►
From YouTube: Kuma Community Call - December 14, 2022
Description
Agenda:
- Kuma 2.0.1, Kuma 1.8.2
- Recap of the 2022
- For ’23: Move from CNCF Sandbox to Incubating
- Q&A
Kuma universal + NSM vl3 integration feedback
https://github.com/kumahq/kuma-tools/tree/master/kuma_nsm_vl3
A
So
yeah,
okay,
hello,
everyone
welcome
to
the
Guma
Community
call,
please
add
your
name
to
the
attend
the
list,
but
I
think
everyone
already.
No,
not
everyone
yeah.
So
please
add
your
name
to
that
and
the
list
and
also
feel
free
to
submit
any
agenda
topic
you'd
like
to
discuss
today.
A
So
I'd
like
to
start
with
the
releases.
Last
week
we
released
to
patch
releases,
Kuma,
2.0.1
and
2.8.2
and
I
think
these
releases
are
great
and
they
have
fixed
for
very
tricky
problem
that
we
had
I
guess
for
a
while,
and
it
was
very
hard
to
debug
but
yeah
kudos
to
Jacob
who
discovered
the
problem
and
fix
it
so
yeah,
and
also
we
fixed
the
problem
with
resource
ordering.
It's
also
pretty
important
fix.
A
A
Okay,
what
else
this
is
the
last
Community
call
of
the
year?
That's
why
we
were
thinking
that
it
makes
sense
to
do
some
recap
of
the
year
and
yeah.
We
have
a
few
slides
so
yeah
Kuma
2022.
What
we
did.
We
did
one
major
release,
which
is
Kuma
2.0.
A
A
These
are
the
most
significant
ones,
so
we
yeah
we
released
on
egress,
we
released
built-in
Gateway
and
Gateway
API
support,
and
we
also
marked
this
I
mean
we
shifted
this
from
experimental
state
to
stable
State.
We
massively
proved
performance,
we
added
inspect
API,
which
is
a
new
way.
You
can
now
observe
your
policies
and
your
mesh
also
mesh
membership,
the
new,
transparent,
proxy
and
cni
are
cross,
merge,
Gateway.
A
Also
a
very
nice
feature.
Now
you
can
easily
set
up
your
communication
between
two
meshes.
We
added
Ram
support.
We
added
app
metrics
collection,
ebpf.
We
released
our
first
policies
with
the
new
format.
At
the
end,
our
GUI
definitely
improved
I
think
drastically.
Since
the
last
year
now,
it's
way
better
and.
B
A
Goodness
to
Philip
and
JC
so
yeah
what
else
we
have?
We
have
stable
triage
process
throughout
the
year.
I
hope
you
noticed
that
that
now
almost
all
issues
are
being
processed
or
within
a
week,
and
we
also
have
we
adopted
madr
format
for
all
new
features.
So
now,
if
you
want
some
features
to
be
implemented,
you
should
create
madr.
We
have
a
rule
that
three,
you
need
three
approval
on
these
and
it
will
be
merged
and
yeah.
A
If
you
didn't
know,
cncf
has
dashboards
for
all
projects,
and
this
is
a
nice
chart
that
shows
how
many
issues
and
PRS
were
open
and
closed
or
in
2022,
and,
as
you
can
see,
it's
a
lot.
I
think
it's
about
800
issues
and
more
than
2,
000,
PRS
and
I
think
it
shows
great
base
that
Kuma
is
moving
so
yeah.
Do
we
have
any
questions
for
this
small
presentation.
A
B
Yeah,
so
one
thing
that
is
worth
mentioning
is
that
we'd,
like
in
the
coming
year
to
move
from
sandbox
to
incubating
one
thing
we
need
mostly
to
get
from
sandbox
to
incubating
is
to
prove
that
we
have
usage.
I
can't
remember
the
exact
wording
from
the
cncf,
but
that
we
have
service
users
of
Kuma
so
being
an
open
source
project.
B
It's
always
tricky
to
figure
out
who
are
the
users
of
your
project
so
like
please
make
yourself
known
to
me
or
like
to
just
someone
else
in
the
commenters
so
that
we
can
start
building
up
this
and
and
being
able
to
submit
this.
This
proposal
to
the
cncf
to
move
from
sandbox
to
incubating
cool
thanks.
A
Yes,
thank
you,
yeah
I,
think
NSM
folks
are
here
and
I
can
give
Awards
to
Daniel
or
which
is
love.
Who
wants
to
take
the?
What.
C
Yeah,
hello,
it
will
be
me,
so
we
showed
you
edema
and
also
we
created
and
merged
pull
request
into
one
of
your
repositories,
and
we
wanted
to
get
some
additional
info
about
your
opinions
and
just
your
feedback
in
general.
So
I
have
a
few
questions.
If
you
can
please
answer,
it
will
help
us
to
improve
this
integration
in
future
in,
in
general,
understand
the
how
well
we
prepared
the
demo
and
prepared
the
dogs
in
for
your
repositors.
So
how
clear
was
the
integration
in
general?
C
Yes,
so
first
of
all,
I
can
probably
send
a
link
into
the
chat
just
in
case,
if
folks
have
not
yet
seen
it,
because
it's
it's
in
one
of
your,
not
main
Representatives
so,
and
probably
people
not
check
it
out
as
often
as
the
main
ones.
So
the
first
question
is:
how
do
you
feel?
Is
it
ready
to
be
used
already
or
not,
because
it
was
an
integration
to
use
NSM
for
as
a
network,
our
wheel,
three?
A
D
A
Do
you
want
this
to
have
as
a
like
question
answer
format
yeah?
It.
C
I
can
I
can
read
all
of
them,
but
I
think
it
will
be
better
if
we
go
one
by
one
yeah.
B
What
do
you
mean
by
how?
How
do
you
feel
it's
ready
to
use
like
I
know,
I'm,
not
really
sure
I
fully
understand
the
question:
is
it
about
integrating
that
into
Kuma
directly
or
if
it's
just.
C
A
ready
solution
for
youth-
or
you
think
it
needs
something
to
improve
before
people
would
use
it
like,
even
just
for
maybe
their
personal
stuff,
not
for
production
registers.
You
can
see
it.
D
Okay,
okay,
yeah,
okay,
so
just
to
share
my
opinion
on
this
one.
The
main
problem
that
I
see
right
now
with
the
approach
that
we
have
here,
is
that
we
manually
need
to
create
and
apply
data
plane,
proxy
and
overall,
that
the
deployment
is
run
with
a
universal
mode
on
kubernetes
right,
which
is
not
something
that
our
users
are
used
to
do
right.
So
usually
you
are
used
to
have
sidecar
injection
and
so
on.
So
it's
much
easier
to
run
this.
C
C
I
thought
about
it
way
before
like,
but
we
found
some
problems
that
we
haven't
figured
out
yet
to
fix
it,
but
I'm
happy
to
hear
that
you
have
an
opinion
so
strong
and
you
may
help
us
in
future
with
it
probably
I,
don't
know,
but
it's
really
great
to
hear
it.
D
Yeah,
just
to
you
know
just
to
give
you
like
honest
perspective
of
our
users
and
so
on
right,
so
I
haven't
seen
a
while
in
a
in
a
while
and
a
universal
deployment
on
kubernetes
I
only
seen
this
in
a
case
when
you
run
in
multi-zone
deployment,
a
global
CP
in
a
universal
mode
right,
but
Global
CP
does
not
handle
DP
connections
and
networking
between
data
playing
proxies
right.
So
that's
like
a
different
thing,
so
yeah!
So
that's
it.
C
Yeah
also,
our
understand
setup
will
help
with
the,
but
it
you
don't
have
to
expose
control
plane.
You
can
use
NSM
IPS
to
get
to
the
control
plane
without
exposing
it
publicly
like
we
handle
all
the
security
stuff
from
in
NSM,
and
you
can
use
also
security
stuff
from
Kuma.
D
D
C
What
exactly
not
also
we
we
provide
support
to
different
by
Metals.
Give
me
a
second
I
will
just
open
our
info
somewhere.
I,
don't
remember
by.
C
C
Yeah,
so
we
support
OS,
special
IV,
VPP
and
kernel,
of
course,
so
you
may
have
even
wider
support
of
VMS
and
metals
than
just
the
basic
VMS
like
Ubuntu
and
such
stuff.
D
I
see
so,
basically
again,
the
main
use
case
is
when
you
want
to
connect
multiple
kubernetes
clusters
or
VMS
with
kubernetes
clusters
and
have
a
rootable
IPS
between
pods
and
VMS.
Is
that
correct,
yeah?
Okay?
So,
basically
our
move
is
on
deployment
right
model,
so
yeah.
So
this
is
one
more
thing
that
I'm
a
little
bit
concerned
here
is
if
the
main
use
case
is
to
have
to
use
this
with
the
movies
on
deployment.
D
What
I
see
that
NSM
solves
here
is
to
have
a
flat
networking
between
between
multiple
kubernetes
clusters
and
so
on,
so
I
guess.
If
we
would
like
to
utilize
this,
we
would
need
to
do
some
additional
work
on
the
kuma
site
to
support
networking
topology
of
flat
Network
between
multiple
kubernetes
clusters.
D
D
Yeah,
true
yeah,
we
could
do
this,
we
could
implement
this.
My
point
is
that
this
is
like.
D
D
So,
of
course
we
could
we
could
internet
sure,
but
it's
additional
work
to
consider.
A
Yeah,
which
side
could
you
please
repeat
what
you
said
initially
I
just
lost
a
little
bit.
A
C
We
also
support
different
kinds
of
metals
like
and.
A
C
A
To
create
a
flat
Network
between
our
kubernetes
clusters
and
I
think
that
Jacob
required
is
that
one
needs
a
support
for
fantastic
kubernetes
support
for
flat.
A
C
Yeah
I
have
another
question
related
to
the
the
solution,
but
I
think
we
already
discussed
it
about
order.
Have
it
have
to
be
done
to
say
that
it's
already
production
ready
solution,
but
I
think
we
discussed
it
so
we'll
skip
this
one,
so
another
one
is:
could
it
be
solved
to
could
be
used
to
solve
problems
in
Kuma
current
scenarios?
Maybe?
C
Yeah,
okay,
then
we'll
give
this
one
too,
and
also
do
you
have
any
questions
about
the
demo
that
we
showed
you,
and
maybe
we
should
so.
We
should
clarify
it
for
the
next
call,
maybe
and
or
just
edit,
the
our
readme
in
your
repository.
So
it
would
be
clear
to
the
users
who
will
check
it
out.
A
Yeah
I
think
the
main
problem
for
us
is
to
that
people
most
people
on
this
call
don't
have
expertise
with
this
Sr.
I
o
v
and
PPP
things,
and
that's
why
it's
not
really
clear
what
a
problems
are
we
solving?
So
maybe
it
would
be
better
to
if
you,
when
you
present
the
demo
to
say
what
exactly
problem
we
trying
to
solve
yeah
already.
We
should
just
read
more
things
about
a
network
mechanisms
to
understand
better.
C
Yeah,
okay,
I
get
it
I,
think
I'll
discuss
it
with
the
team.
Probably
we
will
figure
something
out
just
to
explain
it
in
some
basic
and
general
terms,
so
people
who
are
interested
in
deeper
understanding
they
will
check
additional
for
themselves
but
provide
them
basic
info
in
with
me,
I
think
it
will
be
helpful
in
future.