►
From YouTube: CNCF TOC Meeting 2022-12-13
Description
CNCF TOC Meeting 2022-12-13
B
C
Yeah
I
have
I,
have
guns
through
this
comes
through
it
and
have
some
notes,
so
this
party
it
makes
traffic
interception
at
the
soccer
level
and
The
Leverage
is
ebpf
to
directly
forward
packets
from
one
socket
to
another
socket,
instead
of
instead
of
going
through
the
existing
Linux
kernels
iptable
right,
which
so
it's
it's
speed
up.
The
data
pass
in
the
service
match,
which
I
think
it's
a
it's
a
great
Improvement
and
also
at
the
industry,
is
moving
on
towards
that
direction.
B
E
C
D
C
C
Oh
yeah
I
just
want
to
add
that
you
know
there
are
other
I
see
other
presentations.
You
know
the
tcpip
bypass
and
example
bypass,
which
I
just
you
know
the
same
problem.
You
know
using
this
eppf,
so
maybe
they
can
try
to
expand
that
incorporate
more
contributors
or
or
maintainers
from
other
companies
yeah.
That's
the
one
I
think.
B
Yeah,
okay,
Dev
space
Dev
space
is
interesting.
This
is
from
The
Loft
folks
as
well,
and
the
thing
that
I
liked
here
was
this
architecture
flow
diagram
where
you
start
with
one,
is
a
yaml
file
or
a
Json
file
and
then
essentially
bootstrap
a
work
area
for
yourself
and
to
build
and
push
and
run
on
kubernetes
and
then
so.
E
C
F
C
G
This
and
we
already
have
some
things
like
telepresence
we've
had
for
a
long
time
right,
I
bet
you
we've
got
a
number
of
projects
that
already
fit
into
this
Dev
space
yeah.
We
could
probably
justify
it
today,
but
that's
that's
a
separate
thing.
This
project
has
been
around
since
2018.
It's
got
over
3
000
Stars.
If
you
only
know
how
to
be
an
open
source
project,
yeah
yeah.
B
E
I'm
a
little
confused
and
like
how
this
plays
out
for
other
sandbox
projects,
because
originally
we
had
sent
them
back
to
be
considered
potentially
as
a
sub-project
underneath
Opa
oppa
doesn't
necessarily
feel
like
they're
a
good
fit
because
they
want
to
see
broader
adoption,
which
the
sandbox
provides
a
place
for
vendor
neutral
collaboration,
increasing
adoption.
But
it
is
not
designed
necessarily
to
be
the
adoption
kind
of
like
producer
for
projects
like
that.
That
really
happens
when
they
reach
a
higher
level
of
maturity
and
better
stability.
E
So
if
this
is
accepted
as
a
Sandbox
project
and
Oppa
and
like
it
started,
we
start
seeing
widespread
adoption.
They
get
10
000
users
as
a
milestone,
which
is
interesting,
does
oppa
suddenly
come
in
and
say
all
right.
We
want
you
as
a
sub
project,
and
now
we
remove
them
from
the
sandbox
like.
How
does
that
work?
And-
and
how
do
other
people
feel
about
that?
Potentially
playing.
B
I
think
it
can
go
either
way
right
so
starting
off
by
itself,
and
then
the
Oppa
project
saying
yes
come
join
us
I
think
that
that
is
good
and
yeah.
So
we've
done
both
ways
right
like
so,
we
Helm
was
part
of
kubernetes
and
then
we
split
out
and
then
we
have
like.
For
example,
there
was
one
more
project
in
continuity,
space
that
you
know
came
back
into
containerdy
as
a
sub
project,
and
that
was
outside.
Do
you
remember
which
one
Justin.
H
I
H
F
G
F
F
B
We
can
talk
to
them
about
that
right,
Justin.
We
can
tell
them
that
you
know
you
can't
have
your
own
thing
as
a
default.
We
tell
them
many
projects
anyway.
The
same
thing
like
we
told
cubescape
too,
and
they
came
back
to
us
and
said
it's
okay.
We
will
not
throw
in
a
default
which
points
to
our
own
vendor
stuff.
F
B
J
G
G
B
D
J
B
B
A
B
G
E
Yeah
I
took
a
quick
look
at
it.
I
I
liked
the
project
I.
Think
it's
useful.
However,
I
had
some
concerns
that
it
seems
like
it's
gcp
heavy,
but
I
also
am
not
sure
what
the
potential
adoption
is
going
to
look
like
for
this
project
like
I'm,
not
sure
that
I
can
see
this
actually
going
fully
fledged
and
reaching
all
the
way
out
and
becoming
a
graduated
highly
mature
project
adopted
across
the
ecosystem,
with
the
way
that
it's
currently
set
up
with
just
gcp.
Only.
C
D
B
B
G
Yeah
I
I
like
where
it's
going
and
I
like
the
the
developer
space.
What
I
don't
like
is
obviously
the
low
number
of
contributors
but
the
the
single
vendor
right.
We
talked
before
just
about
you
know
not
favoring
a
single
vendor,
and
this
is
all
targeted
at
gcp
at
the
moment
yeah,
and
so
it's
not
a
multi-vendor
project,
it's
kind
of
a
very
useful
project,
but
one
that
I
think
needs
to
grow
into
to
being
more
multi-vendor
before
it
fits
within
the
cncf.
B
B
C
B
C
B
It's
one
person
doing
it
right,
so
we
can't
really,
like
you,
know,
Drive
hard
on
that
side.
Yes,
when
the
community
expands,
I'm
sure
they'll
get
asked
about
other
languages,
I'm
sure
you
know,
people
are
asking
the
person
already
for
other
languages,
just
maybe
that
is
part
of
the
reason
they
are
coming
here
because
you
know
hey.
We
can
build
a
community
here
in
the
sandbox,
but
then
we
do
have
this
other
restriction
of
like
hey.
You
need
to
be.
E
E
B
B
C
C
That
shows
very
strong
technical
Merit
or
what
is
that
yeah,
the
thing
technical
Merit
that
you
know
they
will
provide
and
the
user,
mostly
from
Russian
speaking
countries,
so
I
think
the
reason
they
want
to
apply
is
they
want
to
be
more
International
and
and
also
is
that
we
are
not
sure
whether
moving
this
part
to
party
to
cncf
is
relevant,
but
we
are
happy
to
discuss
it.
So
I
think
you
know
yeah
that
part.
B
H
F
They
also
have
their
own
tough
implementation,
which
I
had
not
I
mean
one
of
the
one
of
the
things
that's
kind
of
about
this
project
is
that
I've
never
heard
of
it
before
or
before
when
it
came
to
the
sandbox,
I
mean
I,
think
that
bring
it
into
the
ecosystem
and
having
some
more
relationship
with
Helm
and
the
other
things
might
be
quite
positive
for
them
because,
like
I
think
the
impression
I
get
is
they're.
Quite
they've
been
quite
isolated.
G
Yeah
and
my
concern
is
with
forking,
especially
with
things
like
Helm
and
maintaining
a
fork
is
keeping
things
up
to
date
for
security
and
just
a
brief
look
I
think
they're
using
their
Fork
version
of
Helm
has
known
cves
in
it,
and
so
they
need
to
close
the
security
system.
We're
gonna
Fork
the
projects
and
you
need
to
keep
up
to
date
for
security,
I,
think
yeah.
B
F
G
C
I
went
straight
to
so
it's
it's
a
problem
when
the
kubernetes
tenant
needs
more
than
one
name
space.
So
basically
it
groups
a
set
of
namespace
resources
into
a
new
abstraction
and
assigned
assign
it
to
a
tunnel.
I
think
this
technology
is
useful,
because
sometimes
our
tenants
is
not
just.
It
needs
more
than
one
namespace.
B
D
D
B
F
D
B
D
B
H
E
E
Explain
how
it
aligns
with
the
cloud
native
Computing
ecosystem
and
the
justification
that
they
provide
is.
First,
it's
easy
to
declare
a
development
environment
and,
oh
by
the
way
it
turns
into
a
container
and
then
the
second
one
is
really
that
it's
powered
by
next
and
that
they're
rapidly
going
to
overtake
Nyx.
E
Based
off
of
like
the
things
that
I'm
seeing
here
about
the
project,
I
think
the
project
is
cool.
Don't
get
me
wrong,
but
I
don't
know
necessarily
that
any
additional
information
they
provide
about
alignment
and
overlap
with
existing
projects
at
this
point
in
time
would
persuade
us
one
way
or
another
right.
B
G
B
G
D
F
B
F
F
G
C
A
F
Mean
that
that's
kind
of
I
mean
they
say
they
intend
to
be
the
reference
implementation
of
the
OCR
distribution
specification,
but
the
but
distribution
itself
is
in
fact
the
reference
implementation
of
the
SEO
distribution
specification.
I
mean
it's
and
that's,
and
it
always
has
been
the
fact
that
I
mean
you
know.
The
reality
is
that
everyone
implements
some
of
the
historical
Docker
pieces
for
compatibility
reasons,
but
like
that,
doesn't
it's
just
kind
of
where
to
declare
yourself
to
be
the
reference
implementation
of
something
that
already
has
a
reference
implementation
in
cncf.
H
F
I
mean
it
says
the
focus,
Target
distribution
and
go
Harbor
is
the
Legacy
Docker
format,
which
is
absolutely
just
not
true.
It's
literally,
it
literally
is
the
reference
implementation
that
everyone's
using
and
it's
where
all
the
work
on
new
features
is
prototyped,
I.
Think
so
I
just
don't.
Actually
you
know
think
that
I
mean
there's
an
opportunity
for
all
these
solutions
to
incorporate
us
as
a
library
or
back-end
process,
but
none
of
them
have
I
mean
it
just
seems
kind
of
I.
Don't
know
it
just
doesn't
ring
very.
D
B
F
F
F
F
B
On
a
technical
side,
I
don't
see
any
any
reasons
for
not
accepting
right
like
yes,
the
the
some
of
the
aspirational
pieces,
kind
of
definitely
rubs
the
wrong
way,
and
you
know
hoping
that
they'll
play
nice
after
they
join.
That
is,
if
they
have
a
well-diversified
community,
then
that
you
know
that
could
be
of
use
useful.
That
could
be
useful
foreign.
B
Okay,
I'm
gonna
say
on
a
technical
side.
There
is
no
issues
that
we
can
see
right
now,
so
I
think
we
should
go
for
it,
but
you
know
they
should
probably
tone
down
some
of
the
things
that
they
are
saying
here,
but
yeah,
but
that's
not
a
caveat
or
anything
so
I'm
I'm
up
for
going
forward
with
this
any
objections,
foreign.
E
G
They
seem
to
have
a
bunch
of
things
already
in
order
around
code
of
conduct,
governance
and
all
of
that
stuff
in
place
for
a
project
so
as
far
as
operating
as
a
project,
but
what
they
they
have
a
low
number
of
issues.
What
is
it?
44
total
issues
so
they've
had
a
low
amount
of
traction
there
they're
only
at
like
60,
some
total
pull
requests,
foreign
and,
although
they
have
lots
of
commits,
which
makes
me
think
either
now
or
early
on,
they
were
just
pushing
to
master
right.
I
mean.
E
So
I
I
think
part
part
of
the
problem
with
the
project
is
that
they
don't
have
a
wide
community-based
to
actually
like
open
a
bunch
of
issues.
Increase
PR's
they're
not
made
like
going
through
their
contributing
guide
and
trying
to
look
up
whatever
information
I
could
about
the
project.
They
don't
have
like
meetings
to
discuss
a
project
in
some
of
the
roadmap
planning
to
come
up
so,
like
I,
think
the
project
is
valuable,
I
think
there's
more
Community
Development
activities
that
could
be
going
on.
E
B
Yeah
I
I
think
this.
The
part
of
the
problem
is
this:
this
kind
of
solutions
would
be
like
top
down
driven
right
when
the
c
c
exos
want
to
make
sure
that
they're
production
systems
are
doing
okay,
then
they
put
Solutions
like
this.
On
top
of
that,
to
make
sure
that
people
who
can
only
access
they
they
lack,
they
can
access
only
the
things
that
they
should
be
accessing.
G
B
G
G
H
B
B
C
Yes,
yes,
I
went
through
it,
so
it
is,
it
is
used
to
improve
energy
footprint,
it
collapse,
performance
data,
related
data
based
on
ebpf
programs
and
then
Aggregates.
This
data
with
other
stats
is
pause,
those
to
promises,
metrics
and
then
it
uses
the
promise
promise
themes,
promises
metrics
to
Trend
the
machine
learning
models
to
estimate
energy
consumption
by
each
part.
C
So
it
said
it's
lightweight
and
it's
a
little
overhead
I
and
it
also
integrate
with
the
kubernetes
scheduler
and
auto
scalar
to
take
the
power
resource
to
take
the
power
resource
into
consideration,
to
support
better
performance
and
the
resource
utilization
rate.
So
I
think
it's
a
very
good
project.
F
B
F
I
mean
it
just
seems
to
be
the
just
the
Prometheus
exporter
and
not
the
rest
of
it.
That's
being
that's
listed
here
in
the
description
and
the
repo
and
not
the
ml
models
or
the
scheduler
or
any
of
the
other
pieces
which
are
under
the
same
under
the
org,
well,
I'm,
not
sure,
because
they're
also
called
Kepler
hyphen
things
as
well.
So
it
may
just
be
that
they're
confused
by
our
form
that
only
lets
you
put
one
repo.
B
G
H
A
H
E
B
H
E
C
C
G
D
D
D
D
A
B
F
G
B
G
C
G
B
B
A
B
Fine
cubescape
armo,
we
got
a
response
back
from
the
folks
who
submitted
the
project
and
they
clarified
the
things
that
the
questions
that
we
asked.
They
are
okay,
with
renaming
their
service
and
doing
all
the
other
things
that
we
had
asked
them
to
do
in
our
previous
discussion.
And
you
know
they
just
didn't
want
to
do
it
ahead
of
time.
In
case
you
know,
we
still
say
no,
then
it
would
be.
You
know,
effort
that
didn't
get
any
results
as
such,
but
other
than
that
they're.
B
H
D
K
C
G
F
G
Yeah,
so
it's
the
company
is
a
cncf
member
and
I.
Think
that's
what
they're
doing,
because
they're
currently
saying
the
project
in
company.
That's
coupled
they'd
have
to
remove
that
when
they
decouple
it,
but
the
reason
for
it
is.
If
you
look
at
the
reason
on
the
screen
there,
they
want
to
get
the
community
to
standardize
on
Taylor's,
Linux
and
I.
Don't
think
that's
going
to
happen,
their
motivation
for
being
in
the
cncf
I
think
is.
H
H
B
F
I
That
kind
of
extensibility
principles
are
sure,
I
think
in
their
description.
We
really
need
some
details
about
what
are
the
enhancement
or
technology.
They
actually
introduces
to
make
it.
You
know
the
idea
always
for
kubernetes.
This
is
something
we
can
really
find
it
easily
in
the
whole
application.
That
is
my
my
my
rates
from
their
documentation
right
now,.