►
From YouTube: KubeVirt Community Meeting 2022-08-03
Description
Meeting Notes: https://docs.google.com/document/d/1kyhpWlEPzZtQJSjJlAqhPcn3t0Mt_o0amhpuNPGs1Ls/
A
A
If
anyone
is
new
on
this
call-
and
I
would
like
help
getting
something
written
into
the
agenda
and
getting
access
to
read
and
write
to
the
kubert
community
meeting
agenda
feel
free
to
speak
up
and
we
will
help
you
with
that.
A
And
if
you're,
new
and
you'd
just
like
to
say
hello,
and
let
us
welcome,
you
feel
free
to
introduce
yourself
real
quick.
A
A
A
It
looks
like
andrew
you
have
some
things
to
speak
about
for
kubecon.
A
Generally,
it
looks
like
we
are
welcoming
anyone
who
wants
to
present
virtual
office
hours
for
kukan
or
if
you're
interested
in
presenting
or
joining
panel
style,
please
feel
free
to
pipe
up.
Then
we
will
connect
with
you.
A
B
A
And
it
looks
like
there
is
some
review
commenting
activity
on
here.
Do
you
have
any
specific
questions
about
any
of
the
feedback
that's
been
discussed
so
far.
B
A
Right,
we
can
either
do
this
now
or
async.
Does
anyone
have
anything
to
speak
about
on
this
proposal
immediately
or
anyone
interested
in
following
up.
A
I'm
gonna
go
ahead
and
recommend
also
touching
base
on
the
mailing
list
for
that
just
to
make
sure
that
you've
reached
out
to
all
channels.
B
A
If
what
is
that.
B
B
B
A
All
right,
macbeth
mode
for
pod
networking
proposal.
C
Yeah
I
was
talking
about
hello.
My
name
is
andre.
I
was
talking
about
this
feature
already
and
now
I
decided
to
make
proposal.
I
know
that
it's
not
going
to
be
merged
soon,
because
I
want
to
say
that
we
are
going
to
have
another
interface,
for
you
can
open
rendered
version
in
the
first
link.
There
is
some
beautiful
pictures.
C
So
I
just
described
the
every
step
should
be
done
to
make
it
working,
and
why
do
I
think
that
current
networking
design
is
not
good
for
for
binding
port
network
and
how?
Actually
I
made
it
working.
C
C
So
yeah
here
you
can
see
that
macro
tab
mode
is,
has
less
latency
than
bridge
and
masquerade
modes
and
a
little
bottom.
You
can
see
the
final
design
how's.
It
looks
like.
C
A
And
that
looks
like
a
a
lot
of
very
thoughtful
work.
Does
anyone
on
the
call
right
now
have
any
additional
comments
or
feedback
with
the
additional
maturity
of
this
design?.
D
Okay,
all
I
can
say
right
now
is
I'll
review
your
your
proposal,
nevertheless,
but,
as
you
already
know,
we're
trying
to
come
up
with
a
plugable
way
for
interested
parties
to
add
whatever
kind
of
binding
modes
to
wherever
they
want
and
to
keep
that
thing
outside
of
kubert.
So
that's
the
the
thing
that
we
think
the
community
would
really
benefit
from
like
a
way
to
plug
the
like
their
intended
binding
ways
and
have
them
maintain
those
binding
modes
outside
of
the
keyboard
ripple.
D
Use
this
proposal
to
figure
out
the
correct
way
of
doing
that,
I'm
unsure
I
will
do
my
best
to
get
proper
feedback
to
your
proposal
in
the
coming
weeks.
That's
as
far
as.
C
C
Yeah,
thank
you.
I
think
that
making
this
proposal
will
make
you
easy
to
understand,
which
calls
I
actually
need
like
to
set
the
device
permissions,
for
example,
for
the
c
groups,
so
it
will.
C
It
will
allow
you
to
develop
better
design
for
this
plugable
logic
and,
of
course,
in
the
future,
it
will
help
us
to
develop
the
plugin
for
making
this
working
cool.
D
And
I
just
want
to
say,
like
I
think
it's
very
like
the
way
you
just
bypass
the
the
device
plug-in
thing
and
kind
of
adjusted
the
c
groups
thing
on
the
pods
template.
I
think
that's
very.
D
I'm
a
bit
unsure
on
the
how
wise
that
is
or
not,
but
it's
impressive
to
me
to
say
the
least,
so
I'm
very
eager
to
look
at
this
proposal
in
depth.
A
Thank
you
all
right,
andrew
burden.
I
know
that
you
are
having
some
audio
issues.
Did
you
want
to
speak
anything
particular
about
the
agenda
note
for
kubecon.
E
North
america,
hello,
can
you
hear
me?
Yes,
we
can
yeah.
A
E
It's
all
fixed,
not
so
much.
I
think
it's
all
there.
We've
got
another.
I
think
10
days
left
before
or
maybe
nine
days
to
apply
for
the
virtual
office
hours.
I
sent
an
email
last
week,
seeing
if
anyone's
interested.
So
if
I
haven't
heard
back,
if
no
one's
interested,
we
won't
run
it.
So
if
you
do
have
an
interest
in
being
involved
in
in
it
it's
about
40
minutes,
the
the
goal
is
to
create
an
arena
where
we
can.
E
People
who
are
heavily
involved
in
the
project
can
have
like
a
back
and
forth
with
the
with
any
of
the
attendees.
In
a
virtual
thing,
we
can
do
demos
we
could
show
off
some
stuff
just
to
like
do
a
q,
a
it's
entirely
kind
of
up
to
us,
how
we
decide
to
run
it
and
cool
I've
got.
I've
got
one
person,
excellent,
so
yeah.
That
was.
That
was
all
I
wanted.
Thank
you.
A
E
Yeah,
I
think
in
valencia.
I
think
we
had
what
was
it
five
people
and
that
was
really
good,
because
we
had
five
specialties
and
five
separate
demos
are
able
to
run
when
there
wasn't
a
question
and
answers,
and
that
was
great.
I
think,
between
three
and
five
is
perfect.
A
All
right,
it
looks
like
this
house
improve
on
it
right
now.
That
was
just.
B
A
A
B
A
A
I
mean
keep
intro.
Can
we
get
an
overview
on
this.
B
A
A
E
The
link
works
for
other
people,
essentially
as
described
by
the
person
they're
trying
a
vmi
live
migration
triggered
by
kubernetes
cluster
upgrade
and
the
vmi
events.
The
vmi
gets
deleted
training
a
note
manually
with
that
cluster
upgrade
works.
Fine
if
anyone
can
jump
in
on
that
that'd
be
great.
Thank
you.
A
Last
I
recall
we
currently
support
hybrid
clusters.
Is
that
still
correct.
B
F
B
B
A
Something
I
was
actually
looking
at
this
weekend:
we're
not
producing
arm64
builds
of
vert
ctf
for
cuddle
are
we?
I
only
saw
x,
86.
F
I'm
going
to
do
that.
I
do
know
that,
having
built
hubert
on
the
raspberry
pi,
you
can
create
arm
or
arm
64
builds,
but
I
don't
think
we
do
it
automatically.
There
may
be
some
complication
behind
the
scenes
to
actually
pull
that
off,
but
that's
our
problem.
It's
a
reasonable
request.
A
Okay,
when
I
was
looking,
I
I
checked
to
see
if
we
were
using,
go
releaser
for
that
and
we're
not.
So
if
we,
if
we
had,
I
was
going
to
probably
do
the
pull
request,
but
I
didn't
continue
looking
to
see
how
exactly
we're
producing
those
binaries.
A
D
A
So
it
looks
like
they're
describing
a
situation
where
a
vm
with
graphics
device
not
attached,
but
gpu
attached
is
exhibiting
a
normalness
behavior.
B
Here
is
andre
from
the
desk.
I
just
saw
that
and
I
passed
to
my
team.
I
replied
to
you
guys
next
week.
Okay,
great,
thank
you.
F
I
look
see
like
he's,
not
that's,
actually
something
he's
spearheading,
which
is
why
I
immediately
said
his
name.
We
can
track
that
down,
but
there
is
somewhere
a
list
of
all
the
the
check
list,
items
that
we
wanted
in
place
and
they're
not
there,
yet
it
isn't
github
so
it'll.
Take
me
a
minute
to
find
it,
though,.