►
From YouTube: Velero Community Meeting - May 2, 2023
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
Yeah
all
right,
hello,
everyone
today
is
May
the
second
and
that's
the
official
community
meeting
for
Valero
I'm,
the
community
Elite.
My
name
is
Orlin
and
welcome
everyone
to
our
code
today,
with
that
I'm
gonna
share
the
link
to
to
our
community
meetings
into
the
chat.
So
please
like,
if
you,
if
you
like,
add
yourself
in
there
add
your
topics
and
I'm
gonna
share
that
thing.
So
we
can
all
take
a
look.
What
we
have
for
today,
I
hope.
Everyone
can
see
that
and
yeah
all
right.
A
I
have
some
I've
added
a
few
stuff
like
the
super,
late
blog
PR
for
for
the
111
release,
but
yeah
that
that
release
got
me
in
the
middle
of
the
storm
of
kubecon,
so
yeah
sorry,
I
I
just
managed
to
to
put
some
something
in.
So
if
you
take,
if
you
can
take
a
look,
please
it
would
be
nice,
so
we
can
have
it
out
yeah!
That's
in
the
start
of
this,
for
myself
Scott.
They
want.
B
To
take
it,
yeah
yeah,
another
Bunch
for
me,
I
was
again
I,
started
and
and
testing
for
the
new
V2
plug-in
apis
just
before
the
release,
but
with
coupon
and
then
I'm
gonna
pull
in
this
other
ODP
stuff
last
week.
I
haven't
really
gotten
fully
into
this
yet
but
I'm
hoping
to
get
that
done
this
week
and
other
than
that.
You
know
PR
reviews
and
all
that.
But
that's
that's
pretty
much
it
for
me.
C
Hey,
can
you
have
it
yeah,
so
just
basically
Daniel
had
pinged
in
the
group
as
well,
for
this
particular
issue
that
I
have
posted
tracking
that
we
get
this
for
V
1.12.
C
So
basically
you
wanted
opinions
from
Scott
and
sugar.
My
belief
so
just
calling
it
out
here
if
you
folks,
can
have
a
look
if
you're.
Okay,
with
this.
B
Yeah
you
might
as
well
I
mean
if,
for
both
of
us,
when
a
women's
look
at
it,
hopefully
both
of
us,
but
someone
certainly
on
our
side,
we'll
look
at
it
this
week.
A
Thank
you,
yeah
I,
forget
I
started
the
new
discussion
into
the
GitHub
about
the
new
feature,
requests
and
stuff.
So
let's,
let's
monitor
that
and
pradeep
I've
added
you
there.
So
we
can
monitor
in
case
some
someone
comes
up
with
some
right
ideas.
What
we
need
to
add
or
if
they
want
to
work
on
something
with
that
kubecon
Shanghai
and
kubecon
North
America
were
announced
in
in
the
last
coupon
in
Amsterdam,
and
the
thing
about
Shanghai
is
the
folks
from
Claude,
Casa
and
redhead.
A
A
What
do
you
think
it's
in
September,
25th
to
27th
and
it's
gonna
collect
pretty
much
everything
that
can
happen
in
one
go,
so
it's
going
to
be
kubecon
and
it's
going
to
be
open
source
Summit
and
you
name
it.
So
if,
if
folks
are
planning
to
travel
drop
in
line,
what's
do?
Do
you
know
from
your
site,
I'm.
A
Yeah,
do
you
do
you
need
visas,
I
think
everyone
needs
visas
there
I'm
not
sure
about
I
I
definitely
need
visa,
so
I
have
to
apply
for
that
when
I'm.
E
A
B
Yeah
yeah
I
was
I
was
talking
the
way
with
actually
about
this
I
think
it
would
be
great
for
it
to
have
like
one
proposal
mainly
about
the
new
data
mover
between
you
know,
say
myself
and
Daniel
or
whoever
is
available,
and
so
we'll
have
two
maintainers
there.
You
know
from
different
different
companies.
That
would
be
great
kind
of
focusing
on
the
data
mover
aspect
and
then
I
think
Wes
has
some
other
ideas
for
something
else
that
you
know
on
our
side
week
might
be
able
to
propose
it.
B
A
B
B
And
when's
your:
when
is
your:
when
are
you
gonna
be
out
just
in
terms
of
coordinating
with
you
and
this
stuff?
You
said.
A
B
Yeah
and
I
don't
think
on
my
side
I'll
be
available
for
that
one.
But
hopefully
someone
else
would
you
know
and
maybe
on
your
side
or
whatever.
A
Yeah
I'll
try
to
actually
me
and
pradeep
we're
gonna,
try
that
to
see
how
it
goes
within
VMware
with
the
budget
and
everything,
but
most
probably
I
spoke
with
Jolene.
Do
we
have
Jolene
on
the
call,
not.
A
Oh,
it's
yeah.
They
have
the
whole
week.
Labor
yeah
celebration
thing,
but
yeah
I
spoke
with
him
and
he
was
like
super
enthusiastic
about
getting
all
the
maintainers
from
Valero
to
to
join
the
Shanghai
event.
So
yeah
we'll
see,
oh.
B
That's
great
because
yeah
that
was
the
one
thing
I
know
when
we
were
talking.
It
was
less
clear
on
that,
but
I
mean
I
mean
if
they're
going
to
be
going
there.
That
certainly
makes
it
a
lot
more
relevant
to
get
as
many
of
us
as
possible
there
to
actually
have
a
face-to-face.
You
know,
for
you
know
not
if
not
all
the
maintainers,
at
least
most
of
them
and.
A
A
The
thing
is:
I'm
not
sure
how
how
much
from
the
VMware
site
we
can
put
our
bet
on
we're
going
or
not
right
now,
yeah.
D
If
it
wasn't
just
a
curveball,
if
we
didn't
fly
out
for
kubecon
to
Shanghai,
would
flying
directly
to
Beijing
make
it
easier.
B
B
I
mean
I
still
think
Ukraine
would
be
even
better
because
that's
that's
in
the
context
of
lawyer,
but
this
might
be
a
backup
plan
like
if
coupon
doesn't
work.
You
know
if
we
can't
get
it.
You
know,
I,
don't
know
just
because
again,
if
we
had
it
in
Beijing,
then
obviously
all
of
the
other
VMware
maintainers
would
be
available
because
they're
already
there.
F
F
B
D
And
regardless
of
the
kubecon
there,
Cloud
costs
is
very
interested
in
meeting
up
and
having
a
little
hack,
Fest
the
day
before
kubecon,
which
requires
new
tickets.
So.
B
So
we
might
be
able
to
even
if
only
say,
some
of
the
maintainers
go
to
kubecon,
but
maybe
the
entire
team
could
meet
up
the
day
before,
because
that
would
just
be
travel
and
and
not
the
actual
coupon.
You
know.
B
I
An
option
emerges,
I
mean
we.
We
stay
committed
to
getting
a
spot
the
day
before
the
cubecon
for
for
all
of
us
to
meet
and
and
and
work
out,
some
Logistics
on
that
front
from
our
our
site.
A
Satya,
you
mean
like
an
official
thing.
Just
in
the
same
week,
of
coupon
I
mean
yeah.
I
What
we
were
thinking
is
I
mean
even
if
we
are
all
getting
together,
we
still
need
a
hotel
conference
room
for
all
of
us
to
be
together.
We
still
need
you
know
lunch
and
dinner
plans,
and
all
of
that
figured
out
so
we'll
we'll
provide
the
logistical
assistance
to
make
that
happen
and
and
and
get
a
venue
food
and
all
that
figured
out
from
our
side.
I
mean
that'd,
be
our
contribution
to
to
making
this
happen.
A
All
right,
okay,
that
too
make
at
least
looks
like
on
everyone's
side
a
little
bit
more
easier
to
to
justify
the
trouble.
If
we
have
some
more
formal
stuff,
all
right,
okay,
so
everyone
is
driving
the
in
their
companies
and
we
can
talk
again
in
like
two
weeks
or
something
like
that.
A
All
right.
The
next
thing
I
met
with
some
folks
from
the
soda
foundation
on
kubecon
and
for
those
of
you
who
doesn't
know
what
soda
is
I
didn't
know
as
well,
but
it's
like
a
cncf
kind
of
foundation
under
the
Linux
Foundation.
It's
about
data
and
mobility
and
stuff
like
this,
so
they
are
interested
in
Valero
becoming
part
of
the
ecosystem,
which
is
not
like
a
member,
but
it's
like
listed
in
into
the
foundation.
A
So
can
you
folks
take
a
look
at
the
solar
foundation
and
if
that
makes
any
any
sense
for
us
to
to
be
listed
as
a
member
of
the
ecosystem
for
the
sort
of
foundation
we're
listed
as
a
member
of
the
ecosystem
for
the
cncf
already
so
not
sure
I'm,
just
bringing
that
up,
because
there
were
like
super
Keen
into
speaking
with
us
and
getting
us
in.
A
A
But
still
it's
yeah.
F
Firstly,
evaluate
yeah
what
everybody
thinks
they
they
have
kind
of
for
listing.
We
have
to
buy
some
kind
of,
they
have
membership
fees
and
all
we
have
to
kind
of
that's
in
case.
A
Was
picking
us
like
listing
us
as
a
project
that
is
friendly
to
the
ecosystem,
which
is
not
not
no
string
attached,
but
it's
just
yeah
a
topic
for
the
community
to
take
a
look
at
this
one
if
it
makes
sense
or
not,
and
then
on
the
other
side,
the
like
the
big
body
of
VMware,
we
have
to
decide
if
we
want
to
proceed
forward
of
membership
or
not,
but
that's
another
topic,
that's
not
in
discussion
right
now.
A
A
J
Want
to
hi
yeah
sure.
Actually,
if
you
want
to
click
on
the
link,
maybe
yeah
go
ahead.
So
yesterday,
I
was
trying
to
answer
a
question
from
you
know:
one
of
the
users
with
the
long
hand,
snapshots
are
not
being
deleted,
so
I
was
doing
some
testing
in
and
found
that
you
know
what
could
be
a
possible
regression
that
volume
snapshot
resources
according
to
the
documentation,
even
as
I.
J
Remember
it
in
the
starting
from
1.9
welleros
cleans
up
them
deletes
the
volume
snapshots
once
the
backup
is
done,
but
it
doesn't
seem
to
be
happening
anymore,
so
I
tested
that
with
1.11.
So
it's
possible
that
this
is
a
regression
just.
B
B
I
mean
some
the
volume
snapshots
that
are
part
of
the
backup
that
we
created
as
part
of
the
backup
those
are
for
the
CSI
plugin,
especially
in
the
non-datum
river
case.
We
need
those
that
restore
time,
because
that's
that's
the
volume
snapshot
content
is
how
we
do
the
restore
so
I
think
we
do
have
some
code
volume
snapshots
when
the
backup's
deleted.
J
That
is
true,
but
I.
Think
as
per
this
comment,
as
you
can
see,
the
volume
snapshot
resources
are
cleaned
up
from
deleted
from
the
namespace,
but
the
content
is
supposed
to
be
around
and
the
storage
snapshot
is
supposed
to
be
around
and
the
reasoning
behind
this
is
by
mistake.
If
you
delete
your
Source
namespace,
you
don't
want
to
lose
everything
right.
J
B
J
E
Yeah,
the
behavior
should
be
like
the
volume
snapshot
is
deleted
when
the
back
just
before
the
backup
is
completed
and
the
content
is
the
content
remains
in
the
cluster
and
the
policy
is
updated
to
retain
I.
Think.
J
J
Snapchat
then
you
would
lose
everything
so
I
noticed
that
Valero
doesn't
recommend,
or
probably
they
recommend
I
think
the
document
does
mention
retain
some
point,
but
maybe
it
has
to
be
more
proactive
and
it
has
to
make
sure
that
it's
retained
right,
because
delete
has
kind
of
dangerous
semantics
anyway.
So
it
seems
that
you
agree
that
this
is
kind
of
regressive
yeah,
so
we'll
take
further.
E
E
C
And
so
yeah
I
I
didn't
have
the
link
Sandy
so
I,
just
added
them
with
the
course
I'm
going
so
there's.
Basically,
the
difference
is
one
conversation.
I
was
having
with
Linden
tiger
and
I'm
in
the
CI,
so
it
was
around
and
the
parallelism
today.
So
as
of
today,
the
CSA
snapshot
right,
it
is
a
it
is
linear
in
manner
you
take
one
PVC,
you
take
a
snapshot.
C
You
wait
for
10
minutes
for
it
to
complete,
and
similarly
you
do
for
n
snapshots
and
even
I
realized
further
in
the
code,
this
polling
happens
twice
once
in
the
CSI
plugin
and
once
in
the
core
code.
C
So
this
takes
around
two
into
n
into
10
minutes
for
like
for
the
backup
to
complete,
which
is
like
very
very
long
in
some
sense
for
failures
scenarios.
So
we
had
a
lot
of
discussion
around
how
we
can
fix
it
and
it
seems
like
the
overall
conclusion
is
to
go
through
the
Bia
V2
route
for
CSI,
plugin
and
I
also
went
through
it.
The
code
seems
neat
enough
to
implement
in
the
CSI
plugin.
C
B
So,
as
far
as
I
know,
no
one
started
that
work
yet,
but
as
I
understand
it,
that's
part
of
the
data
mover
work
that
that's
kind
of
being
done
as
part
of
the
item
ever
work.
That's
part
of
the
data
member
design
is
talking
about
changing
that
so
I
believe
that's
supposed
to
be
done
as
part
of
1.12.,
but.
G
E
So
if
you
take
a
look
at
the
data
mode
design
right
there,
Linden
is
not
on
the
call,
but
he
has
mentioned
that
we'll
be
porting.
The
CSI
plugin
right.
E
B
C
The
reason
am
I
asking
is:
basically
we
are
trying
to
see
if
we
need
to
prioritize
this.
If
we
do,
we,
we
might
implement
it
ourselves,
but
again,
I
did
not
wish
to
duplicate
work
with
everyone.
I.
B
Would
say
if
you're
interested
in
that
talk
to
Linden
about
it,
it
might
be
that
he
can
use
your
help
doing
that
as
kind
of
working
in
parallel
with
the
other
data
River
work
is
on.
G
B
Working
on
that
might
be
helpful,
but
I
again,
I,
don't
really
have
any
insight
on
the
specifics
of
when
they
planned
on
doing
that.
As
part
of
this.
B
C
It
will
take
some
while
for
now.
E
C
Yep
I've
got
it
but
I'll
just
maybe
didn't
direct
keeping
Linden
if
they
have
any
updated,
ETS
on
it
or
otherwise.
I'll
see
how
it.
B
Can
work
with
yeah
and
again
I
think
as
mentioned
earlier,
that
I
think
they're
mostly
off
this
week
on
holiday,
but
certainly
when
they
get
back
yeah.
C
Okay,
cool-
and
that
is
the
first
one.
The
second
one
is
it's
a
PR
specific
to
the
Azure
plugin
I'm,
not
sure
how
many
folks
in
this
call
have
a
idea
about
it,
but
venka
is
already
reviewing
it.
If
anyone
else
wants
to
help
review.
This
we
are
like
please
go
at
I,
can
walk
through
this,
but
I'm
not
really
sure.
If
there's
anyone
in
the
audience
who
who
will
be
interested.
So
this
this
is
the
other
PR
just
to
request
for
review.
C
I'll
give
a
quick
overview
so,
basically,
as
of
today,
the
older
code
that
we
had
it
used
to
assume
that
the
blob
endpoints
for
Azure
are
in
the
specific
pattern.
It
is
in
the
pattern
of
windows.co.net,
that's
specific
URI,
but
that
assumption
no
longer
holds
the
URI.
Is
that
Azure
uses
for
its
storage
account
can
kind
of
vary.
It
can
have
some
zonal
zxx.azio.storage.net.
C
The
URI
pattern
has
changed
and
it's
not
a
fixed
pattern
anymore,
so
this
code
kind
of
makes
it
Dynamic
to
fetch
the
blob
endpoint
on
the
Fly
and
then
try
to
use
it
instead
of
using
using
a
autogint
like
kind
of
create
using
a
regex
to
form
it
in
some
way.
So
that
is
one
part
of
the
change.
The
second
is
the
AED
changes
that
these
folks
had
already
done
as
part
of
pr11.
That
was
I.
C
Don't
remember
the
name
of
that
person,
but
so
that
approach
I
also
brought
in
a
few
changes.
So
there
what
that
canvas?
If
you
do
not
provide
the
subscription
or
resource,
then
they
used
to
default
to
aad.
But,
as
per
the
discussion
we
kind
of
had
right,
it
was
not
making
a
lot
of
sense
because,
firstly,
we
wanted
to
be
a
deterministic
behavior.
We
don't
want
to
say
that
if
you
don't
provide
a
specific
parameter
will
not
we'll
use
this
particular
route.
C
So
so,
basically,
we
I
introduce
a
new
field
which
is
use
aad,
because
the
subscription
and
RD
are
required
to
fetch
the
blob
endpoint.
We
need
to
do
a
get
properties,
call
and
get
the
blob
endpoint
from
the
search
account.
So
that
is
second,
second,
is
the
the
first
part.
Was
we
fetched
the
Broadband
Point
through
using
an
API
call?
Second,
is
for
aadv
explicitly
set
a
flag,
and
the
third
thing
is:
there
is
one
customer
issue
where
basically
the
we
in
in
this
particular
course.
C
Up
till
today,
we
used
to
make
a
list.
Keys
call,
and
the
list
case
call
was
I
had
to
go
through
the
management,
plane
of
azure
and
kind
of
gets
and
end
up
ends
up
getting
throttled
a
little
bit.
So
this
AED
route.
What
did
it
was
doing
so
far?
Is
it's
directly
talking
to
the
storage
account
the
underlying
data
plane
so
now
introduce
introducing
this
get
properties?
Again,
we
were
running.
We
will
run
into
that
sortling
issue.
So
just
to
summarize
right.
C
The
third
point
is
the
user
can
directly
provide
the
storage,
URI
The,
Blob
endpoint
of
the
storage
form,
and
we
will
directly
use
it.
I'll
summarize
quickly
right
so
first
thing
is,
instead
of
generating
a
URI
on
the
basis
of
a
specific
pattern,
we
try
to
fetch
it
second
is
we
have
an
explicit
flag
for
using
Aid
and
the
third
is
the
user
can
choose
to
specify
the
endpoint
themselves?
In
that
case,
you
will
not
make
any
unnecessary
calls
which
could
throttle
their
account.
C
I
have
the
chat
conversation
also
with
on
slack
everything
is
discussed
there.
I
can
point
you
to
that
as
well.
I'll
see
you
there.
If
anyone
else
is
interested,
I
can
also
ping
you.
There.
E
C
So
this
is
another
issue
which
we
encountered.
Basically,
as
of
today,
I
already
talked
about
the
the
linear
behavior
of
the
CSI
plugin
right.
We
do
10
minutes
per
snapshot
and
wait
for
it.
But
the
issue
is
that
after
that,
10
minutes
right,
we
don't
clean
up
that
snapshot.
So
what
that
ends
up
doing
is
in
the
cluster.
That
creates
a
lot
of
bad
snapshots
which
are
in
a
bad
shape.
C
So
if
you
think,
if
you
do,
let's
say
10
backups
per
day
and
you
do
it
for
let's
say
three
days
and
they
continuously
fail
and
let's
say
each
backup
had:
let's
say
three
snapshots:
you
are
unnecessarily
creating
90
snapshots,
which
are
like
useless,
pretty
useless
for
the
user.
So
what
that
ends
up
doing?
Is
it
throttles
the
CSI
drivers
because
they
have
to
keep
because
CSA
driver
Works
in
an
infinite
way?
C
It
will
have
an
exponential
backup,
but
it
will
if
every
five
minutes,
or
so
it
will
try
to
take
that
snapshot
if
it's
in
a
bad
shape,
so
that
ends
up
has
a
like
a
bad
effect
which
leads
to
throttling
the
CSI
driver
so
I.
So
we
don't
have
an
issue
for
this
right
now,
I'll
create
it
in
the
on
the
repo,
but
basically
the
proposal
here
is:
if
a
snapshot
fails
after
10
minutes,
we
clean
it
up,
instead
of
letting
it
stay
in
the
cluster.
Let
me
know
if
you
folks
feel
otherwise.
G
C
The
part
that
you
also
mentioned
I
am
aware
of
that
code.
I'll
also
have
a
look
at
it
if
anything
has
changed
yet,
but
yeah
this.
This
is
again
similar
lines
as
well,
but
I
I
want
to
take
this
particular
work
in
conjunction
with
the
Pia
ready
thing,
because
we
are
also
concerned
about
the
parallelism
of
where
everybody
are
all
those
three
things
are
in
similar
lines
along
the
CSI
product,
yep.
C
C
Definitely
any
other
thoughts
books,
if
you
think
otherwise
is
this
Behavior?
Okay,
with
you
folks.
J
J
I
just
have
a
quick
one
comment
about
the
meetings
themselves,
I
mean
because
I
don't
join,
or
at
least.
J
Been
joined
regularly
so
I'm
having
really
tough
time
finding
out
whether
this
given
week
is
you
know,
U.S
friendly
time
zone
or
the
Beijing
friendly
time
zone,
or
the
bestest
solution,
of
course,
is
to
have
a
calendar.
Invite
right
just
like
other
kubernetes
things
have
all
of
them.
Have
that,
but
short
of
that,
can
we
they
add.
You
know
next
to
the
date
in
these
notes.
Can
we
add
the
time
zone
then
at
least
that
will
immediately
tell
us
what
is
the
previous
meeting
you
know,
then
then
we
can
easily
figure
out.
J
B
Or
something
like
that,
yeah
I
think
we
should
do
that.
That
would
be.
That
would
be
an
easy
thing
to
add
to
just
these
notes
here,
but
there
is
a
calendar
invite
on
the
Google
group,
the
Valero
development,
the
Valero
project.
Google
group
I
think
that
I
think
that
currently
does
have
a
calendar
invite.
But
it's
basically.
J
B
B
Yeah
so
so
we
say
May,
2nd
and
then
in
parentheses
you
know
the
U.S
friendlier
or
it
might
make
sense
to
indicate
in
the
beginning,
you
know
which,
which
time
zone
this
was
yeah
exactly
that
just
makes
it
another
another
place.
Someone
can
look
at,
you
know
very
quickly
to
say:
okay,
if
they're
not
looking
at
the
calendar
to
know
which
week
is
which.
A
A
I
think
on
the
previous
one,
I
was
expecting
that
that
you
received
that,
but
obviously
that's
not
how
groups
works,
so
we
have
to
figure
out
some
other
more
proper
way
to
to
send.
These
invites
out
not
sure
if
I
can
do
that
on
the
on
the
slack
notification,
stuff
I'll,
take
that
and
I'll
try
to
fix
it
in
or
make
it
a
little
bit
more
reusable
I'm.
Sorry.
A
I
was
expecting
to
happen
the
same,
but
it's
not
so.
If
it's
not,
then
I'll
try
to
figure
out
right.
J
B
G
G
K
K
E
I
think
that's
what
I
did
as
well
once
the
invites
were
changed
like
six
months
back.
B
Yeah
right
right
because
because
the
amount
sends
the
list,
so
you
click
on
that
in
your
email
and
then
you
say
add
to
calendar
and
then
then
your
personal
calendar
has
that
invite
from
that
point
forward.
But
I
guess
the
challenge
is:
if
you
weren't
on
the
list
when
that
was
sent,
you
might
not
have
seen
it
or
noticed
it,
because
it's
not
going
to
retroactively.
Add
it
to
your
calendar
because
you
exercise
a
list.
K
B
A
So
you
mean:
did
you
create
in
that
in
the
groups?
Okay,
I'll.
Take
this
one
and
I'll
try
to
figure
out
until
next
next
week,
not
sure
how
to
approach
that
right
now,
right
now,
but
yeah.
G
K
Right
so
if
you
have
something
like
this,
you
can
add
events
to
it
and
then,
if
you
wanted
to
add
this
to
your
own
calendar,
you
just
in
the
bottom
right,
you
see
a
plus
button.
B
A
I'm
just
sure,
okay,
yeah.
K
And
then
you
can
say,
add
yeah
and
then
that
that
would
add
add
this
calendar
to
one
of
your
calendars
on
the
left
which,
which
you
can
check.
You
know
if
you
want
to
see
it
or
not,
and
this
will
be
always
be
updatable
by
you
and
everyone
else
to
join
us.
We
will
have
an
updated
event.
G
K
K
Just
generated
from
Zoom,
but
okay
yeah
you
can
you
can
make
a
new
one
by
pressing
Plus
on
the
other
calendars
on
the
on
the
left.
K
K
G
A
K
With
you
no.
A
A
No,
no
I
know
how
to
do
that.
I'm,
just
thinking.
What's
the
best
choice
for
everyone
all
right,
I'm,
sorry,.
A
Okay,
anyone
else
any
any
other
topics
for
today.
A
Nope
all
right,
then,
thank
you
very
much.
Everyone
hope
you
have
a
great
rest
of
the
day
and
talk
to
you
in
one
week,
bye.