►
From YouTube: CNCF Network Service Mesh Meeting 2019-10-08
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
CNCF Network Service Mesh Meeting 2019-10-08
A
A
A
A
Also,
as
always,
if
you
guys,
if
folks
have
things
they
would
like
to
see
added
to
the
agenda,
then
please
feel
free
to
go
ahead
and
add
them,
in
particular
we're
trying
to
track
and
report
on
the
things
that
landed
the
past
week
and
the
things
that
are
in
progress.
So
if
there
are
things
that
are
not
already
in
the
agenda
that
matches
those
criteria,
please
add
them.
There
I
want
to
make
sure
we
give
everybody
calls
out
call-outs
for
the
good
work
they
do
and
that
we
get
some
senses
of
community
of.
C
D
B
Yep,
let's
get
started
so
welcome
to
the
next
network
service,
so
much
meaning.
So
we
have
this
particular
meeting
which
occurs
every
week
at
8
a.m.
Pacific.
We
are
also
involved
in
the
telecom
user
group,
which
occurs
every
first
Monday
at
8
a.m.
and
every
third
Monday
at
4
a.m.
we
are
also
involved
with
the
C
and
C
of
networking
working
group,
which
is
currently
being
rebooted.
So
we
will
have
time
we'll
know
what
time
that
comes
on
pretty
soon
so.
B
A
D
B
So
we
also
have
so
we
also
completed
sorry.
We
also
have
open
source
summit
coming
up
in
Leon
with
a
way
they
talked
accepted
by
Ivana
and
Radha
Slav,
which
is
introduction
of
anism.
So,
if
you
are
going
to
be,
there
definitely
feel
free
to
talk.
Whoever
is
sharing
we're
currently
seeing
your
code,
not.
B
Oh
cool,
sorry,
my
audio
is
doing
a
couple.
Weird
things.
Can
you
still
hear
me
yep
perfect?
So
we
also
have
cube
Khan
and
cloud
native
Khan
where
we
have
NSM
Khan
announced
and
we
have
the
Culver
papers
have
already.
Proposals
have
already
been
closed
and
the
talks
have
already
been
posted
for
both
Cube
Khan
and
NSM
pond
so
feel
free
to
take
a
look
at
the
schedule
and.
B
B
D
A
B
D
B
B
D
B
C
Great
okay,
thank
you.
So
the
last
week
we
gained
a
few
more
followers
14,
followed
about
15,
more
and
posted
tweets
and
retweets
21
and
I
posted
and
there's
links
to
all
the
new
posts
so
close
to
the
pre-registration
for
NSM
con.
Be
sure
to
add
that
to
your
cube,
con
registration
and
the
$50
fee
will
go
to
the
CN
CF
diversity.
C
Scholarship
friend,
the
I've,
also
posted
on
the
webinar
October,
2nd
I
thanked
our
lunch
sponsor
Juniper
Networks,
and
also
sent
out
a
thank
you
to
the
four
different
network
service,
prong
and
network
service
mesh
con
sponsors.
So
this
week,
I
plan
to
promote
the
individual
sessions
and
speakers
that
were
selected
in
the
lineup
for
Ana
Semin
I'll,
promote
the
sessions
at
cube
con
and
sends
more
reminders
to
pre-register
to
NSM
con
and
also
promote
the
sessions
at
open
source
summit
can
Europe
later
on
in
this
month.
C
So,
when
available
I'd
like
to
share
a
link
to
the
telecom
TV
interview
from
ons,
if
there
was
a
video
of
the
5g
panel
from
Linux
Foundation,
networking
I'd
like
to
share
that
as
well.
I
did
retweet
the
ons
keynotes
link
and
the
contributors
podcast
when
that's
available
I'd
like
to
share
that
as
well.
A
A
A
I
believe
we
do,
these
are
great
bugs
for
beginners,
because
they
pretty
much
tell
you
exactly
where
to
look
and
they
tend
to
be
things
like
there
is
an
you
know.
Some
sort
of
unusual
mill
check
that
needs
to
go,
there's
a
mill
trick
that
needs
to
happen
or
something
like
that.
So
these
should
be
very,
very
good.
Beginner
bugs
for
people
to
get
started,
we've
been
sort
of
intentionally
holding
them
open,
so
the
beginners
can
grab
them
or
obviously
we'll
go
and
fix
them
before
the
release.
A
A
So
this
is
what
we're
gonna
focus.
Please
feel
free
to
add
your
stuff
I
not
try
caught
everything
going
by,
but
the
first
one
I
wanted
to
mention
is
we
you've
probably
noticed.
We've
been
some
remain
testability
issues
that
came
out
of
the
blue.
That
didn't
appear
to
be
related
to
any
code
changes
that
we
made
and
they
were
all
happening,
I
believe
in
AWS
and
so
I
think
Artem
has
finally
chased
them
down.
A
We've
got
a
candidate
or
what
may
be
the
roof
having
to
do
with
something
wonkiness
in
an
update
that
AWS
me
to
it
see
and
I.
Do
you
want
to
say
a
few
words
about
that?
Artem.
E
F
A
A
So
effectively
are
done
figure
this
out.
It's
a
very
obscure
kind
of
problem,
he's
really
good
at
those.
We
think
this
is
the
root
cause
and
he's
currently
attempting
a
fix
that
would
downgrade
to
the
previous
releases
he
and
I
it's
a
pretty
good
indicator.
It
was
the
root
cause
that
the
upgrade
of
USC
and
I
happened
at
exactly
the
same
time
that
our
inter-domain
tests
started
having
issues,
but
this.
This
is
why
we
test
on
a
bunch
of
different
platforms
so
that
we
catch
things
like
this,
and
currently
the
tests
have
been
disabled.
A
While
we
work
on
fixing
the
issue
is:
there's
no
point
in
failing
code
changes
that
have
nothing
to
do
with
this
field.
This
CI
failure
cool
any
questions
on
that,
alright,
so
things
that
I
noticed
that
land
it
this
week.
So
do
you
want
to
say
a
little
bit
about
a
Slav
about
the
colonel,
fort
or
support
for
metrics.
G
Yeah
yeah
we
quit
slow
down
in
Qatar
bits
and
but
yeah
against
is
not
merged,
so
it
introduces
metric
support
for
to
come
forward.
So
I
I
used
the
the
model
that
DP
was
using
with
this
source.
Something
and
destination
does
something
it's
not
moral,
so
is
the
same,
and
it
should
yeah
interchangeable,
fabulous.
A
Have
you
talked
to
Matthew,
Rohan
I,
don't
know
whether
or
not
he's
planning
on
doing
any
updates
to
the
skydive
support.
Oh.
A
Might
be
worth
reaching
out
to
him
and
see
if
he'd
be
interested
in
showing
some
of
the
metrics
on
those
links.
It's
the
kind
of
thing
that
he
often
likes
he's
been
very
kind
to
us
on
the
skydive
front,
plus
he
knows
the
skydive
community
people
they
may
even
do
it.
The
skydive
community
has
been
incredibly
generous
with
us.
H
G
D
D
D
A
A
D
F
F
A
Okay,
that's
awesome,
so
no
metrics
are
super.
Important
observability
is
super
important,
so
I'm
delighted
to
see
the
stuff
finally
landing
and
the
good
news
is.
We
should
hopefully
get
a
fix
for
VPP
agent.
That
will
looks
those
metrics
as
well,
so
that
we
can
get
that
working
too,
but
I'm
glad
that
you
guys
continued
moving
all
this
forward,
because
it's
important
to
get
it
done.
I.
A
E
D
A
I
mean
III
different,
some
local
lovers
myself
before
and
after
because
I'm
serious
and
I
have
almost
a
perfectly
optimal
build
environment.
I
got
a
very
fast
laptop
I
have
a
very
fast
internet
connection,
and
if
I
cleared
out
my
daugher
cash
before
inserted
from
a
cold
docker
cache,
it
was
taking
me
15
minutes
to
build
everything.
You
know
make
a
safe
with
these
50
15
I
thought
yeah.
With
with
this
new
patch
that
vastly
improves
the
build
times.
It's
it's
down
to
a
minute
or
less
from
a
cold,
build
it's
just
incredibly
faster
right.
A
B
A
H
A
bunch
of
interesting
improvements
at
did
the
both
ways
DK
and
SM,
and
actually
I'm
almost
finish.
It
is
decay
like
approach
for
MSM,
request
and
close
it's
in
a
pull
request
at
the
moment
and
could
be
look
at
by
you
guys,
so
it
do
the
same
as
we
have
a
decay
for
the
end
points,
but
do
for
the
entire
chain
of
very
questing
and
closing
wear
network
service
requests
so
for
now
open
tracing
shows
pretty
cool,
actually
traces.
You
can
have
a
short
demo
if
I
can
share
my
screen.
H
Okay,
do
you
see
my
screen
yeah,
okay,
nice,
so
this
trace
for
pretty
complex
scenario
when
with
that
the
plane
is
died
on
the
remote
side.
So,
firstly,
we
do
requesting
a
connection,
so
we
can
figure
out
on
any
place
what's
happening
on
by
looking
to
attach
it
with
locks
and
so
on.
After,
for
example,
we
see
with
that
the
plane
down
stuff.
H
We
know
what
is
doing
sending
remote
connection
update
if
the
plane
is
down
and
on
kubernetes
master
node,
we
receive
events
and
when
we
receive
even
when
the
plane
is
died,
we
know
what
to
do
with
a
link
and
do
the
healing
itself
if
all
the
stuff-
and
you
can
figure
out
what
happening
and
mostly
in
real
time
and
if
any
error
will
occur,
it
will
be
market
here
in
up
interesting,
cuz,
the
error.
So
it's
more
obvious
what
to
do.
If
you
always
chain.
A
It's
really
amazing
about
this
is
rather
than
having
the
fish
to
wants
to
figure
out.
What's
going
on.
When
you
have
any
kind
of
a
problem,
you
can
literally
get
down
to
the
individual
sub
part
of
the
code
and
what
it
was
doing
and
where
the
error
originated
through
the
system
and
or
you
can
also
see
what
goes
in
and
goes
out
at
every
step,
plus
any
logs
that
are
happening
at
any
stuff.
Yeah.
H
One
more
interesting
point
here
is:
we
just
need
was
spam
identifier
from
our
unit
container.
So,
for
example,
client
do
they
request
the
first
time
and
any
healing
like,
but
the
point
down,
and
so
on
happening
inside
the
nest.
Man
we'll
be
automatically
added
to
the
first
traces
when
we
requested
a
connection.
So
all
the
lifecycle
with
healing
will
be
automatically
attach
it
here.
A
H
A
H
A
A
D
A
True
I
mean
there's
definitely
offer
upper-upper
ability,
considerations
here
as
well,
because
now,
if
you've
got
some
problem
that
goes
wrong
in
the
system,
you
could
be
exactly
where
wrong.
It
should
hopefully
lead
to
much
more
interesting
bug
reports
as
we
start
getting
those
in
from
the
field
yeah.
D
D
I
A
H
D
A
D
A
D
I,
don't
have
anything
that
Leonard
this
week,
but
I
have
a
question
because
I
would
have
to
probably
drop
in
a
couple
of
minutes,
so
I
just
wanted
to
ask
I
put
it
on
the
bottom
here,
but
I
just
want
to
bring
it
up
with
the
community
in
general.
Cause
I
think
that
it
would
be
a
longer
discussion.
Probably
we
should
you
know
next
time.
Probably
so,
essentially,
I
was
talking
to
some
guys
around
here
and
I
got
a
pretty
complicated
question
that
essentially
boils
down
to.
D
How
do
we
do
it
noisy
neighbors
neighbors,
like
we
don't
have
any
quality
of
service,
no
like
SLA
guarantees
and
what
we
have
been
saying?
Okay,
so
this
is
not.
This
is
a
function.
This
is
what
our
ticket
sitter,
but
essentially
providing
some
guarantees
about
the
bandwidth
or
something
probably
is
not
that
far
from
what
we
want
to
do
with
NSM.
So
I
don't
know,
maybe
maybe
it's
a
long
excursion.
We
will
educate
the
PR
spec
whatever,
but
they've
been
just
in
kind
of
initial
thoughts
from
from
the
community
here.
What
what?
What
everyone
is?
D
A
A
very
initial,
it's
it's
definitely
an
important
question
and
it's
a
very
interesting
question
and-
and
some
of
it
is
I-
think
you've
partitioned
it
correctly
right,
because
there
are
two
places
you
might
stick
costs
related
stuff.
One
of
them
is
in
the
innocent
borders
and
then
the
other
one
would
be
in
the
particular
asses
and
in
fact,
I
think
we
have
a
talk
at
an
assembly
on
where
someone
has
actually
stuck
some
interesting
cost
discovery
stuff
into
a
Tennessee
and
they'll
be
talking
about
their
that
their
so-called.
A
That's
the
question
of
the
four
aside:
I
was
thought
that
the
comet
I
was
making
about
the
NSC
said
I
think
there's
potentially
something
there
for
both
of
them.
The
other
two
comment:
I'll
make
having
dealt
with
a
lot
of
costs
over
the
years,
is
a
really
deep
networking
guy
and
having
dealt
with
a
lot
of
really
low-level
forwarders.
Is
that.
A
Cost
is
almost
never
the
answer.
Sometimes
it
is,
but
it's
almost
never
the
answer,
because
it
turns
out
simple
cost
things
like
police
ORS
are
very
useful
right,
basically
saying
I
won't
let
you
flood
me
out
yeah
complicated
cost
things
like
RSVP
or
you're,
trying
to
reserve
the
bandwidth
across
a
complicated
system.
Those
are
almost
universally
a
bad
idea.
A
You
know,
and
because
the
interesting
thing
about
costs
is
in
a
really
high-performance
forwarder
up
doing
whatever
you're
doing
for
cost
actually
consumes
way
more
resources
than
it
would
take
to
just
service
the
packets
appropriately
anyway
and
to
offer
foreigners,
but
police
sirs
are
potentially
very
interesting.
A
You
know,
police
are
some
shapers,
are
potentially
very
interesting
in
the
system,
so
I
think
there's
definitely
a
really
interesting
conversation
to
be
had
here
around
what
we
might
want
to
do
so
yeah.
That
is
interesting,
but
the
reason
I
sort
of
made
the
initial
comment
of
cost
is
almost
never
the
answer.
Is
it
east
half
the
time
when
someone
taught
wants
to
talk
to
me
about
class?
They
want
to
talk
about
the
complicated
RSVP
version
of
cost.
That's
been
sort
of
a
universal
fail.
B
We
can
also
look
at
some
of
the
building
blocks
as
well.
So
if
we
take
a
look
at
how
an
SM
is
built,
we
we
know
which
streams
come
from,
where
not
only
from
an
individual
endpoint,
but
also
across
a
whole
flow
of
packets,
and
we
also
have
all
of
the
the
monitoring
primitives
there
as
well.
So
it
may
be
possible
to
have
something
that
can
monitor
the
the
overall
quality
of
a
connection
at
a
high
level
from
end
to
end.
B
And
if
we
see
that
there's
issues
with
noise
or
something
similar,
we,
it
should
be
possible
to
get
enough
context
on
any
given
connection
and
any
given
node
that
we
can
then
have
something
try
to
try
to
remediate
and
in
a
variety
of
different
ways
that
the
first
start,
the
first
part,
is
getting
all
the
primitives
there
so
that
we
can
measure
measure
it
without
destroying
the
the
overall
performance
of
the
of
the
system.
And
so.
A
This
is
almost
certainly
going
to
be
more
interesting.
Coming
from
an
NF.
U
perspective
than
an
enterprise
perspective,
you
guys
potentially
could
produce
an
amount
of
traffic
that
would
produce
these
kinds
of
noisy
neighbor
problems
as
they're
flowing
through
a
CNF
there's
just
no
way
that
an
enterprise
app
is
going
to
produce
enough
traffic
to
matter
hey.
A
So,
but
this
this
this
needs
to
be
even
more
interesting
when
we
start
getting
the
hardware
next
stuff
in
because
right
now,
if
you
really
want
to
be
driving
enough
traffic
through
the
box,
to
have
this
be
an
issue,
you
can't
have
the
kernel
in
the
way
right.
It
just
doesn't
work
so,
but
I
think
this
is
a
very
good
question
and
you
know
I'm
trying
to
think
of
like
the
places
you
might
stick
it
in.
D
A
I
mean
it's
sort
of
the
difference
between
the
RSVP
approach
versus
the
the
police
or
some
shapers
that
actually
do
kind
of
work,
because,
if
you
say
I
can
guarantee
you
bandwidth,
that
could
mean
a
number
of
things.
It
could
mean
things
like
I
guarantee
you
bandwidth
through
your
fort
I
guarantee.
You
been
with
out-of-the-box,
you
know
from
from
here
to
the
actual
physical
network.
What
those
people
who
want
to
at
guaranteed
bandwidth
really
want
is
end-to-end
guarantees,
but
that
you're
into
the
world
of
RSVP,
and
we
all
know
how
that
story
ends.
A
So
you
know,
but
it's
an
interesting
it's
an
interesting
problem
to
think
of.
If
we
can
think
of
something
smart,
there
I'm
I'm
totally
open
to
exploring
possible
solutions.
As
long
as
we
don't
go
down,
I
mean
maybe
we'll
even
come
up
with
a
smart
solution
that
gives
you
the
things
you
wanted
from
RSVP,
that
isn't
RSVP
and
that
would
be
glorious.
D
A
B
I
suspect
some
of
these
requirements
come
from
from
some
historic
areas.
Like
you
look
at
something
like
ATM
like
ATM,
you
can
give
you
a
guarantee
that
a
certain
circuit
will
hit
a
certain
level
of
bandwidth
and
when
you
look
at
the
line
of
thinking
over
time,
the
ATM
crowd
shifted
towards
the
MPLS
and
then
eventually
are
shifting
towards
something
like
sr
a
survey,
6
and
so
on,
and
so
some
of
those
thoughts,
I,
think,
are
still
prevailing
from
that
concept
of
I
want
a
dedicated
circuit.
D
I
see
where
this
is
slightly
more
a
telco,
telco,
I
kind
of
wanted
you
and
it's
not
absolutely
valid
today.
What
you're
saying
I
think
this
does
specific
context
when
I
hit
these
conversations,
mostly
about
okay,
I
live
in
the
cloud,
and
you
give
me
a
network
link.
I
want
to
be
sure
that
I'll
be
able
to
send
my
five
packets
where
I
want
them,
not
that
don't
the
bandwidth
will
be
taken
by
my
neighbor
that
actually
he's
doing
some
weird
stuff
there
and
yeah.
A
It's
interesting
is
it's
not
just
the
ATM
guys
anybody
who's
ever
dug
into
taxes.
How
cable
modems
work
like
those
guys
like
they
do
cost
like
you
would
not
believe
if
the
DOCSIS
guys
tell
you
that
you've
been
have
10
bytes,
every
10,
milliseconds,
plus
or
minus
one
millisecond.
That
is
exactly
what
you
will
get.
They
are
not
around
here.
I
have
no
idea,
I
mean
I'm
sure
they
have
applications
where
it
matters,
but
man
are
they
really
serious
about
that?
A
A
A
So
moving
back
to
in
progress,
so
the
API
discussion
stuff
is
continuing.
I
think
what
I'm
probably
going
to
do
just
to
start
moving,
that
ball
forward
is
break
out
just
the
API
piece
as
a
PR,
so
we
get
something
we
can
merge
and
then
do
one
for
the
helpers
and
then
do
one
for
the
sort
of
converters
and
compatibility
layer,
and
then
we
have
the
pieces
in.
So
we
can
start
looking
at
the
strategy
we
discussed
last
week
of
building
of
wrapping
things
and
they
had
various
adapters.
A
So
we
did
the
extensive
discussion
of
the
API
stuff
last
week.
If
folks
want
to
go
back
and
look
into
that,
and
then
you
know
the
slides
are
actually
up
there,
you
can
go,
take
a
look
at
them
as
well
they're
linked
here.
The
next
up
is
for
you,
Andre
the
network,
styled
SDK
style,
refactor
of
network
service
manager,
yeah.
H
I
think
if
they
are
mostly
ready,
so
they
have
working
lost
to
test
fail
it
inside
it,
so
I
hope
you'll
pass
VCI
and
it
could
be
looking
more
carefully.
I
have
made
a
few
attempts
to
make
it
into
smaller
pieces,
but
it
actually
took
you
took
more
more
and
more
time,
so
I
would
I
think
ask
you
guys
to
try
review
it
in
the
current
shape
about
16
changed
files.
H
At
the
moment,
most
of
em
are
pretty
simple,
since
it's
just
a
different
chains
of
operations
like
not
a
plain
programming
or
assigning
a
connection
or
creating
across
connect
and
so
on.
So
most
complicated
changes
related
to
monitors
and
how
it
is
handled,
but
in
general
it
become
more
easy
to
read
them
to
understand,
rather
before
so
I
hope
it
will
go
in
this
way.
Oh.
A
That's
awesome
so
then
excellent,
so
we
we've
got
the
ethernet
contact,
stuff
and
I
think
this
is
still
in
progress.
I
know,
Denise
doesn't
have
his
audio
right
now,
so
he
probably
can't
talk
about
it,
but
he
basically
decided
this
was
easier
to
do
after
the
data
plane
refactor,
and
this
effectively
was
put
in
to
solve
a
problem
we
had
on
he'll,
where,
if
we
healed
you
to
a
different
NSE,
you
might
have
the
ethernet
address
of
the
previous
NSC
in
your
cache,
and
so
you
might
not
immediately
get
service.
A
A
So
sr
v6
support,
that's
still
in
progress.
I
think
Artem
is
working
on
that
it's
currently
blocking
on
a
fix
from
VPP
still,
but
they
now
have
figured
out
what
they
think
the
problem
is.
Is
there
should
be
a
PR
shortly?
I
know
there
are
certain
people
on
this
call
who
are
very
interested
in
the
SR
v6
stuff.
So.
A
A
E
Things
are
going
not
that
fast,
but
progress
and
today
I'm
I,
think
I
got
success
with
helm
charts.
So
we
are
continuing
and
integration
tests
running
okay
and
after
the
whole
CI
thing
will
stabilize
where
I
become
continue
to
to
build
in
a
parallel
as
your
as
your
pipeline,
CI
beside
the
circle
CI
and
then
to
switch
to
a
new
CI.
Something
like
that.
That.
A
Would
be
great,
I
mean
the
Azure
pipelines?
What
are
the
advantages
they
have
is
that
Microsoft
has
very
kindly
donated
enough
capacity
for
us
to
run
our
CI
on
them
for
free,
which
is
quite
quite
kind
of
them,
so
hopefully
we
can
get
switched
over
and-
and
you
know
they,
it
looks
like
quite
a
nice
system.