►
From YouTube: Kubernetes Community Meeting 20161013
Description
We have PUBLIC and RECORDED weekly video meetings every Thursday at 10am US Pacific Time.
https://docs.google.com/document/d/1VQDIAB0OqiSjIHI8AWMvSdceWhnz56jNpZrLs6o7NJY
Demo on incubator project node-feature-discovery; SIG Cluster Lifecycle; Release 1.4.2 update; Release 1.5 alpha today and some new processes; 1.5 features scrub; Kubecon update
A
Okay,
good
morning,
all
that
is
the
Cooper
Nettie's
community
meeting
today
is
Thursday
octobre
13
and
we
have
a
demo
from
the
Intel
team
on
node
feature
discovery
by
biology
and
Connor
this
morning,
and
then
we
will
go
on
to
the
rest
of
our
meetings.
So
well
as
you
please
introduce
yourself
and
let
us
know
what
you're
going
to
be
talking
about
sure.
B
B
B
Can
go
to
our
Gator
page
and
look
at
the
project
if
you
want,
and
so
first
off
I
will
start
with
why?
What
did
we
do
in
this
project?
So
it
is
a
software
project
which
discovers
the
hardware
features
and
configuration
on
the
community
snowed
and
the,
and
it
exposes
these
harbour
feature
and
configuration
for
workload
scheduling.
B
So
why
do
we
do
this
all
nodes
in
a
community's
cluster
or
the
same?
The
eight
virginity
can
stem
from
the
hardware
features,
as
well
as
the
platform
configuration
we
want
to
enable
users
express
preference
for
certain
types
of
nodes
or
certain
types
of
configuration
in
their
job
solutions,
and
through
that
we
want
to
improve
the
efficiency
in
terms
of
performance
and
also
reduce
performance
variability
of
applications.
So
how
does
it
work?
It
will
basically
deploy
a
container
in
each
of
the
communities
node.
B
The
container
discovers
certain
set
of
hardware
features
and
configuration
and
adds
adds
labels
for
each
of
these
feature
as
mode
labels
for
each
of
these
features
and
configuration.
So
this
is
a
look
at
the
CLI.
You
have
certain
options
you
can
use
with
the
discovery.
Software,
for
example,
you
can
look
at
the
version
of
the
discovery
software.
The
sources
feature
sources
from
which
we
are.
B
So
in
these
node
labels,
so
we
encode
certain
pieces
of
information.
The
first
one
we
encode
is
a
namespace
to
denote
which
vendor
or
providers
providing
this
feature.
For
example,
it
could
be
not
Alfred
or
intel
com
and
then
the
version
of
the
discovery
code.
Currently
we
auto
version
the
discovery
code
using
the
commander
displayed
in
the
slide,
and
it
also
encodes
the
source
the
feature
source
of
the
feature.
For
example,
it
could
be
cpuid
and
finally
it
it
encodes
the
feature
itself.
If
so,
it
could
be.
For
example,
the
harbor
feature
a
ESN.
B
So
how
does
it
really
work?
Take,
for
example,
this
set
of
that
we
three
nodes
in
the
communities
cluster
on
the
host
one,
the
API
server,
the
scheduler
is
also
running,
so
what
you
would
do
is
you
would
create
a
discovery,
job
the
discovery
pod
lands
on
all
of
these
nodes.
It
figures
out
one
of
the
hardware
features
and
configuration
available
in
each
node.
It
contacts,
API
server
and
decorates.
Each
of
the
hosts,
with
the
available
node
features
and
platform
configuration
so
to
illustrate
this
I'll
go
to
an
ask
a
cast
here.
B
B
For
a
start,
we
can
look
at
what
are
the
labels
that
are
already
available
for
each
of
these
nodes,
so
there
are
three
set
of
labels
available
for
each
of
these
notes.
So
I
am
just
getting
the
I'm
just
displaying
the
labels
of
sync
with
each
of
the
nodes
here
and
I
am
also
showing,
if
any
part
is
running
in
the
cluster.
None
of
it
is
running.
We
can
deploy
the
discovery
pods
in
each
of
the
nodes,
using
a
simple
script
that
is
available
in
the
grid.
B
Oh
crap,
oh
so,
as
you
can
see,
contain
yourself
started
on
each
of
the
node
in
the
cluster,
so
node
1,
node,
2
and
node
3
us.
So
when
the
container
finishes
you
will
see,
you
will
see
that
it
decorates
the
nodes
with
the
hardware
features
and
a
platform
configuration
available
on
each
of
those
notes.
B
So
you
can
see
that
it
has
decorated
the
node
with
a
set
of
features
so
to
reiterate,
earth
or,
to
recap
the
what
we
encode
in
the
label.
We
have
the
names
case,
you're
the
version
of
the
discovery
code,
the
feature
source
and
finally,
the
feature
itself.
So
this
feature
is
available
in
this
node.
So
we
have
a
value
of
true
for
that
feature.
B
So,
coming
back
to
the
slides,
so
now
we
have,
we
have
deployed
the
discovery
pod
and
we
have
to
get
to
the
node
lib
nodes
with
the
labels
of
harbor
features
and
configuration.
But
what
is
the
benefit
of
using
or
Peter
discovery?
So
in
order
to
order
to
show
an
example,
we
chose
the
example
feature
of
turbo
boost.
So
what
we
do
here,
we
get
two
applications
which
benefit,
and
one
of
the
applications
will
benefit
from
rubber
boots.
Scallop
won't
benefit
from
rubber
boots
and
we
can
use
this.
B
The
hardware
disk
the
labels
used
by
the
hard
disk
are
sorry,
not
feature
discovery
to
look
at
the
hour
to
run
the
plum,
the
applications
in
the
inner
in
an
efficient
fashion.
So
the
first
example
is
affinity
to
the
feature
set
up
is
the
same
arm,
so
it's
three
hosts,
but
I
will
intentionally
disabled
a
turbo
boost
in
one
of
the
nodes.
B
So
what
this
means
is
the
three
hosts
are
identical
in
terms
of
utter
feature,
except
for
turbo
boost,
so
only
one
of
the
one
of
the
nodes
have
turbo
boost
enabled
for
the
first
example
here,
I
am
running
a
benchmark
from
parsec
benchmark
suit.
It
doesn't
image
similarity
search
really,
so
the
metrology
you
use
here
is:
we
run
the
benchmark
ten
times
without
with
and
without
affinity,
turbo
boost
and
wish.
B
B
The
next
example
I
want
to
show
you
is
anti
affinity
to
a
feature.
It's
the
exact
same
setup
and
the
exact
same
methodology
methodology
where
we
run
the
benchmark
10
times
with
without
anti
affinity.
The
benchmark
used
in
this
case
is
cloverleaf.
The
man
tivo
benchmark
suit.
It's
a
mini
app,
which
solves
hydro,
dynamic
equation.
B
B
So
again,
like
we
run
the
benchmark
10
times
with
and
without
anti
affinity,
and
it
should
be
performance
here
again.
So
it's
a
similar
plot,
where
the
y-axis
shows
the
percentage
variability
in
normal
execution
time
and
the
left
hand.
Side
of
the
plot
is
where
we
use
indefinitely
to
a
feature
and
the
I-10
side
is
without
anti
affinity
to
feature.
You
can
see
that
there
are
some.
B
So,
in
the
example
case,
we
only
saw
the
affinity
and
and
definitive
a
feature
that
is
turbo
boost,
but
we
also
expose
other
features
are
using
our
discovery.
Software.
You
can
look
at
the
set
of
features
that
we
expose
by
going
to
our
little
bleep,
oh
and
finally,
like
we
would
like
to
welcome
contributions.
You
can
create
an
issue
by
going
to
our
get
up
page
and
you
can
also
submit
a
PR.
B
You
can
add
a
new
source
internally.
These
sources
are
implemented,
so
these
sources,
the
feature
sources,
implement
an
interface.
The
additional
sources
can
be
added
by
implementing
the
same
interface,
and
you
can
also
contact
us.
So
my
github
and
Ellis
pelagius
monium
and
olive
oils
data
panda,
lezcano
Doyle
and
you
can
get
in
touch
with
us
where
the
slack
or
the
the
video
sort
of
meeting
is
available.
A
A
A
C
The
question
was
why
labels,
instead
of
annotations
and
I,
think
the
answer
was
well.
You
could
have
you
know
if
you
need
annotations
use
that
and
I
think
that
was
sort
of
not
really
the
question.
The
question
is
that
I've
seen
this
question
come
up
a
few
times
in
different
design
discussions
which
is
or
should
we
use
labels,
or
should
we
use
annotations
and
I
think
it
would
be
helpful
for
everyone?
If
you
have
some
comment
on
it
to
understand
why
you
picked
one
versus
the
other.
E
Are
there
concerns
with
having
that
many
labels,
because
my
understanding
is
the
you
know?
The
big
difference
here
is
that
labels
are
indexed
and
you
can
write
queries
against
them
versus
annotations,
which
are
more
freeform
and
so
I
mean
as
we're
looking
at
scaling
up
in
the
amount
of
load
on
that
CD
I.
Just
worry,
you
know
how
many
labels
are
we
adding
on
a
node
by
default.
D
Yes,
I
would
recommend
not
adding
a
gazillion
labels
by
default
and
in
general
my
recommendation
would
be
only
add
the
ones
you
actually
think
you
need.
So
this
is
a
component.
You
know
in
its
own
repo
and
I
haven't
looked
at
it
that
carefully
to
be
honest,
but
this
did
come
up
in
a
number
of
discussions
and,
yes,
my
recommendation
would
be
to
create
a
white
listed
filter
mechanism,
so
you
actually
get
the
labels.
You
want
babe.
Any
additional
labels
that
get
added
to
the
nodes
are
effectively
a
liability
for
the
operator.
D
D
B
E
One
idea
that
may
make
sense
here
is
to
put
all
the
information
that
you're
discovering
into
an
annotation,
because
that's
useful
for
debugging
and
logging,
and
you
know
just
knowing
what
the
heck's
going
on
and
then
and
then
you
know
having.
Oh
so
Tim
is
saying
no
I've
been
having
a
way
of
like
select
out
of
like
one
big
annotation
sort
of
what
parts
you
want
to
hoist
into
labels.
Sure.
E
F
D
Know
I
mean
it's
more
of
a
topic
for
Notre
or
I.
Don't
know
actually
one
of
the
other
things,
but
there's
a
lot
of
data
that
you
may
need
for
debugging
codes,
we're
not
going
to
show
up
all
that
stuff
in
the
API
anyway
I.
My
recommendation
would
be
the
this
agent
just
actually
log
all
the
information
and
there
are
gazillion
logs.
You
need
to
collect
from
nodes
in
order
to
tell
what
was
going
on.
So
you
know,
until
we
sort
of
have
a
stronger
case
for
having
more
information
the
API.
B
So
you
answer
your
first
question
about
like
looking
at
what
labels
are
being
published,
you
can
run
the
discursive
tour
in
a
dry
run
mode
and
you
can
just
use
something
like
a
coupe
cuttle
logs
and
look
at
what
labels
are
actually
being
discovered
and
later
on.
We
had
the
white
listing,
it
should
be,
you
can
choose,
we
can
choose
what
labels
are
supposed
to
be
advertised.
A
B
H
An
incubator
project
and
we
need
evidence
usership
to
graduate.
So
if
you
try
it
out,
please
look
back,
especially
if
you
start
using
it
in
your
clusters.
We've
have
a
wiki
page
where
you
can
just
mention
like
hey
I'm
using
this,
and
we
can
use
it
when
we
go
to
make
our
case
for
becoming
a
real
project.
A
Everybody
trying
to
get
all
grown
up-
it's
all
awesome,
which
is
it's
an
excellent
point,
because
that
is
part
of
the
idea
of
incubator
is
to
make
sure
that
we
are
building
the
right
things
and
the
people
start
engaging
with
the
community
and
using
it
offering
user
feedback,
and
then
it
can
graduate
to
become
a
cougar
nidhi's
project.
So
thank
you
for
that
reminder
all
right
up.
Next,
we
have
a
update
from
Sega
cluster
lifecycle.
Ilya.
A
I
So
yeah
just
a
quick
one.
Well
yeah
we
had
been
pretty
busy
etsy
Gloucester
lifecycle.
We
have
released
cube
ADM
along
with
one
point:
1.40
release
of
Copernicus,
it's
technically
in
the
1.5
British
window,
but
the
binaries
are
released
alongside
and
it's
targeting
1.4
and
yeah.
Thanks
to
all
the
members
for
their
hard
work
on
the
we
managed
to
make
this
on
time
for
14
days
on
the
many
many
users
tested
it
and
reported
bugs,
and
we
really
would
like
to
thank
all
the
users
who
reported
this
early
alpha
bugs
this.
I
This
is
really
important
and
we've
been
busy
fixing
those
bugs
and
we're
hoping
to
cut
another
release
of
snapshot,
release
of
cube
ADM
coming
days
and
yeah.
We
really
appreciate
some
help
if
people
want
to
join
us
on
the
Seacoast
lifecycle
and
slack
and
to
stay
up
to
date
or
I
like
to
contribute.
Please
talk
to
us.
Thanks
is.
I
I
A
Yeah
well
as
you
as
you
may
know,
the
your
ego,
which
will
hear
a
little
bit
more
about
closed
for
the
1.5
feature,
planning
or
at
least
the
future
proposal
space.
So
we'll
talk
more
about
that
in
the
releases
section.
But-
and
this
may
be
separate
because
sig
cluster
or
couplet
is,
is
it's
a
standalone
piece?
So
thank
you.
Ilya
and
everybody
can
find
sit,
closer
life
cycle
on
the
community
meeting
calendar
and
slack
and
email
lists
and
all
the
usual
ways
going
on
to
releases.
We
have
a
super
short
update
from
1.4.
A
Jessie
was
not
able
to
join
us
this
morning,
but
she
wanted
everybody
to
note
that
we
cut
the
one
point
4.2
beta
and
would
like
testers.
So
if
anybody
is
playing
with
things
today,
please
try
and
play
with
the
one
point:
4.2
beta
they're,
looking
to
she's
looking
to
cut
the
one
point.
Four
point
two
point
release
tomorrow
friday.
A
So
any
feedback
on
the
beta
would
be
very
helpful
today
with
that
we
can
go
on
to
1.5
work
and
sod
and
caleb,
which
I
get
I
get
to
start
by.
Introducing
the
fact
that
sod
is
are
really
socked
for
1.5
and
caleb
during
the
retrospective
meeting
for
1.4
agreed
to
be
silenced
right-hand
person
to
help
us
find
all
the
places
where
it
the
release
is.
Our
work
still
gets
mired
in
being
being
or
more
so
more
closely
associated
with
google.
A
Then
we
want
it
to
be
so
so
we
now
have
a
a
two-headed,
I
can
say
monster,
but
forgive
me
may.
The
two-headed
releases
are
that
is
that
is
helping
so
sod
being
believed
and
caleb
stepping
up
to
do
as
much
as
he
can
from
the
outside
and
help
us
find
all
those
places
where
help
from
community
members
is
harder.
So
sudden
you
want
to
talk
about,
you
had
a
proposal
yeah.
J
So
the
proposal,
one
of
the
goals
that
we
want
to
add,
is
code.
Freeze,
let's
not
have
planned
code
changes,
that's
obvious
for
features,
but
it's
not
been
so
obvious
for
buzz
and
in
the
past
weeks
the
folks
try
to
sneak
in
codes
as
planned
bugs.
J
J
This
includes
flakes
and
non
flakes,
and
within
three
days
we
would
love
to
get
acknowledgement
that
this
bug
has
been
looked
at
and
has
been
triaged
and
whether
it
is
or
is
not
a
release
blocker,
and
if
we
can
get
that
I
think
that
will
make
our
lives
a
lot
easier.
I
know,
cabled
has
agreed
to
try
to
keep
track
of
all
the
p0
p1
bug
flakes
and
what
are
their
blocking
or
not
trying
to
get
the
separator
signal
from
the
noise,
because.
K
So
you
use
the
phrase
feature
freeze
there.
I
was
curious
if
you
could
define
the
difference
between
feature,
freeze
and
code,
freeze,
I,
guess,
I
was
sort
of
under
the
impression
that
there
was
like
a
deadline
for
everybody
to
get
there
they're
like
features
into
the
feature
repo
on
those,
maybe
monday
la
neige.
Okay,
so
is
that
the
feature
phrase
you're
talking
about?
Are
you
talking
about
code.
J
M
That
work
explain
a
bad
situation
about
this
fish
recent
code
freeze.
So
unfortunately,
we
had
this
bit
async
reading
these
two
freezers,
so
we
are
going
to
sing
Chris
Caliban
and
said
regarding
taste,
is
to
two
stages,
also
provided
some
situations
to
some
PRS
to
our
competitors.
Community
and
the
communities.
Fishery
police
wondered
five
road
map,
so
we
are
going
to
describe
these
features,
please
specifically
to
phrase
also
in
our
official
documentation,
the
surfer
tis
confusion,
but
we
like
we're
going
to
improve
all
this
stuff,
not
to
be
unsynchronized
in
all
these
activities.
K
That's
totally
cool,
so
the
reason
I
was
asking
about
the
feature.
Crees
thing:
I
apologize
that
I
wasn't
able
to
attend
the
retrospective
last
time,
but
if
we
think
we
now
have
sort
of
gated
what's
in
the
features,
repo
now
would
be
a
great
time
to
start
drafting
the
release,
notes
for
Cuba
Nettie's
15
from
a
product
and
feature
owner
perspective,
while
the
code
is
still
ongoing
that
whole
high
level
summary
that
I
did
like
during
the
burn
down
as
we
were
trying
to
cut.
This
is
something
that
could
absolutely
happen
now.
M
J
Yeah,
that's
a
great
idea
to
get
started
early,
especially
considering
the
features
are
now
frozen.
So
right.
All
the
other
thing
I'd
like
to
add,
is
we're
planning
to
cut
an
alpha
release
a
1.5.
Today
there
is
a
very
early
alpha
release.
I
went
out
in
September.
First,
we
tried
to
do
releases
every
two
weeks,
but
they've
been
blocked
on
at
that
build.
So
hopefully
we
can
get
that
out
today,
a
second
alpha
release,
and
then
that
is
all
for
for
the
release.
4.1
awesome.
A
But
in
that
retrospective,
one
of
the
things
that
was
suggested
was
anyone
who
has
an
open
book
or
an
open
featured
during
the
feature
ordering
code
freeze
is
asked
to
attend
the
burden
down
meetings
to
make
sure
the
status
is
given.
So
it's
kind
of
a
compliment
to
your
one
day
turnaround.
So
if
you
have
something
that
is
still
open,
you're
asked
to
participate
in
the
burndown
meeting,
just
to
make
sure
that
we
know
exactly
where
the
status
is
as
we're
going
into
release.
A
A
Sweet
just
one
in
check
fantastic
a
or
do
you
want
to
talk
about
the
feature
fries
from
Monday?
What's
going
on,
there
yeah.
M
So
today,
if
you
like
half
an
hour
before
we'd,
have
started
this
meeting,
we
have
finished
our
previous
mission
that
was
initiated
by
a
product
management
group
regarding
the
features
that
are
going
to
be
included
into
one
at
pfeiffer
lace.
So
we
host
our
strumento
the
features
in
github,
repo,
deter
marketplace,
milestone,
105
and
the
financial
status,
and
so
on.
M
So
in
the
current
moment,
to
like
we
have
this
something
like
drug
information,
because
we
have
to
provide
also
some
more
details
seem
too
into
the
github
issues
to
get
industry
from
so
with
the
conformity
we
have
55
features
it's
going
to
be
headed
in
in
the
year
relation
to
what
at
pfeiffer
list.
That
doesn't
mean
that
all
the
features
absolutely
brand
here.
Some
of
these
features
were
listed.
The
previous
cabinet
is
versions
as
alpha
a
beta
stage,
so
in
jerz
released
we
expecting
to
see
them
in
the
most
stable
state.
M
So
misread
treaty
is
like
the
column,
with
the
release
notes
mentioned
so
I'm,
asking
every
future
owner
to
provide
a
few
few
lines
about
that.
What
what
did
switch
is
about
what
you
can
see
in
the
release
notes
about
your
future
and
possibly
I
hope
that
we,
we
may
start
collaborate
with
aaron
right
now
and
with
that
and
creating
the
early
snows
generating
there
from
get
started
with
having
amount
in.
A
D
A
A
Open
shift
is
having
a
monday
workshop
on
monday,
oddly
enough
and
then
that
coupon
is
tuesday
and
wednesday.
There
is
the
Q&A
DS
developer,
some
on
Thursday,
which
invites
went
out
for
last
night.
We
have
a
few
people
on
the
waitlist,
but
for
the
most
part,
everyone
who
is
interested
is
everyone
who
has
expressed
interest
to
date
and
or
people
who
were
shall
we
say
volunteered,
because
people,
because
other
people
wanted
to
make
sure
they
had
conversations
with
them,
should
be
able
to
get
invites
over
the
next
couple
of
weeks.
A
We're
getting
a
couple
of
no
thank-yous
today,
and
so
the
waitlist
I
suspect
will
clear
with
those
no
rsvps.
So
my
suspicion
is
everyone
who
has
already
expressed
interest
in
going
or
people
who
were
requested
to
attend
by
other
groups
have
had
invites
and
are
likely
to
be
able
to
attend
travel
costs
aside.
A
We
saw
those
invites
have
gone
on.
The
schedule
will
go
up
in
the
next
week
or
so
of
the
formalized
topics
that
are
going
to
be
covered
at
the
dev
summit.
Those
are
topics
for
discussion,
they're
not
going
to
be
presentations.
So
the
idea
is
here's
a
topic
and
then
we'll
have
some
sort
of
conversation
about
it
in
a
room.
There's
also
going
to
be
a
fair
bit
of
open
space
available
in
this
in
the
dev
summit
area,
so
that
people
can
peel
off
and
have
smaller
discussions.
A
Smaller
groups
decide
things
on
the
spot
that
they
need
to
continue
a
conversation
or
take
a
conversation
in
a
different
direction.
So
we
will
have
more
about
the
logistics
of
that
for
Thursday
in
the
next
week
to
two
weeks
published.
But
those
of
you
who
have
invites
should
please
RSVP
so
that
the
few
of
you
that
are
on
wait
lists
can
get
rolled
in
and
if
there's
anyone
that
someone
thinks
that
we
have
missed
it.
A
M
Yes,
I
have
faced
the
question,
the
problem
and
not
not
me
personally,
but
also
some
people
from
community,
and
we
have
discussed
it.
For
example,
is
Michele
Slaton
see
gaps
that
there
is
no
structured
information
in
the
general
availability
are
getting
the
schedule
of
sick
millions
of
sick
face-to-face
meetings
during
club
card,
because
some
of
the
six
have
scheduled
your
meetings
privately
in
their
melon,
based
at
the
same
time,
it's
not
so
easy
for
people
from
the
outside
scheduled
a
time
schedule.
Their
logistics
journal
keep
cone
and
to
define
the
best
place
to
visit.
M
So
I
prepared
this
brief
spreadsheet,
where
I
am
asking
all
the
sig
leads
or
like
responsible,
sick
people
to
fill
it
with
the
actual
data
regarding
date
and
time
where
you
keep
confess
to
face
meetings,
collecting
could
happen.
Also,
if
you
don't
know
exact
time,
you
might
take
a
look
at
different
6l
just
to
find
a
free
slot
when
you'll
be
able
to
run
it
segmented.
M
A
Team
at
feelin,
X
Foundation,
has
also
agreed
to
help
us
find
space
for
these
over
the
course
of
the
week.
So
I
don't
know
that
they
knew
quite
what
they
were
getting
into
when
they
said
that
they
would
help
us,
but
we
will
do
our
best
with
them
to
find
space
and,
as
I
said
on
Thursday,
there
may
be
a
little
bit
of
space
as
well.
A
So
if
you
run
into
if
you
as
a
cig,
lead
or
as
a
cig
run
into
a
challenge
finding
space
to
have
this
meeting,
please
feel
free
to
reach
out
to
me
and
Cameron,
and
he
and
I
can
help
facilitate
either
with
the
Linux
Foundation
events,
people
or
directly
with
the
hotel
around
all
of
this
so
know
that
you
have
resources,
don't
don't
hesitate
to
ask?
Does
anybody
have
questions
about
coop
con,
the
community
developer
summit
or
the
cig
face
two
faces
that
are
happening
around
that.
K
I
think
there
was
a
lottery
around
the
kook
on
topics
to
Key.
Has
that
actually
closed?
Do
we
know
what
the
final
topic
list
is
to
be?
Are
we
still
sorting
that
out
the.
A
A
You
haven't
missed
that
notice.
There
are
going
to
be
12,
formal
topics,
then
there
are
going
to
also
be
six
small
room.
A
A
Can't
believe
we've
been
that
articulate,
but
thank
you
all
right.
Let's
see
next
next
thing
on
the
agenda.
I
have
nothing
further
on
the
agenda.
Does
anyone
have
any
notices
or
upcoming
things?
This
I've
been
working
all
all
things
toward
coupe
con.
A
All
right,
so
1.5
features
from
cigs
I
will
be
reaching
out
to
sig
leaves
more
for
next
week,
because
we
need
to
get
a
little
more
visibility,
a
little
more
water,
all
the
different
sinks
doing
and
so
I'll
get
more
sick
is
on
next
week.
Thank
you.
All
have
a
great
yeah
enjoy
the
extra
time,
and
it's
exactly
where
intensive
I
give
you
the
gift
of
time
have
a
great
week
and
I'll
see
you
all
next
week,
thanks.