►
From YouTube: CNCF Telecom User Group Meeting - 2020-05-18
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
That's
five
after
thank
you
for
your
patience
very
happy
to
have
a
number
of
folks
here
for
our
more
asia-friendly
time
zone,
I'll
point
out
that
it's
7
a.m
here
in
new
york
city-
and
this
is
one
of
the
harder
meetings
for
me
to
do,
because
I'm
going
in
to
get
blood
work
done
in
a
couple
hours,
just
checking
my
cholesterol
and
also
finally
going
to
get
a
the
covid19
serological
tests
to
see
if
I've
had
it
in
the
past
and
have
the
antibodies.
A
A
Well,
while
I
do
it,
but
I
do
want
to
thank
all
of
you
for
being
here
and
particularly
our
friends
of
europe
and
asia,
and
I
don't
have
a
ton
of
updates
since
our
call
a
month
ago,
we
are
getting
ready
to
issue
the
initial
white
paper,
the
prologue
white
paper,
but
I
I
think
the
biggest
thing
going
on
by
far
is
taylor
and
his
team's
work
on
cnf
conformance
and
then
the
integration
that
we're
now.
A
I
think
working
much
more
closely
with
cntt
and
the
other
work
from
elf
networking
so
I'll
hand
it
off
to
taylor.
I
think
the
main
ask
I
would
have
for
folks
here
is
that
we're
just
about
at
the
point
that
we're
ready
for
you
to
start
evaluating
this
cnf
conformance
framework
and
test
suite
against
your
own
internal
cnfs,
that
you
might
have
that
you
might
be
working
on.
A
So
we
would
love
for
vendors
to
look
at
theirs
and
for
operators
who
are
already
have
just
written
some
and
and
have
some
functionality
to
look
at
that,
and
really
the
whole
idea
of
the
cnf
conformance
is
to
have
it
become
a
a
real
open
source
project
where
we
have
multiple
people
involved,
who
are
making
contributions
and
fixing
bugs
and
such,
and
so
that
almost
always
requires
some
bootstrapping,
which
taylor
and
folks
have
been
doing
a
great
deal
job
on,
but
we're
just
about
there
now
on
really
needing
help
to
prove
that
the
the
functionality
of
the
code
is
useful.
B
C
C
B
B
Based
on
that
now,
the
current
snapshot
is
of
of
the
requirements
is
an
earlier
version
of
the
ra2,
so
they're,
not
if,
if
you
look
if
you're
familiar
with
this
spreadsheet,
you'll
see
it's
based
on
earlier
commits.
B
But
what
we're
interested
in
right
now
is
starting
to
get
some
feedback
on
how
the
test
suite
gets
set
up
and
the
initial
usage-
and
this
works
with
this-
should
be
the
similar,
whether
it's
workload
or
platform,
eventually
for
the
type
of
testing
that
we're
gonna
do
so
at
the
moment
that
the
biggest
feedback
would
be.
Are
you
able
to
get
up
and
running
using
the
software
and
test
with
some
type
of
example,
cns
so
the
workload
and
maybe
more
specifically
a
a
component
of
that?
B
B
So
some
of
the
things
that
we're
already
working
towards
is
moving
as
much
as
possible
in
the
initial
setup
into
the
conformance
configuration
files,
which
is
a
yml
for
finding
the
different
parts
of
the
test,
so
that
when
you
get
started,
there's
essentially
a
couple
of
commands
and
you
can
get
going
starting
going
from
here.
I
think
some
of
the
items
that
we're
going
to
look
at
on
platform,
which
we've
been
talking
to
the
cntt
rc
rc
folks,
everyone
that's
working
on
that-
would
be
integration
with
other
tests.
B
We
don't
have
packages
for
specific
os's.
That
might
be
something
coming
down
the
line,
so
these
are
binary
releases.
You
have
steps
for
also
testing
with
a
source.
So
if
there's
some
other
platform,
that's
desired,
then
that's
possible,
but
if,
if
you
want
the
simplest
method
and
you're
on
amd
64,
then
using
those
binary
releases
and
then
stall
guide,
which
I've
posted
into
the
meeting
notes,
is
probably
the
easiest
way
to
go.
C
B
B
Okay
from
reuben
is
there
a
basic
setup
to
just
build
and
run
the
suite
likely?
Yes,
so
I
was
trying
to
bring
that
for
some
reason.
C
B
So
I
can't
bring
it
up,
but
the
install
guide
itself
there's
a
set
of
instructions
that
are
building
everything
from
source
and
then
the
other
instructions
are
using
the.
C
B
We
have
right
now
the
sweetest
running
to
test
most
everything
that
we've
been
we've
had
examples
work
in
kind,
including
recently,
michael
pedersen
at
intel,
I
think
maybe
on
this
call-
has
tested
some
with
multis
inside
of
kind.
So
we
have
a
few
examples
that
use
multis
and
that
looks
like
we're
good
to
go.
B
There,
we've
also
tested
on
packet
on
clusters
that
are
using
machines
there
and
those
are
provisioned
using
the
cnf
test
bed
tool
chain,
which
has
a
part
of
that's
using
cube
spray
for
the
cluster,
but
those
are
on
bare
metal
machines,
so
ruben.
I
think
you
should
be
good
to
test
locally.
I'm
I'm
not
sure
what
your
environment
is,
but
the
requirements
right
now
for
the
the
cluster
that
would
run
the
workload
are
pretty
straightforward.
C
B
B
B
And
then
slack
for
getting
feedback
that
can
go
into
either
issues
or
some
type
of
form
on
specific
things.
But
if
you're
interested
in
getting
started
earlier
on
trying
it
out,
please
go
to
the
probably.
The
easiest
way
is
via
the
github.
Repo
and
you'll
have
a
link
to
the
install
guide
and
either
start
with
the
binary
version
or
the
source.
B
A
A
Update
we
might
give
is
for
some
of
our
asian
colleagues
who
weren't
on
the
call
two
weeks
ago
to
tell
them
about
the
cnt
meetings.
We
could
ask
gurgley
to
do
that
or
you
might
talk
about
it
to
some
degree.
A
Sorry,
my
other
comment
is,
I
think
it
would
be
nice
to
do
a
demo
next
time
of
the
cnf's
passing
your
family
or
such.
B
B
C
B
C
B
B
B
A
lot
of
the
efforts
are
on
the
initial
bootstrapping
and
trying
to
cover
the
hardware
provisioning
phase.
I
think,
on
the
the
side
of
tooling
for
provisioning
kubernetes
itself.
Those
are
there's
some
good
choices
out
there
that
are
are
not,
as
I
guess,
as
big
of
concern
as
self-hosted.
B
So
most
of
the
bootstrapping
for
hardware
provisioning
has
either
been
the
cnf
testbed,
which
is
targeting
packet
and
there's
a
little
bit
of
efforts
on
how
that
can
extend
to
provide
support
for
other
another
environment
or
a
self-hosted
lab,
and
then
their
opnfv
is
trying
to
look
at
how
to
provide
support
via
other
tooling.
I
think
there's
the
project,
maybe
xc
xc1
bill.
You
may
be
on
that
or
gergey
yeah
and.
A
B
At
different
stages,
the
the
once
we
get
up
to
the
actual
certification
level,
then
that's
that's
where
the
conformance
suite
would
come
in
to
provide
the
cloud
native
testing.
B
So
I
think
that
has
to
do
with
integration
for
output,
and
I
don't
I
don't
think,
there's
a
full
decision
there,
but
it's
more
of
discussions
on
how
can
we
add
support
to
test
different
things
that
are
desired,
so
there's
going
to
be
integration
testing.
I
think
the
last
two
weeks
there's
been
discussions
about
benchmarking
and
bill
or
anyone
else
who
may
have
some
more
input
on
that
and
gray.
I
know
I
just
mentioned
xc,
so
if
you
have
that,
then
feel
free
to
speak
up.
C
Yeah
so
axia
is
a
is
a
ci
framework.
Dyno
pnfv
and
the
main
target
of
xci
is,
is
to
be
able
to
pull
different
components
from
different
open
source
projects
and
integrate
and
test
them
together.
C
So
that's
why
it's
it's
planned
to
be
used
in
in
in
rc2
about
the
the
april
event.
I
posted
the
the
schedule
and
the
and
the
materials
link
to
the
chat
of
the
meeting
you
can
reach.
C
You
can
check
all
the
sessions
and
you
can
reach
the
ether
pad
and
and
the
powerpoints
of
each
each
session
and
in
theory
we
should
have
also
the
recordings
up
in
the
table,
but
they
are
not
there
yet.
But
elephant
guys
promised
to
publish
the
recordings
of
the
meetings.
Also,
so
all
of
the
meetings
were
recorded
about
future
events.
There
will
be
an
other
virtual
elephant
developer
and
testing
forum,
which
will
cover
against
cntt
and
and
the
rc2
work
which
will
happen
in
in
june.
C
I
just
don't
have
to
22nd
to
25th
of
june
and
now
it's
time
to
propose
topics.
So
I
just
post
the
link
to
the
event
also
to
the
chat
of
the
meeting,
and
if
you
have
any
topics
what
you
would
like
to
discuss
about
rc2,
for
example,
you
can
put
it
to
the
to
the
list
of
proposed
topics.
C
C
C
C
C
So
borderlands
is
basically
the
result
of
the
of
the
last
six
months
of
of
evolution,
of
all
the
all
the
all
the
documents.
So
it's
like
it's
the
same
release
for
the
reference
model
or
the
reference
architectures,
the
reference
implementations
and
the
reference
conformance
tests.
C
B
There
was
a
pretty
major
re-architecting
of
ra2
and
cleanup
and
re-architecting.
I
guess.
C
C
Yes,
the
main
main
target
of
that
activity
was
to
connect
all
the
requirements
to
the
reference
model,
because
in
in
earlier
versions
of
of
the
reference
lecture,
we
had
some
requirements.
What
we
just
you
know
created
for
reference
architecture
too,
and
now
we
have
like
all
the
requirements
are
derived
from
a
reference
model
architecture.
C
C
So
it's
it.
It
was
music,
basically
a
requirement.
Traceability
improvement.
What
what
we
did
in
the
in
the
last
part
of
the
bold
release.
B
Had
a
chance
to
look
at
the
requirements
spreadsheet,
which
was
the
it's
more
of
let's
say,
conformance
potential
conformance
test
and
cmtp
ra2
requirement.
C
B
So
there's
a
lot
of
te
from
the
platform
side,
there's
a
lot
of
things
that
wouldn't
immediately
get
covered
based
on
if
you're
on
kubernetes
and
you
think
it
could
be
there.
And
so
the
very
first
thing
is
well
kubernetes
conformance
test
and
we
can
map
a
lot
of
that
and
then
beyond
that,
there's
the
kubernetes
e
to
e
tests
which
go
well
beyond
what
conformance
does
you
have
even
more
coverage?
B
So
then
we're
saying
what
is
not
covered
and
there's
going
to
be
some
things
on
rc2,
which
could
be
out
of
scope
for
cloud
native
testing
because
they're
integration?
So
that's
fine!
So
then,
what
are
the
parts
that
we
actually
expect
that
we
would
want
to
test
for
and
be
thinking
about,
I'd
love
to
get
your
earlier
feedback
so
that
we
can
start
working
towards
that
first
set
of
tests
that
we
want
to
start
providing
rc2
and
the
conformance
itself.
B
C
And
should
we
look
it
from
cnf
conformance
point
of
view
or
from
cntt
rc2
point
of
view.
B
Yes,
so
right
now
that
she
is
kind
of
a
blend,
so
it's
it's
was
intended
to
help
the
efforts.
I
guess
in
all
the
work
streams
on
tnt
and
with
the
conformance
suite
efforts,
so
the
the
way
that
you
can
read
it
is
probably
starting
at
the
the.
C
B
There
at
the
top,
so
you
have
the
the
first
two
columns
for
cnf.
A
B
This
is
saying
is
from
the
idea
of
requirements,
and
maybe
what
are
we
going
to
ensure
that
we
have
as
we're
we're
going
for
each
level,
so
we
may
say
on
the
our
the
ri
and
implementation,
we
may
say
we're
going
to
make
sure
and
have
all
mess
requirements
absolutely
in
the
implementation
when
we
start
now
it's
going
to
take
a
while,
but
this
would
be
priority
of
implementation.
B
So
the
idea
on
that
the
cnf
conformance
priority
would
be
a
similar
thing.
But
what
is
in
scope?
So
you
can
kind
of
think
of
those
priorities
as
in
scope
for
implementation
or
in
certification
for
the
cnt
side,
and
then
is
it
in
scope
for
implementation
on
the
cloud
native,
not
whether
or
not
it's
important
for
a
telecom.
C
B
Of
network
interface,
so
that's
a
good
one
so
that
the
the
idea
here
is
support
for
anything
from
having
a
network
interface
that
has
different
bandwidth
characteristics
or
maybe
even
a
specific
interface
type.
There
could
be
all
sorts
of
things,
but
the
idea
from
the
requirement,
which
potentially
is
updated
after
the
change
in
the
ra
2..
But
when
this
was
taken
on
that
commit,
the
idea
was
different
types
that
could
be
offered
from
the
platform
side
for
cnt,
and
this
was
a
must
requirement.
B
C
B
Really
care
about
it,
so
we're
going
to
make
sure
and
have
tests.
So
the
idea
with
this
is
what's
in
scope
and
who
cares?
Who
what's
important
enough
for
someone
to
implement
something?
So
then
someone
maybe
from
opnfv,
says
we're
already.
C
B
Network
interface
testing,
so
we
want
to
go
ahead
and
work
on
this
versus
on
the
cloud
native
side.
We
may
say
well
we're
working
on
the
virtual
compute
testing
and
that's
going
to
be
something
higher
priority
right
now,
so
it's
it's
primarily
to
decide
help
people
to
decide
on
priority
for
implementations
and
who's
can
collaborate
on
something
and
then
the
other
would
be
discussions
about
different
topics.
That
may
not
be
as
clear,
so
one
of
them
was
virtual
networking.
So
this
is
something
that
is
marked
as
a
must
on
both
of
these.
But
it's.
B
If
you
look
at
some
of
the
other
columns,
it's
there's
notes
that
come
from
cnt
to
clarify
some
of
these
there's,
some
cnf
conformance
notes:
that'll
communicate
either
potential
tests
that
we
could
do
or
what
does
this
mean
from
a
platform
implementation
for
cloud
native?
Well,
some
of
these
are
not
going
to
have
kubernetes,
specific
or
built-in
core
pieces.
They
may
have
an
extension
like
cni
that
supports
it.
B
B
I
I
believe
that
there's
some
items
in
here
that
are
going
to
be
gone
completely,
based
on
the
efforts
within
the
ra.
They
may
have
moved
completely
out
or
to
other
chapters,
but
we
found
that
we've
already
made
some
notes
and
they
may
be
a
it's
a
test
that
we
actually
want.
So
we
didn't
want
to
lose
anything
here,
but
we
can
definitely
update
it
to
baldi
and
I'd
I'd
appreciate.
You
know
help
in
doing
that
from
anyone.
B
C
B
A
Okay,
why
don't
we
hold
it?
There
thanks
everybody
for
your
time
today,
we'll
talk
to
you
in
in
two
weeks.
I
guess
the
last
thing
I
was
going
to
mention
is
I
I
was
looking
at
switching
these
meetings
from
being
twice
a
month
to
once
a
month.
I
don't
know
if
there
was
any
feedback
on
that.
I
might
have
them
alternate
once
in
a
year
friendly
time
and
once
on
a
asia
friendly
time,
but
I
I'm
not
sure
we
have
quite
enough
volume
of
material
here.
A
I
think,
as
we
start
having
people
working
on
the
cns
we'd
like
to
to
have
do
that
on
github
and
such
so
any
objections
or
pushback
on
it.
If,
if
things
start
speeding
up-
and
we
feel
like
once
a
month's
not
enough,
we
could
always
bring
back
a
more
frequent
call
again.
A
We're
gonna
alternate
the
two
times
is
my
thought
alternate
other
two
times.
So
could
you
go
ahead
and
is
lucina
on
or
ask
her
to
make
that
change
for
us
where
she
could
cancel
the
call
in
two
weeks
instead
do
a
call
in
in
one
month,
but
at
the
other.
B
A
Taylor,
the
other
taylor
wagoner,
can
do
it
yeah.
Okay,
thanks
everybody
for
your
time
today
we
appreciate
it.
Thank
you.
Bye,
bye,.