►
From YouTube: Cloud Custodian Community Meeting 2021-08-31
Description
Our community meeting is public and we encourage users and contributors of Cloud Custodian to attend! You can find the notes for this meeting on our github repo: https://github.com/cloud-custodian/community/discussions
To get an invite to the meeting join the google group and you'll receive one via email: https://groups.google.com/g/cloud-custodian
A
B
B
Maybe
but
welcome
everybody
to
the
clack
custodian
community
meeting.
This
is
the
community
meeting
that
we
have
every
two
weeks
we
started
on
a
weekly
cadence
and
we're
kind
of
trying
to
figure
out
what
the
best
pace
is
here.
But
the
kind
of
mission
for
this
meeting
is
for
us
to
get
together
as
a
community
in
a
like,
slightly
organized
fashion,
talk
about
what's
happening
in
cloud
custodian
and,
of
course,
give
the
community
members
an
opportunity
to
race
things.
B
You
know
if
you
have
a
pr
or
an
issue,
and
you
prefer
a
more
high
bandwidth
discussion.
We
kind
of
have
this
hour
every
two
weeks
to
communicate
in
this
kind
of
high
bandwidth.
You
know
if
you
have
an
idea,
you
want
to
toss
out
to
the
community
that
sort
of
things,
and
it
helps
us
cover
things
like
announcements
and
cool
stuff-
that's
happening
in
events,
so
we
have
an
open
agenda
here.
If,
first
before
we
get
started,
we
do
introductions.
B
B
From
looking
at
the
names,
I
think
everyone's
been
here
before,
so
that's
good,
all
right.
Moving
on
to
the
agenda
items
first,
we
have
our
regular
weekly
doc
sprint
sessions.
This
is
when
myself
and
liz
kind
of
on
fridays
pick
an
hour
find
something
that
we
is
annoying
to
us
in
the
docs
or
somewhere
in
github,
and
then
attempt
to
do
a
fix.
B
For
this
past
week,
we've
been
looking
at
moving
the
moving
the
issue
templates
from
the
kind
of
old
style
github,
it's
just
a
bunch
of
markdown
and
then
the
template,
and
then
we
asked
users
to
fill
it.
Github
now,
of
course,
has
github
forms
which
gives
us
a
nicer
user
experience.
So
I've
opened
the
pull
request
for
that.
I've
had
some
people
look
at
that.
B
If
you
have
opinions
on
that,
please
feel
free
to
look
at
that
and
I'm
hoping
to
close
that
out
after
I
get
some
more
feedback
from
reviews
there
does
anybody
want
or
have
an
opinion
on,
github
forms
for
issues
and
whatnot?
If
you
look
at
the
pr
I'm
trying
to
not
make
it
so,
you
have
to
fill
out
an
entire
survey
monkey
to
file
an
issue,
but
it
will
be
useful
for
us
as
far
as
making
sure
that
we
ask
people
what
cloud
provider
they're
on.
For
example,
anyone
have
comments
here.
B
That's
what
I
like
to
hear
nice
and
nice
and
simple
no
brainer
there.
The
second
bit
is,
as
I've
been
doing,
documentation,
pull
requests
and
things
like
that
kind
of
by
default.
The
reviewer
is
kapil,
so
I'd
like
to
propose
for
us
to
have
an
at
cloud
custodian,
docs,
team
and
github,
so
that
we
can
add
more
than
one
person
there.
So
it's
not
blocking
on
anyone.
D
And
that
seems
like
a
good
call
to
me
happy
to
jump
in
there
if
you're.
Looking
for
members
for
that
group
and
also
on
the
since
you've
got
that
open
pr,
I
can
drop
a
link
to
my
custodian
fork
if
anybody
wants
to
see
how
those
issue
templates
look
live.
If
that's
useful.
B
Yeah,
that
would
be
great
drop
down
or
in
the
notes,
yeah.
So
people.
D
Put
in
chat-
and
we
can
put
it
over
there-
I
mean
that
will
be.
That
will
not
be
an
always
live
link
yeah
sure,
but
for
for
now
it
will
work.
B
All
right,
so
I
went
ahead
and
added
those
links
to
the
notes
kapil.
How
should
I
do
this
custodian
doc?
Should
I
follow?
Shall
I
file
an
issue
for
you
to
create
it
or.
B
B
And,
of
course,
if
anybody
else
wants
to
add
to
join
the
docs
review,
team
just
start
reviewing
doc
prs
and
then,
after
a
few
we'll
we'll
we'll
put
you
in
there,
we
don't
really
have
like
a
formal
policy
like
if
you
come
for
kubernetes.
Like
you
know,
you
have
to
do
x,
amount
of
pr's
in
order
to
be
doing
by
a
group.
I
think
we're
still
small
enough
or
we
just
kind
of
know
you
and
what
do
we
think
as
far
as
or.
C
In
the
group
be
part
of
the
group,
just
from
a
reviewer
perspective,
you
still
need
to
facilitate
the
container
flipped
on
the
repo
which
there's
like
about
a
dozen
people
out
in
the
world.
As
far
as
people
in
this
group
right
now,
aj
wayne
mandeep,
I
think
you're
an
emirate,
that's
something
you
want
to
try
to
do
another
pr.
Then
you
can
get
back
to
active,
fyi,
yeah
and
then
todd
actually
as
well
as
far
as
active
committers
that
I
see
in
this
meeting.
B
I
feel
you
all
right
anything
else
on
docs
and
docs
review
and
as
usual
on
fridays,
I'll
usually
announce
on
the
getter
about
an
hour
before
we
go
live,
hey
we're
kind
of
looking
at
this
issue
or
whatever
people
to
dive
in.
So,
if
you
want
to
join
me
and
liz,
it's
fine.
We
just
hop
on
kind
of
shoot.
The
breeze
look
at
the
docks
file
issues
on
things
we
find
and
try
to
think
about
ways
how
we
could
restructure
things
to
make
them
look
better.
So
any
other
comments.
D
A
I
can't
remember
now,
I'm
sorry
but
yeah.
If
someone
has
come
across
something
like
please
let
us
know
so
that
we
can
address
it
on
friday.
If
there's.
C
The
recent
topic
around
documentation,
which
I
just
don't
know
that
we
have
a
good
answer
yet
definitely
would
want
to
raise
it
for
the
broader
group.
The
stadium
has
a
bunch
of
server-side
filters,
good
luck,
trying
to
find
out
what
they
are
unless
you're
you
know,
sourceful
lunking
and
all
of
those
are
both
resource
and
source
specific.
C
So
if
you
have
any
thoughts
about
how
to
how
to
do
documentation
on
the
server-side
filtering,
I
think
we
should
at
least
get
an
example
of
using
it
and
exposing
the
fact
that
it
exists
into
some
form
of
narrative
doc,
but
as
far
as
getting
to,
I
think
the
question
would
then
be:
how
do
we
generate
reference
material
for
it
for
on
the
january,
docs
and
right
now?
That
is
still
an
opening
question
just
because
all
of
those
are
implemented.
C
B
Gotchu
and
when
we
get
to
those
two
issues,
it'll
be
more
clear:
how
we,
how
kapil
arrived
to
the
this
conclusion,
all
right
anything
else.
B
Okay,
awesome
moving
on
that's
really
kind
of
the
end
of
the
formal
agenda.
If
anyone
has
anything
to
add,
I
will
leave
you
a
space
there.
I'm
gonna
take
some
time,
however,
to
talk
about
code
as
governance
day
governance
day.
B
I
messed
this
up.
Every
single
government
promise
I'm
not
doing
that
on
purpose,
and
this
is
the
day
before
kubecon
slash
cloud
native
con
as,
as
we
said
before,
I'm
just
gonna
repeat
this.
Every
meeting
you
can
register
now
for
the
link
there.
The
event
is
free
to
attend
whether
you
attend
physically
or
virtually
and
just
a
reminder.
You
don't
have
to
have
a
kubecon
cloud
native
con
ticket
to
attend,
so
you
can
click
on
that
and
we
are
still
looking
for.
B
We
have
a
few
spots
left
open
for
talk,
so
if
you
want
to
submit
a
p
submit
a
cfp,
I've
got
the
link
there
in
the
notes
as
well
and
umir
has
been
generous
enough
to
extend
the
deadline
to
september
8th.
I
know
some
people
are
still
kind
of
mulling
in
the
in
the
prototyping
stage
and
want
some
extra
time
to
do
that.
B
So
the
deadline
for
talk
submissions
will
be
on
september,
8th
and
then
will
give
us
a
day
to
kind
of
organize
and
plan
the
schedule
and
then,
on
september,
10th
we'll
be
publishing
the
schedule
so
we're
hoping
to
have
a
nice
diverse,
set
of
content.
B
There
for
you
to
consume,
do
we
have
any
questions
about
that
at
all
and,
of
course,
if
the
kubecon
committee
decides
to
not
have
the
physical
event,
we
will
of
course
have
the
fallback
to
the
virtual
platform,
which
is
what
we've
been
using
for
our
workshops
anyway
and
has
been
generally
pretty
good
for
us.
So
it
won't
be
a
problem
for
us
to
switch
over
to
virtual
anyone
any.
E
Comments,
yeah
and
if
you
want
to
attend
virtually
or
present
virtually
that's
fine
too,
like
even
right
now
we
pretty
much
have
all
the
presenters
virtual.
So
again,
if
you're
interested
in
a
talk,
you
know
just
send
us
a
note
or
let
us
know
I'll
put
in
my
email
in
there
too.
Just
if
somebody
else
is
interested
got
you.
A
Does
it
specify
anywhere
I'm
just
looking
in
the
event
page
that
you
don't
have
to
that
talks
are
not
required
to
be
in
person?
E
Wrote
it
down
like
as
a
hybrid
event
just
by
default,
but
we
can
write
it
down
too.
A
Yeah,
I
don't
know
if
that
might
I
don't
know
it
might
help
clear
up,
like
any
confusion
around
that
in
case,
like
some
people
have
been
hesitant
to
submit
a
talk
because
they
don't
want
to
have
to
be
there
in
person.
C
I
did
I
I
found
the
reinforced
just
happened.
I
guess
last
week
yeah,
it
was
last
week
and
I
just
found
the
spot
in
the
keynote
that
the
hbo
max
guy
was
talking
about
how
how
they
use
cloud
custodian
to
help
them
achieve
remediation
at
scale.
B
B
Yeah,
fantastic
all
right,
moving
on
one
second,
I
lost
my
place
when
I
had
to
scroll.
Let's
go
over
the
backlog
here,
really
quick.
We
don't
expect
progress
in
these
all
all
the
time,
so
we'll
just
go
right
through
them.
Contributor
guide
liz.
I
know
that,
generally
speaking,
every
friday
we're
looking
at
this
but
yeah
anything.
A
A
Not
yet
for
this
it,
we
have
been
just
focusing
on
just
the
user
documentation,
but
it
might
be
interesting.
I
think
this
friday,
maybe
to
take
a
look
at
what
we've
got
already
existing
in
the
contributor
guide
and.
C
A
While
we
still
it
sounds
like
we
might
need
to
do
a
little
bit
of
thinking
about
how
we
want
to
surface
the
stuff
around
filtering
right
and
so
in
the
meantime,
taking
a
look
at
the
contributor
docs,
I
think,
would
be
a
good
thing
to
do
this
friday.
B
C
Yeah,
I
I
I
don't
know
that
we
have
a
great
answer
right
now,
yeah,
if
you
were
interested
in
turning
to
a
provider.
Generally
speaking,
we
we
have
tagged
all
the
provider,
specific
issues,
there's
a
bunch
on
azure
and
a
couple
on
gcp.
C
B
Got
you
and
probably
the
same
with
renaming
master
domain.
C
That
one-
I
am
more
interested
in
I've,
been
doing
some
more
reading
and
yeah
would
like
to
get
this
done.
Someone
hold
the
mickey
to
the
fire
either
you
know
publicly
or
privately
on
getter
and
we'll
try
to
get
down
before
the
next
community
meeting.
B
Yeah,
that's
why
I
left
it
in
the
backlog
notes
to
force
ourselves
to
do
it
any
any
questions
on
that
one.
This
seems
pretty
straightforward.
B
Make
getter
easier,
so
I've
been
kind
of
on
the
side
when
I
get
a
few
minutes
here,
investigating
every
single
matrix
client
that
I
that
I
can
there's
a
lot
of
movement
in
this
space.
So
I
don't
really
have
anything
to
report
other
than
I
am
keeping
track
of
all
the
web
kind
of
chatty.
B
Clients
that
we
could
be
using
on
top
element
seems
to
be
making
a
lot
of
progress
here,
but
really
nothing
to
report,
and
when
I
do,
when
the
feature
set
kind
of
starts
to
match
what
we
need
I'll,
probably
start
to
take
a
more
opinionated
approach
on
the
client
just
give.
B
We
are
due
for
a
release.
We
did
that
last
two
weeks
ago.
Did
the
conversations
around
pip
x
ever
go
anywhere?
Is
this
another
reminder
to
ourselves
in
the
future.
D
Some
folks
were
interested
in
trying
it
before
we
officially
recommended
it
anywhere.
I
mean,
I
think,
where
we
left
off
was.
I
thought
it
was
cool
and
everybody.
C
It's
a
great
question,
though
I
think
on
how
do
we
improve
install
experiences
like
we
have
like?
We
have
a
lot
of
users
who
know
nothing
about
python
and
we
are
either
giving
them
a
docker
experience
or
we're
forcing
them
to
at
least
figure
out
how
to
do
some
python.
C
I
think
there
are
some
two
interesting
tools
in
the
python
ecosystem
that
can
produce
self-contained
binaries,
I'm
not
probably
butchering.
That
name
seems
to
be
one
that
seems
to
work.
Okay
and
potentially
produces
can
produce
a
self-contained
binary
can
produce.
A
binary
depends
on
system
python.
C
I'm
just
wondering
like
pivx
is
clearly
the
easy
one
of
the
easier
ways
to
do
this,
but
still
requires
pip
and
solid
pivx
pivx
install
this.
Like
yes
see.
Did
you
just
tell
me
to
install
your
tool
by
installing
your
tool?
Yes,
yeah
and
yeah,
so
I
mean,
I
think,
there's
an
open
question
on.
B
B
C
We
do,
I
don't
know
that
it's
been
in
ages
decades.
Even
I
forgot
how
we
did
that
I'd
have
to
go
through.
Okay,
I
think
we
actually
have
camera
if
we
did
it
in
their
repo
or
our
repo,
but
there
was
an
open
issue.
I
don't
know
that
we've
updated
that
as
a
distribution
channel
to
the
extent,
though
it
exists.
C
C
Like
nitroco,
did
it
directly
into
core
of
homebrew
and
it's
currently
on
a
fairly
old
and
50
15
month
old
version,
and
I
don't
know
that
we've
updated
it.
D
Now
that
that's
all
fair,
though
I
mean
the
the
the
docker
option
too,
I
think
at
one
point
we
talked
about
that
that
docker
was
a
more
popular
way
of
getting
custody,
and
that
kind
of
surprised
me,
I
just
assumed
people
were
installing
it
more.
C
I
mean
looking
at
various
download
stats
on
pip
and
versus
docker.
Yeah
definitely
gets
is
ahead
by
a
little
bit,
but
okay
part
of
that
just
might
be
how
we
document
the
two.
I
you
know
like.
D
F
Why
not
both
the
statistics
might
be
a
little
misleading
given
usage
patterns?
If
people
are
more
likely
to
do
pip,
install
cloud
custodian
and
have
that
running
for
a
lot
longer
than
spinning
up
shutting
down.
F
C
Like
this
is
a
that's.
D
C
C
Like
this
is
just
the
classic
oss
issue
and
outside
of
doing
phone
home
type
of
opt-in,
opt-out
semantics,
like
reality
is,
is
you
download
the
bits
once
you
cache
them
locally?
Be
that
a
docker
hub
or
download,
and
that's
just
the
way,
the
world
otherwise
known?
As
how
do
you
count?
What
are
you
measuring
with
open
source
yeah?
I
think
the
kubernetes
t-shirt
and
thanks
for
the
comment.
C
Oh
well,
I've
got
an
adventure
t-shirt
on
so
nice
baseball
rag.
One
yeah.
I
don't
know
that
like
we,
we
track
download
numbers,
but,
like
we,
you
know,
we
clearly
know
they're,
not
authoritative,
in
a
real
fashion.
I
I
don't
know
that,
given
the
user
population,
that
we
want
to
go
to
anything
else
as
far
as
opt-in
or
opt-out
stats,
tracking
of
from
environments
like
that,
doesn't
I
feel
nice
yeah
yeah.
B
C
A
C
To
publish
the
test
pipeline
test
all
the
packages
and
then
publish
for
the
real
and
cut
the
release.
We
probably
need
a
bit
of
automation
to
create
a
pull
request
into
brew.
A
C
Get
rico
to
create
the
update
itself.
I
think
absolute
that,
like
I
said,
the
existing
version.
That's
in
there
looks
about
15
months
old,
so
clearly
lacking
in
on
the
tracking.
C
Sure
that's
your
code,
I
think
you
get
a
username.
Do
the
last
one
or
did
the
upstream
push
into
homebrew.
A
C
And
also
helped
us
set
up
our
reddit
forum,
which
made
I
don't
know
my
fans
actually
visited
it
recently,
but.
C
B
Yeah,
I
think
it's
worth
investigating.
Maybe
we
know
someone
who's
who's
into
this.
I
know
that
there
is
a
lot
of
activity
in
in
brew
with
open
source
packages.
So
surely
someone
out
there
would
might
be
interested
all
right,
anything
else,
installation,
wise
and
then
we
have
this
policy
testing
spec
with
marco
who
hasn't
been
able
to
attend
in
a
while.
So
I'm
going
to
table
that
unless
anyone
has
new
information
boom
all
right.
I
have
two
issues
to
review
today
from
this
week.
B
Let
me
get
those
open
aj.
You
were
running
lead
on
this.
D
Yes
and
now
you're,
okay,
you're
opening
them
yeah.
These
were
too
related,
because
I
think-
and
and
these
were
just
I
mean-
I
should
call
out-
I
mentioned
to
the
person
who
reported
them-
that
these
were
just
really
well
done.
Bug
reports,
so
yeah,
so
really
well
done
thorough
included
all
the
context
we
needed
to
look
into
it,
so
that
was
that
was
great.
That
was
really
nice
to
see
shout
out
to
robbie,
demuth
and
they're
related.
I
think
it
looks
like
he.
He
had
two
two
going
into
them.
D
At
the
same
time
it
looked
like
oh
yeah,
it
looked
like
the
first
time
he
was
looking
at
the
snapshots,
so
looking
at
described,
snapshots
and
found.
Oh,
this
is
this
is
the
second
one
so
yeah
we
can
talk
about
this
one
first.
B
D
B
D
Isn't
we're
not
pulling
we're,
not
paginating
this
response,
it's
just
trying
to
pull
back
all
the
hundreds
of
thousands
of
snapshots
in
one
call
and
that's
why
we're
hitting
a
timeout?
That's
why
it's
failing
and
looking
into
it
it
looked
like
you
can
under
the
hood
in
custodian.
We
do
try
to
get
a
photo
three
paginator
and
use
that
to
get
responses
and
pages
rather
than
pulling
everything
at
once,
but
in
the
case
of
ebs
snapshots,
it
won't
paginate
your
response.
D
Unless
you
tell
it
how
many
results
you
want
back,
sometimes
that
just
happens
implicitly
but
with
ebs
snapshots
it
didn't
so
it
looks
like
we're
going
to
have
to
add
that
and
there's
there's
a
lot
more
context
in
there
that
we
need
to
add
an
option
to
the
enum
spec
kapil.
If
you've
got
any.
D
I
would
say
the
the
wrinkle
that
he
identified
there
and
that's
what
I'm
what
I'm
calling
out
in
the
last
comment
there
is
that
if
we
add
that
option,
where
it's
a
max
results
option,
then
we
have
to
pull
that
out
when
we
go
to
query
individual
snapshot,
ids
otherwise
providing
explicit
ids
in
combination
with
max
results
will
blow
up.
E
G
D
A
little
bit
surgical
like
I
think
we
can
do
it
if
we
update
the
the
enum
spec
in
the
resource
type
and
then
kind
of
kind
of
back
out
and
get
resources.
But
I
don't
know
if
that's
the
best
way.
E
D
And
so
as
he
was
as
you
can,
you
can
read
this
issue
and
watch
him
falling
into
the
the
rabbit
hole
of
ebs
snapshots
and,
as
he
was
going
through,
he
thought
well.
We
should
be
able
to
pull
back
fewer
results
in
the
first
place
if
we're
taking
advantage
of
server-side
filters,
rather
than
pulling
back
hundreds
of
thousands
of
snapshots
and
then
running
client-side
tag
filters.
Can't
we
just
ask
the
api
only
give
me
snapshots
with
these
tags
set
so
that
we
don't
have
to
go
through
all
those
results
locally.
C
So
before
we
get
off
the
page
inner
bit
into
the
service
of
the
querying,
it's
worth
us
doing
a
investigation
as
far
as
other
ec2
resources
that
have
that
same
behavior.
I
suspect
it's
possible
that,
like
instances,
armies
there's
other
things
that
potentially
are
running
into
this
potentially
historical,
hysterical
reason
type
of
behavior,
where
they're
just
trying
to
keep
api
compatibility,
and
it
doesn't
scale.
D
D
D
So
open
four
factor:
seven,
four,
so
six,
eight
seven
four
was
okay,
there's
an
issue
with
pagination,
but
hopefully
we
won't
have
to
deal
with
this
anyway,
because
we
want
to
filter
on
the
server
side
and
not
pull
back
as
much
pull
back
as
many
resources
in
the
first
place,
and
so
he
did
what
looked
looked
like
a
logical,
a
logical
thing
to
do
you.
D
And
it's
not
it's
not
really
obvious
to
find
that
through
the
documentation.
I
think
if
it
is
documented,
it's
it's
just
a
quick
note
here
there
I,
if
you're,
going
to
come
across
how
to
use
server-side
filters,
it's
through
the
the
query
block
of
a
policy
which
has
some
overlap
with
the
aws
config
source,
and
you
pretty
much
have
to
read
the
code
today
to
understand
how
to
use
it.
D
B
D
Yeah
that
one
right
there
and
there's
a
link
to
the
spec
where,
where
custodian
defines
which
filters
are
acceptable,
are
valid
in
the
in
the
schema.
So
I
think
this
is
useful
and
you
know
robbie
commented
back
and
said:
oh
you
know
I
didn't
know
he
could
do
this,
which
is
totally
not
his
fault.
We
can
we
can.
We
can
surface
this,
so
I
I'm
hoping
that
this
issue
discussion
is
useful
context,
for
maybe
a
doc
sprint
call
I'm
happy
to
jump
in
with
some
of
that.
B
D
E
D
I
know
I
know
a
few
people
have
used
them.
I've
seen
it
come
acro,
but
only
if
you
happen
to
read
the
code
or
or
know
exactly
what,
if
you
had
an
existing
ec
api
call
or
cli
call,
and
you
wanted
to
replicate
it
in
custodians.
You
just
kind
of
search
the
code
for
where
that
shows
up,
and
you
say
oh
dang,
look.
I
can
do
this
in
a
query,
but
that's.
C
Yeah
the
only
time
people
run
into
it
is
when
they
they
bang
their
head
up
against
the
wall
and
they
go
start
splunking
and
or
they
come
into
slack.
That's
right,
get
her
and
ask,
and
someone
who
knows
thousand
men
like
it
is
very
not
discoverable.
G
Well-
and
I
ran
into
this
problem
where
I
started
using
it,
but
then
because
we
do
all
of
our
policies
like
inside
of
a
code
build
run
right,
so
we
have
tons
of
different
policies
running.
I
actually
stopped
using
it
so
that
we
could
just
pull
all
the
data
and
then
use
custodian's
caching
internally,
so
that
we
had
all
the
resources
and
not
a
subset
of
them
right,
and
so
it
didn't
actually
help
in
the
overall
run.
G
C
C
So
if
you
have
differentials
to,
if
you
are
doing
server-side
filtering
variants
across
your
policies,
you
may
find
that
it's
actually
faster
and
more
efficient
from
the
api
call
perspective
to
remove
them,
which
I
think
was
part
of
the
reason
justification
at
least
essentially
the
time
to
not
document
them
outside
of
people.
I
mean
know
about
special
cases.
D
Todd
yeah.
That's
that's
a
very
good
point.
That's
something
we'll
have
to
document.
If
we
do
expose
this
stuff
is
yeah
is
when
it
can
bite
you
and
some
of
the
I
mean,
even
in
this
case
it's
they're
looking
for
some
some
questions
about.
Is
this
tag
present
or
or
some
high
level
attribute,
but
then
for
some
of
the
more
advanced
stuff,
looking
doing
value,
checking
or
or
conversions
to
dates
and
checking
ages
on
things
that
the
best
we
can
do
is
a
presence
filter
through
server
side
and
then
do
the
more
advanced
stuff
locally.
D
D
That's
all
I've
got
for
this,
but
but
I
do
think
it
was
a
really
interesting
pair
of
issues.
A
Yeah,
I
was
just
taking
notes
on
that.
How
interesting
yeah,
aj
I'll
have
to
sit
with
you
or
you
know,
whoever
wants
to
hang
out.
A
Dive
into
this
a
little
bit
more
to
make
sure
I
understand,
before
than
a
dealing
or
addressing
it
in
the
docs,
so
we'll
plan
on
that.
B
D
B
B
All
right,
those
were
really
the
only
two
pr
slash
issues
I
wanted
to
bring
anybody
have
anything
else
burning
on
fire
or
anything
they'd
like
to
see.