know
in
the
load
from
the
number
of
crds
we're
installing,
etc.
That
goes
into
some
really
deep
details
about
more
upstream
kubernetes
stuff
and
the
api
server
and
std,
and
you
know
performance
and
stuff
like
that,
so
a
whole
bunch
of
really
really
cool
content
there.
One
of
one
of
the
things
that
I
was
kind
of
surprised
about
at
kubecon
is:
is
you
know
that
the
the
conference
organizers
were
talking
about
how
it's
like
70
of
the
attendees?
A
It
was
their
very
first
kubecon
ever
and
so
that
made
me
think
that,
like
oh,
my
gosh
there's
like
a
lot
of
people
here
that
are
just
getting
started
like
that.
Aren't
super
familiar
with
what's
going
on
in
the
kubernetes
landscape,
but
I
didn't
find
that
to
be
the
case
actually,
because
I
think
that's
that
statistic
is
perhaps
a
bit
misleading,
because
you
know
we
haven't
really
had
a
lot
of
qcon
opportunities
to
visit
in
the
last
two
years.
A
So
I
think
people
have
been
adopting
cloud
native
technologies
and
you
know
using
cncf
projects
and
building.
You
know
running
their
workloads
on
kubernetes,
etc
for
a
couple
of
years,
but
just
haven't
been
to
a
kubecon
yet
so
you've
got
70
of
people
that
are
there
that's
their
first
kubecon
but
like
they're
already
building
stuff,
they
already
are
absorbed
and
involved
into
the
ecosystem.
A
B
A
A
lot
of
the
conversations
that
we're
having
too
just
a
lot
of
people
a
lot
of
a
lot
of
expertise,
a
lot
of
experience.
It
was
really
cool
to
share
that
with
folks.
A
So
hopefully,
we'll
be
able
to
see
some
more
people
at
the
next
q
con
as
well,
the
north
america
one
that's
in
detroit.
You
know
I'm
I'm
hoping
to
be
there
myself
as
well
and
get
to
see
people-
and
I
guess
that's
a
good
reminder
too,
is
that
the
cfp.
A
The
call
for
proposals
for
kubecon
north
america
is
the
deadline,
is
tomorrow
friday
june
3rd.
I
think
probably
midnight
pacific
time.
Don't
quote
me
on
that?
One,
that's
what
it
typically
is,
and
so,
if
anybody
wants
to
submit
a
talk
about
crossplane
or
integrations
with
the
rest
like
other
projects
in
the
ecosystem,
or
anything
like
that,
I
am
happy
to
personally
review
and
give
feedback
on
people's
proposals.
So
you
can
just
directly
reach
out
to
me.
Send
me
a
dm
or
whatever
and
I'll,
be
happy
to
review
your
proposal.
A
Each
each
kubecon
has
its
own
local
flavor.
Detroit
will
have
its
own
local
flavor
as
well.
It'll
be
cool.
I've
been
to
that
city
in
a
while.
Last
time
I
went
was
for
a
red
wings
game
and
that
was
pretty
sweet.
I
had
a
very
good
time
this
detroit's
pretty
tight,
all
right
so
yeah.
That's
everything
for
kubecon,
then.
So
the
next
agenda
item
that
I
see
is
about
this
pull
request
here.
A
B
So
I
added
this
to
the
agenda.
This
is
not
actually
my
pull
request,
co-worker
of
mine
and
I'm
just
kind
of
interested
in
to
hear
what
is
there
anything
that
we
can
add
or
anything
steps
that
we
can
take
to
help
push
this
push
this
forward.
A
Yes,
good
question:
let
me
let
me
see
if
there
is
okay,
I
guess
it
hasn't
really
gotten
much
like
feedback
on
it
at
all.
Yet
right,
so
that's
yeah.
B
It's
less
first
step,
yeah
less
about
the
actual
content
of
the
pr
more
just
about
the
process
of
how
we're
going
about
what
we
do
differently.
Maybe.
A
Yeah
yeah
yeah,
so
I
think
yeah,
like
everything
here,
looks
to
be
pretty
reasonable
in
terms
of
what
it's
covering
and
what
it's,
what
it's
like,
describing
the
content
and
how
it's
been
tested,
et
cetera.
So
I
think,
like
in
in
general,
you
know
with
requests
across
a
number
of
different
providers.
Some
providers
get
more
attention
than
other
ones
do
like
provider.
Aws
has
a
whole
lot
of
folks
looking
at
it
and
traction,
and
you
know
continuous
updates
to
it,
etc.
A
So
gcp,
maybe,
is
a
little
bit
less
than
than
aws.
So
yeah.
I'm
sorry
that
this
has
fallen
to
the
cracks
here,
but
we
can,
I
don't
think,
there's
anything
about
the
approach
here
that
is
is
preventing
it
from
from
getting
getting
traction.
I
think
it's
more
of
just
you
know.
Community.
C
A
And
maintainers,
you
know
spread
out
across
a
number
of
different
providers
and
this
one
hasn't
gotten
attention.
Yet.
B
I
understand
also
kubecon,
as
you
said,
yeah
very
busy,
so
yeah,
just
thanks.
Thanks
for
that
cool.
A
A
Let's
go
ahead
and
bring
this
one
up
too,
for
bob.
D
D
It
it
blocks
up
the
the
workspace
so
that
it
can't
you
know
it
can't
do
any
more
work
without
some
manual
intervention.
This
was
just
one
idea
I
had
to
basically
generic
make
it
make
it
generically
possible.
To
put
you
know,
post
start
and
pre
pre-stop
hooks
in.
I
know
there
was
a
suggestion
to
maybe
do
something
in
the
provider
itself,
and
I'm
just
you
know,
I'm
just
kind
of
looking
for
maybe
some
pointers
on
how
to
catch
signals.
D
You
know
in
in
the
in
the
provider
itself
and
then
do
something,
because
I'm
not
a
go.
I'm
not
a
go
expert
by
any
stretch
of
the
imagination,
and
so
I
was
just
you
know,
just
looking
for
some
ideas
on
the
best
way
to
you
know
catch
the
the
terminate
signal
so
that
we
can
cleanly
shut
down
the
terraform
cli
commands
before
the
pod
goes
away.
A
Got
it
got
it
got
it
yeah
that
makes
sense
bob
like
what
the
what
the
like
scenario
is
here.
Muafik.
Do
you
have
some
some
opinions?
You
can
share
about
like
general
approach
here
for
being
like
being
able
to
deal
with
life
cycle
events
for
for
the
the
providers.
C
But
we
do,
we
do
handle
the
this
state
in
different
ways
in
intel
yet,
but
that
is
because,
like
you
know,
we
kind
of
hide
away
a
lot
of
telephone
details.
You
wouldn't
even
know
if
it's
running
terraform
or
not
behind
the
scenes,
but
in
this
case,
like
your
terraform,
is
exposed
to
the
users
in
the
in
the
provider
terraform.
C
So
it's
a
little
bit
like
you
know
more
challenging
to
to
to
play
with,
like
you
know,
tf
state
or
parameters
file,
etc.
I
think
I
haven't
looked
into
this
specific
pr
yet,
but
one
thing
that
comes
to
my
mind
is
to
use
a
like.
You
know:
change
provider
terraform
to
use
something
like
secret
as
their
backhand,
so
that
because,
like
natively
knows
how
to
work
with
secret
because
it's
a
valid,
valid
state
store.
D
That's
actually
what
it's
using
today
and
the
issue
is
that
when
you
kill
terraform,
while
it's
using
that
the
the
kubernetes
back
end,
it
leaves
the
lease.
So
it
uses
a
lease
as
a
locking
mechanism
right
to
only
allow
one
terraform
process
to
to
access
the
tf
state
at
a
time
and
when
you
kill
terraform,
you
know
non-gracefully.
D
C
Yeah
gotcha
yeah,
well,
one
thing
that
we
do
in
internet
that
might
be
applicable
in
that
sense
is
that
we
said
because
we
we
save
state,
like
you
know,
in
different,
like
in
a
city
like
you
know,
in
the
custom
resource
and
and
for
sensitive
information
in
the
secret.
C
We
tell
it
to
not
hold
the
lease
essentially
so
that,
like
you
know,
even
if
you
kill
terraform
workspace
is
gone
and
there
is
no
file
system
at
all.
We
are
able
to
reconstruct
the
tf
state
and,
like
you
know,
keep.
B
C
So
maybe,
like
you
know,
yeah
we
can.
We
can,
like
you,
know,
experiment
with
that
option,
like
you
know,
seeing
how
like
essentially
automating
what
you're
doing
manually
right,
automating
your
intervention,
what
you
do
is
essentially
remove
the
lease.
So
if,
if
removing
the
lease
is
always
the
option,
it
is
always
the
viable
option.
Like
you
know,
maybe
we
can
even
just
tell
it
not
to
do
that.
D
D
C
Like
you
know
something
that
I
don't
really
know
regarding
provider
terraform,
like
you
know
how
it
manages
workspaces
because,
like
you
know
normally,
when
you
use
like
a
terraform
cli
like
if
it
has
certain
like
you
know,
restrictions
right
like
you
can't
lose
tf
state.
C
The
the
workspace
folder
is
important
and,
like
you
know
there
is
this
whole
backhand
story
around
it
right.
So
you
know
as
it's
like
a
bit
delicious,
we
use
only
like
you
know,
specific
flow
of
execution.
That's
why
we're
kind
of,
like
you
know
more
liberate
whatever,
like
you
know,
we
do
with
terraform,
but
in
provider
telephone
this
is
less
possible.
So
I
I
need
to
like
you
know,
look
into
the
looking
to
go
and
to
be
clear
like
no.
I
haven't
I
haven't
been
in
in
the
provided
server
form
code
base.
C
B
Yeah
yeah,
but
a
life
cycle
hooks
like
a
generic
mechanism
that
bob
is
incorporating
with
this
pr
into
into
cross
plane.
It
seems
reasonable,
like
to
add
this
ability
to
hook
into
a
pod
life
cycle.
It
will
be
a
control
control
report.
Even
if
it's
not
specific
to
provider
terraform,
it
seems
to
be
a
reasonable
functionality.
B
C
E
This
sound,
this
is
a
design
feature
of
terraform
such
that.
If
a
reply
is
interrupted,
the
state
is
not
reliable
and
it
is
actually
a
decision
by
terraform
to
force
users
to
go
and
manually,
remove
the
lock
after
ensuring
that
the
state
is
is
even
recoverable
that
they're
they
haven't
been
left
in
a
condition
where
their
infrastructure
will
get
stomped
on.
E
D
Yeah
I
mean
in
in
this
specific
scenario,
at
least
from
what
I've
read
right.
If
you
send
terraform
a
sig
hub
signal
one,
while
it's
running
it
is
supposed
to
cleanly
shut
itself
down
and
release
the
lock
and
and
leave
state,
you
know,
leave
the
state
sane.
So
that's
what
I'm
really
trying
to
do
is
I'm
really
trying
to
you
know
the
pods
going
away
like
it
or
not.
D
You
know
in
the
in
the
deployments
that
we're
planning
and
the
the
scale
that
we'll
be
working
on
there's
just
no
way
we're
going
to
be
able
to
handle
these
issues
you
know
manually.
Obviously
we
want
to
move
away
from
provider
terraform
as
much
as
we
can,
but
you
know,
there's
always
going
to
be
something
left
behind.
E
Well
and
there's
and
you're
always
going
to
be
in
a
case
where
terraform
might
not
be
able
to
clean
up
in
the
60
seconds
that
it
has
like
even
a
life
cycle
hook.
You
know
if
a
pod
is
being
is
receiving
a
termination
signal,
because
a
node
is
being
preempted
or
something
you
have
a
limited
amount
of
time
for
terraform
to
even
try
to
clean
itself
up
and
if
it's
cleanup
procedures
take
longer
than
that.
You're
still
gonna
wind
up
with
the
least
not
cleaned,
because
the
pod
will
get
killed
and
removed.
C
D
E
D
Yeah
I
mean
there
is
a
lot
there
are.
There
are
configuration
options
in
kubernetes
that
will
let
you
adjust
that
timer.
I
don't
know
what
I
mean.
Obviously,
there's
got
to
be
an
upper
limit.
Obviously
there's
scenarios
where
the
worker
goes
away
and
there
there
is
no
timer
right,
it's
just
gone
and
you
know
so,
and
you
know
maybe
this
is
just
an
artifact
of
the
fact
that
we're
running
a
coi
command
in
the
background
and
there's
always
going
to
be
some
level
of
instability.
A
Yeah,
you
know
it
would
be
also
pretty
maybe
useful
in,
like
the
general
sense
here,
to
get
to
get
dan's
opinion
on
like
adding
this
to
controller
config.
You
know,
I
know,
there's
been
some
thoughts
about
in
like
a
general
way
of
like
of
how
we
approach
controller
config
and,
I
think,
there's
an
issue
for
it.
A
We'll
have
to
look
and
find
that
issue
but
see
if
there's
something
that
makes
sense
to
kind
of
approach
this
in
a
holistic
way
of
general
controller
configs
design
concerns,
but,
like
that's
one
thing
to
consider,
but
in
general,
like
in
general,
specifically
like
this
doesn't
seem
like
a
large.
You
know
departure
from
some
of
the
other
things
like
the
environment
variables,
and
you
know
things
that
are
part
of
controller
config
already,
so
it
doesn't
seem
like
this
is
like
a
poor.
D
A
Cool
yeah
so
we'll
have
to
we'll
have
to
follow
up
on
that,
but
yeah,
that's
that's.
My
initial
gut
feel
at
least
on
it.
A
Cool
yeah,
thanks
for
bringing
this
up
bob
and
thanks
for
continuing
to
contribute
to
and
driving
driving
ideas
into
the
project
too
man.
It's
my
pleasure
and
was
this:
was
this
the
same
thing
here?
Bob.
A
A
All
right,
then,
if
that's
everything,
we
can
go
ahead
and
adjourn,
we'll
keep
moving
forward
on
the
1.9
release
cycle
and
we'll
be
talking
on
slack
and
such
in
the
meantime
and
see
everybody
here
in
two
weeks.
The
next
meeting.