►
From YouTube: CNCF CNF WG Meeting - 2022-02-28
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).
B
B
Let's
see
so
we
can
get
started.
I
don't
know
if
today
is
a
holiday
for
some
folks,
no,
the
well.
We
got
a
few
more
coming
in
hey
tom,
hi
ben.
B
B
B
We've
heard
that
that
should
be
good
for
some
people.
It
may
not
be
as
good
for
others
half
the
year,
whatever
all
right,
cute
connie,
you,
let's
see
cloud
native
telco
day,
cfps,
are
closed
at
this
point
for
that
and
we
did
get
quite
a
few
sessions
submitted
as
well.
Those
are
in
review.
B
All
right
so
the
tag
network,
cncf
technical
advisory
group
network
they're,
looking
for
speaker
around
air
gap
deployment.
B
B
All
right:
well,
I
was
planning
on
giving
an
update
on
cnf
certification,
covering
some
things
like
categories
and
tests
that
we're
looking
at
for
the
certification
I
kind
of
feel
like
we
should
maybe
wait
until
there's
a
larger
group.
D
Taylor,
I
noticed
that
someone
opened
an
issue
regarding
multiple
interfaces,
so
I
don't
know
if
you
want
to
talk
about
that
or
maybe
we
can
wait
also
for
a
greater
audience
or
you
need
a
number.
I
that
I
have
here.
B
Okay,
so
daniel's
not
on
the
call
today
I
don't
know
his
schedule
if,
if
it's
going
to
be
easier
once
the
daylight
savings
time,
but
I
anticipate
it
probably
will
be
for
him.
I
believe
he's
equivalent
to
pacific
time
in
canada.
B
Anyways
yeah
this
one's
a
good
one,
and
I
see
that
you
posted
some
stuff
on
it.
I
it
seems
like
this
one
might
be
able
to
be
talked
about
in
a
more
generic
fashion.
It's
fine
that
it
says
multi-interface
from
now,
but
I
know
that,
with
your
experience
with
network
service
mesh,
you
know
that
we
could
talk
about
this
in
a
different
way.
The
requirements
that
daniel's
listing,
I
think,
are
still
valid,
even
if
you
don't
use
multiple
interfaces.
B
The
multiple
interface
is
an
implementation
for
a
problem,
a
potential
solution
for
a
problem,
so
traffic
segmentation,
isolation
performance-
you
know-
that's,
maybe
arguably
a
stronger
one,
but
still
there's
potential
ways
of
getting
around
actual
interfaces.
B
Hardware
dependency.
All
right
that
one,
this,
how
we
did
it
peanut
venus,
okay,
yeah!
Well,
these
are,
I
think,
questions.
Is
this
four
right,
so
this
is
good.
B
B
So
one
of
the
things-
and
you
probably
have
recall
this-
have
run
into
the
reason
for
host
deployment
interfaces
and
I'm
going
to
drop
interfaces
and
I'm
going
to
say,
connections.
The
reason
why
you
may
want
to
add
a
new
connection
to
an
existing
pod
is
well
I'll.
Give
one
example,
one
example
would
be
if
you
have
a
set
of
pods
that
are
running
providing
a
service
together.
B
They
work
together
in
concert
to
provide
some
larger
service,
and
then
you
have
one
of
those
pods
die
for
whatever
reason,
and
it
can
come
up
somewhere
else
or
you
may
say
we
need
to
scale
it
up.
That
would
be
a
related
one
and
you
need
to
then
connect
from
one
of
those
pods
or
maybe
multiple
pods
to
the
either
newly
deployed
one
or
a
maybe
you're
scaling
up.
B
So
now
you
need
connections
to
new
pots
that
were
not
there
before,
so
that
the
communication
can
continue
without
having
to
redeploy
all
the
other
ones
for
network
connections.
B
But
I
think
the
problem
comes
whenever
the
the
services
go
beyond
what
the
flat
network
provides,
and
now
you
need
the
additional
ones.
So,
whenever
it's
outside
of
what
would
normally
be
there
is
when
we
run
into
situations
where
you're
either
going
to
do
it
custom
or
you're,
going
to
use
something
like
it
could
be
nsm
or
anything
else
that
can
dynamically
insert
new
connections.
B
Anyways
lots
of
good
things
here,
but
what
do
what
do
we
want
to
do
on
this
right
now.
B
I
think
this
one,
this
topic
needs
some
user
stories,
use
cases
and
user
stories
added.
B
We
need
to
merge
these,
I
think,
under
the
dock
directory,
I'm
going
to
open
up
matrix.
B
B
C
C
B
Can
you
please
add
some
information
into
the
description
of
the
pull
request
sure
either.
C
So
the
general
idea
is
that
I
want
to
create
a
set
of
you
know
best
practices,
documents
around
kubernetes,
core
components,
okay,
and
how
to
deploy
them
securely
and
what
are
the
do's
and
don'ts?
And
you
know
I'm
playing
this
to
be
a
you
know,
a
series
of
of
of
pages
or
or
even
same
pages,
series
of
subjects.
I've
started
with
the
api
server
and
how
to
deploy
api
server
in
a
secure
way.
If
I
haven't,
you
know,
touched
something,
you
think
that
it's
important
you
can
open
for
any
comments
or
know.
C
If
you
want
yourself
to
add
something,
I'm
I'm
fine
with
it.
I'm
going
to
continue
with
the
cubelet
deployment
and
by
the
end
I
want
to
have
you
know
a
set
of
best
practices.
Okay,
for
you
know
for
our
on-prem
deployments,
you
know
the
kind
of
deployments
we
are
working
on,
usually
not
covered
by
others,
because
all
who
are
using
kubernetes
as
a
managed
service.
B
This
looks
really
good
ben,
thank
you
so
for
those
who
haven't
seen
the
best
practice
proposals,
the
idea
here
is
to
provide
a
whole
lot
of
context
and
information
for
anyone.
That's
trying
to
adopt
a
best
practice.
B
You
may
already
think
a
best
practice.
Is
you
understand
it?
You
agree
with
it
you're
ready
to
adopt,
but
you
may
be
talking
with
a
colleague,
that's
not
quite
sure.
So,
ideally,
this
will
be
content
that
are
helping
that,
and
there
may
be
someone
you
know
that
totally
disagrees.
So
we
want
to
be
able
to
communicate
all
that.
So
we
have
context
that
may
go
beyond
what
we
see
in
other
user
stories
be
specific.
B
We
can
reference
user
stories
and
we
want
to
be
able
to
one
one
thing
that
you
want
to
look
at
ben.
That
we'll
want
to
add
into
this.
Is
some
of
the
this
looks
like
it's?
Maybe
the
start
of
the
idea
here
for
the
best
practice,
and
as
we
get
as
you
get
more
feedback,
we
can
iterate
on
it.
What
we'll
end
up
wanting
to
do
eventually
is
have
we
want
to
cover
all
of
these
areas
that
are
marked
as
required,
so
just
go
through
right
here,
so
is
it
implementable?
B
Is
there
do
we
have
a
summary
and
the
motivation
you're
already
providing
a
lot
of
that?
So
implementable,
I
think,
is,
looks
like
it
doesn't
go
through
all
the
details,
but
it
looks
like
those
are
covered.
Do
we
have
a
way
to
test
it?
I
think
we're
already
covered
on
this
specific
one,
but
that's
we
just
want
to
add
that
in
there
scoring
is
probably
something
we
need
to
adjust
off
of
here,
because
that
was
early
early
on
the
other
ones,
though
we
have
some
things
like
non-goals
are
important
to
point
out.
B
If
there's
something
that's
out
of
scope,
we
don't
want
people
to
think
we're
covering
everything
trying
to
cover
everything.
So
this
one.
B
This
is
on
the
non-root
point
out
some
things
here
and
then
just
what
is
the
main
proposal.
So
that's
the
summary
area
user
stories.
We
normally
just
reference
those
this
one
on
the
least
privilege,
we've
been
pointing
to
a
lot
of
the
supply
chain
attack
user
stories,
so
if
there's
existing
user
stories
feel
free
to
reference,
those
we
don't
have
to-
and
this
isn't
just
for
you
ben
this
is
for
anyone.
That's
trying
to
propose
that
would
like
to
help
propose
a
best
practice.
B
If,
if
we
already
have
a
user
story
that
works,
then
you
can
reference
existing
ones.
If
we
need
a
new
one,
then
we
can
work
on
those.
Ideally,
the
user
stories
are
general
enough
that
they
can
work
for
many
best
practices.
Of
course,
if
there's
one,
that's
very
specific,
that's
fine!
You
can
inline,
it
just
put
it
right
in
there
and
then
another
thing
to
point
out
is
when
we're
doing
something:
what
are
the
cons?
B
B
We
do
have
some
discussions
at
the
top
and
if
we
had
a
best
practice
around
network
interfaces
or
connections
or
something
we
would
probably
reference.
The
issue
that
victor
mentioned
earlier
we'd
go
to
the
discussion
and
stuff
like
that,
but
we
also
reference
other
documents
like
the
tag
securities
white
paper.
There's
the
pretty
sure
it's
somewhere
in
here.
B
Well,
I'm
not
seeing
it,
but
the
security
recommendation
that
that
armo
references
for
cubescape
in
a
lot
of
of
the
work
that
y'all
do
so
we
want
to
put
those
in
there
so
that
people
see.
Why
is
this
important?
Why
do
I
need
to
know?
Why
should
I
be
recommending
this?
B
B
Maybe
I
I
went
away
from
it
the
one
that
you're
working
on
been
a
lot
of
people
may
agree,
but
we
want
to
make
sure
and
back
it
up
with
as
much
as
possible,
so
that
if
people
have
contacts
from
wherever
they're
coming
an
understanding
of
why
it's
important,
okay,
okay,
sure,
eventually
what
we'd
like
is
put
put
the
proposal
over
in
this
directory
by
the
way.
B
B
I'm
going
to
kind
of
move
through
some
of
this
try
to
go
a
little
quicker.
Let's
see
who
else
has
maybe
joined
anyone,
so
I
think
everyone
on
the
call
is
at
least
familiar
with
the
notion
of
the
certification
that's
going
to
be
coming
out,
so
this
certification
is
going
to
be
launched
by
kubecon
eu
officially
another
open
source
program
from
lf
cncf
like
the
kubernetes
certification.
B
These
are
the
categories
where
all
of
the
tests
reside
cloud
native
principles
overlap
into
multiple
categories.
These
categories
are
primarily
to
help
us
as
humans
to
just
focus
in
on
one
area.
We
realize
there's
overlap,
but
these
apply
to
the
areas
that
we've
been
hearing
about
for
years
with
problems
and
concerns
life
cycle
onboarding
of
the
cns.
This
networking
applications
reliability
of
the
overall
services.
And
how
do
you
approach
that
in
a
cloud-native
way?
It's
not
that
it's
not
being
done
it's!
How
do
you
do
it
in
kubernetes
in
the
most
effective
way?
B
B
The
test
suite
is
leveraging
a
lot
of
upstream
tools
to
do
this.
So
if,
if
you
want
to
contribute
and
you're
familiar
with
any
of
these,
then
it
helps
cubescape,
including
thanks
ben.
Over
there
we
got
a
lot
of
stuff
around
security
and
other
things,
and
we've
been
getting
contributions
from
all
over
and
I
consider
it's
a
snapshot,
but
we
have
contributions
a
lot
of
contributions.
I
think
that
have
gone
in
that
aren't
captured,
maybe
in
commit,
because
someone
was
collaborating
with
someone
else,
and
I
appreciate
and
would
like
to
encourage
any
type
of
contribution.
B
I
think
everyone's
aware
of
these,
but
we're
here
on
the
working
group
call,
of
course,
but
the
contributor
meeting,
if
you're
interested
in
getting
more
involved
on
the
technical
part
of
the
certification
and
the
test
suite.
We
have
a
call
on
thursdays
at
14,
15,
utc
and
anyone's
welcome
to
join
that
call
and
to
talk
about
how
you're,
using
the
test
suite
or
how
you
want
to
contribute
new
test
updates,
docs
whatever
and
there
are
select
channels
for
any
of
the
initiatives
that
we're
running
there's
going
to
be
a
public
slack
channel
on
cncf.
B
I
already
mentioned
cloud
native
telco
day:
that's
it
so
just
a
quick
overview.
Does
anyone
have
any
questions.
B
B
Sometimes
you're
you're,
muted
and
practicing
no
it'll
be
okay.
I
just
have
to
do
a
re-do
right,
so
we
have
over
50
application
workload
focus
tests
in
the
test.
Suite
we've
narrowed
it
down
to
a
smaller
set
of
tests
that
will
be
used
in
the
first
iteration
of
the
certification.
I'm
sure
there's
going
to
be
multiple
iterations
as
more
tests
come
in
and
coverage,
the
tests
will
be
in
all
the
different
areas.
B
We
have
an
abundance
and
security
and
the
reliability
and
availability
areas
and
we're
continuing
to
add
them
into
all
of
the
areas.
If
anyone
has
ideas
for
tests,
we
would
love
to
get
feedback
on
that
so
specific
things
to
cover,
and
that
could
strictly
be
like
problem
statements
and
we
try
to
match
it
up.
B
B
B
We
also
have
a
document-
that's
maybe
related
to
the
work.
That's
been
happening
in
the
working
group
around
I'll,
say:
supplemental
content
that
helps
users
understand,
so
the
documentation
which
is
slower
to
build
out
because
of
you
know,
we're
doing
a
lot
of
effort
to
make
sure
we
cover
everything
in
the
working
group
will
eventually
provide
a
lot
of
context
and
and
understanding
the
this
document
rationale
is
similar,
but
a
smaller
subset
around.
Why
we're
doing
a
specific
test?
What
is
it
doing,
and
why
are
we
doing
it?
B
B
Well,
we
don't
have
much
documentation
there
yet,
but
this
could
end
up
linking
to
the
best
practices
and
in
the
working
group,
the
documentation
there
and
similar
to
the
one
ben
that
you're
working
on.
I
think
we
have
gonna,
maybe
have
a
harder
time
on
this.
There
we
go
there's
a
few
things
about
apis
and
stuff,
so
there
could
be
a
test
that
we
end
up
here
and
we'll
point
it
to
that
new
best
practice,
documentation.
B
All
right
does
anyone
have
any
thoughts
or
ideas
or
anything
right
now,.
B
If
you're
interested
in
helping,
let
us
know,
definitely
could
use
help
on
the
documentation,
site
and
driving
in
specifically
to
the
tests
that
are
there
right
now,
that'll
be
part
of
the
certification
and
then
ideas
for
new
tests.