►
From YouTube: Network Service Mesh WG Meeting - 2019-01-29
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
A
So,
to
start
off
with
the
view
hack,
if
you
have
not
added
yourself
to
the
attendees
list,
please
do
so.
The
meeting
notes
are
listed
on
both
on
the
github
page
and
there's
also
been
posted
to
the
to
the
chat
window,
so
welcome
to
our
network
service
master
meeting,
and
so,
as
always,
we
begin
with
agenda
bashing.
So
is
there
anything
that
anyone
would
like
to
discuss
that
is
not
on
the
agenda.
I.
A
We
also
have
a
polo
credit
event
at
Q,
Khan
EU.
We
have
the
final
mini
summit
which
will
be
on
May
20th.
The
location
will
be
in
Barcelona,
but
the
final
location
in
Barcelona
is
to
be
determined
and
there
are
no
call
for
papers
yet
listed
as
far
as
I.
As
far
as
I
know,
we
also
have
Mobile
World
Congress
coming
up
at
the
end
of
February
every
25
through
28.
This
tends
to
be
more
demos
on
stances
research,
writer,
centric,
so
the
entity
you
cases
are
more
interesting.
A
If
you
have
a
booth
there
and
are
willing
to
showcase
that
brick
service
match,
let
us
know
and
we'll
see
what
we
can
do
to
help
you
put
together,
something
that's
and
it's
successful.
We
also
had
the
Open
Networking
Summit
ons,
North,
America
and
San
Jose
on
April
3rd
through
5th.
He
call
for
papers
is
also
just
closed
recently,
we'll
see
what
the
schedule
is
and
actually
that
February
5th
looks
wrong
for
notifications,
because
we
need
a
little
more
time
than
that.
So
double
check
the
tape
and
then
we'll
check
the
dates
on
that.
A
A
We
also
have
FOSDEM,
which
is
also
going
to
be
streamed,
live
them
on
February,
2nd
and
3rd,
and
actually
is
that
correct,
sorry,
there's
fog
them
streaming
and
then
there's
just
looks
like
there's
multiple
conferences
listed
and
I'm,
sorry
yeah.
So
we
have
pods
I'm
streaming
live
on
February,
2nd
and
3rd.
We
have
upper
side
conferences,
MPLS
FCM
ffv
in
Paris,
in
April,
9th
through
12th.
The
deadline
was
was
sometime
or
late
last
year,
so
but
be
interesting.
A
A
And
so
we
have
serviced
much
today,
call
for
paper
is
going
to
close
on
February
8th.
This
is
conveniently
located
near
near
May,
so
I
will
definitely
be
putting
on
a
talk
for
that.
We
have
Q
cond
in
cloud
native
called
open
source
summit
in
China
and
Shanghai,
which
will
be
from
June
24
through
26.
So
if
someone
wants
to
travel
to
China
and
bring
the
word
of
metric
service
miss
there,
then
February
15th
is
your
in
your
deadline.
A
A
D
D
Yeah,
that's
it
so
I
suggest
that
okay,
we
don't
go
into
details
here
now
the
PR
is
I,
don't
remember
the
number,
but
it's
there.
The
link
is
there.
So
if
someone
has
any
objections,
ideas,
suggestions
just
put
them
in
the
in
the
pin
the
PR
and
then
we'll
do
merge
it
in
a
couple
of
days.
If,
if
nothing
major
shows
up.
D
Okay,
so
the
other
thing
that
I
wanted
to
quickly
just
kind
of
update,
inform
announce
was
about.
We
had
a
quick
chat
with
it,
and
so
I
have
proposed
this
a
long
time
ago
to
start
having
a
separate
examples.
Repo,
so
I
have
something
going
on
this
PR,
which
I
call
proxy
and
C,
which
is
essentially
HTTP
proxy,
which
accessories
and
I
see
you
on
the
other
side.
D
This
example
in
this
new
repo
and
then
we
can
share
it
and
see
how
it
works
or
it
integrates
with
the
main
airport,
how
the
integration
testing
can
be
done
and
all
these
things.
So
this
is
kind
of
where
what
is
going
to
happen
the
next
day,
I
hope
so,
maybe
by
next
Tuesday.
We
have
this
new
very
point
with
some
initial
feeling
of
how
it
looks
like
to
have
four
separate
examples.
D
For
the
first
initial
version,
it
will
be
just
the
new
examples
that
I'm
I
have
a
new
example
and
I
will
just
put
it
in
in
a
separate
repo
and
see
how
this
works,
and
if
this
goes
well,
we
can
discuss
if
we,
if
we
want
to
move
some
of
the
already
existing
things
there.
We
should
keep
part
of
them
in
the
in
the
current
repo,
but
it
is
I
think
that
for
the
future
are
all
the
new,
the
new
things
that
we
wanted
to
publish
to
go
in.
A
A
But
what
ends
of
what
the
idea
is.
Is
that
once
we've
once
we
have
enough
feedback
here
and
people
are
generally
happy
and
we've
reviewed
and
we
as
a
community
compute
it-
and
this
is
something
that
that
the
core
team
wants
to
incorporate.
We
will
take
this
document
converted
to
mark
down,
and
we
will
then
commit
this
into
a
to
be
determined
directory
within
the
network
service
mesh
project
that
oldest
event.
A
A
B
A
Yeah
so
so
our
hope
is
to
try
to
make
it
as
easy
as
possible
for
people
to
contribute,
and
this
is
also
a
nod
to
the
fact
that
some
contributions
are
are
not
necessary
for
on
the
coding
side,
but
also
on
the
design
side.
So,
for
example,
if
we,
if
we
have
something
that's
related
to
s,
r
io
v,
like
I'm,
pretty
sure
ye
and
would
have
a
lot
to
do
to
contribute
in
that
space.
And
so
then
that
would
allow
somebody
if
and
then
have
does
not
have
the
time
to
implement
such
a
thing.
B
You
know
I
know
you're
super
interested
in
getting
the
SR
iove.
I
started
the
start
on
the
SRB,
the
v
sr
v
6
remote
mechanism
in
and
quite
honestly,
I
know
you're
super
busy,
you're
not
gonna,
have
time
to
code
that
yourself,
but
if
we
can
get
that
worked
up
as
a
spec
that
somebody
can
follow,
who
wants
to
go
write
code?
That
gets
to
be
super
easier
for
people
to
pick
up
and
get
done
and
I
think
the
same
thing
holds
for
some
of
the
stuff.
B
Geoffrey
I
know
you
care
a
lot
about
MPLS
and,
although
I
think
you're
a
little
more
prone
to
write
code,
the
Daniel
is
but
you're
probably
wondering.
Where
do
you
start
right,
and
so
we
can
help
shake
some
of
that
out
as
well.
Does
that
make
sense
to
folks,
especially
the
folks
in
the
community
who
are
looking
to
pick
up
a
shovel
and
write
some
code.
F
Basically,
now
there's
one
thing:
I
wanna:
now
we
are
discussing
about
adding
those
as
our
v6
or
something
like
that.
So
what
we
are
playing
seeing
is
that
we're
gonna
like
say
to
implement
those
like
vnf
functions.
Well,
basically,
we
just
like
to
encapsulate
those
things
and
try
to
route
things
to
so
I'm.
Still
looking
confused
about
that
part,
yeah.
B
So
when
we
say
something,
that's
our
v6
we're
actually
talking
about
a
super
simple,
a
little
stupidly,
simple
thing
in
network
service
mesh
where,
instead
of
using
the
X
land
to
carry
your
your
your
packet
or
frame
from
a
client
to
a
network
service
endpoint,
you
could
use
a
36
or
you
could
use
MPLS
over
Ethernet
UDP
Jiri,
whatever
they
invent
next
right.
So
these
are
all
all
within
the
purview
of
network
service
mesh.
B
Now
there's
been
a
ton
of
interesting
stuff
that
somebody
might
write
into
a
network
service
in
point
4,
a
c
enough
that
does
much
more
confidence
with
MPLS
or
SRB
6.
Then
we're
ever
going
to
do
and
part
of
what
I
think
Frederick
intended
with
this
specification
process
was
say
you
wanted
to
write
a
networks,
just
recent
point
that
did
some
complicated
MPLS
thing.
B
In
that
case,
you
might
have
you
know
you
might
say:
okay
well,
I
mean
I,
know
that
I
want
to
do
this
thing,
and
I
actually
want
to
do
this
thing
over
here
in
my
own
repo,
but
it
would
be
good
to
get
advice
from
the
network
surface
mesh
community
on
what
they
think
might
be
an
optimal
way
to
do
it,
and
this
gives
you
a
venue
in
which
to
go
and
actually
start
the
conversation
and
basically
say
look
I'm
building
some
complicated
SR
v6
network
service,
endpoint
I'd,
like
some
advice
about
how
to
make
that
work
optimally.
A
And
it's
not
required
for
you
to
emit
something
to
this,
if
you're
doing
something
as
a
third
party
endpoint.
So
if
you
don't
feel
comfortable
or
you
don't
want
to
or
you
feel
you
are,
you
already
understand
what's
going
on,
then
you
don't
have
the
post
here.
So
it's
not
a
gate.
It's
something
where
people
can
ask
for
advice
like
yeah
I'm,
a
newcomer
or
I
have
some
complicated
thing
and
I
want
to
make
sure
that
I'm
thinking
of
this
in
the
right
way,
so
I
would
like
feedback.
Here's
my
stuff
yeah.
B
B
Six,
you
know
network
service
endpoint
that
did
some
complicated
routing
on
a
36
headers
right,
so
it
understood
a
lot
about
them
and
I
was
pushing
information
into
it
about
my
physical
network,
because
I
wanted
particular
SRV
six
headers
added
to
my
payload
right.
So
a
packet
comes
out
of
the
client
and
I
want
payload
to
have
a
36
header
is
not
the
thing
that
gets
it
to
the
network
service
endpoint.
In
that
case,
one
way
to
approach
this
would
be
to
write
the
network
service
endpoint
and
the
SDK
is
super
helpful
for
this.
B
By
the
way
and
my
network
service
endpoint
would
have
its
own
VDP
that
processes
the
packets
when
they
get
delivered
and
does
whatever
the
more
complicated
network
behavior
is
your
network
service
mesh
is
only
about
that
were
service.
Mesh
is
managing
virtual
wires
between
things
right,
and
so
you
get
one
in
plugs
into
the
client
you,
the
mother,
in
that
plugs
into
the
network
service
endpoint.
Once
that
packet
arrives
the
network
service
endpoint,
then
you
have
all
kinds
of
interesting
processing.
You
could
do
with
it.
E
E
G
Yeah,
maybe
I
can
explain
it
away,
as
in
the
net
insert
and
I
want
to
burn
a
network
service,
specialist
or
some
other
thing.
We're
talking
v6,
but
one
notion
of
segment.
Routing
v6
is
the
service
programming
aspect,
so
you
create
a
mesh
of
services.
Try
to
address
the
function
you
want
to
create
and.
F
G
Won't
find
them
using
local
authority
v6
addresses.
So
if
you
look
at
this,
that
I
can
create
a
mesh
of
services
using
ipv6
addresses
using
SR
v6,
but
I
still
need
to
program
the
control
to
make
it
happen.
So
we're
where
an
SM
comes
in.
It's
easy
to
say:
I
want
to
create
a
mesh
so
that
the
the
workloads
know
what
to
what
to
request
and
what's
get
and
then
I
can
translate
that
back
into
SR
v6
either
at
endpoints,
so
I
know
are
today
they're
planning
to
make
it
stick
together.
G
B
E
Does
I
just
read
into
the
documentation
and
stuff?
It's
been
a
little
bit
fuzzy
on
me,
like
kind
of
where
the
actual
like
CNF
logic
lives
like
if
it
lives
in
that
end
point
where,
if
the
end
point
is
just
some
dumb
forwarding
plane
that
I'm
stacking
services
behind
and
then
using
an
SM
to
stitch
it
to
it,
and
that
kind
of
clears
it
up.
For
me.
B
E
I
think
I'd
be
good
like
and
I'm
definitely
willing
to
help
with
data.
Just
you
know
have
some
of
you
smarter
people,
kind
of
make
sure
that
my
information
is
correct,
but
I'm
I
would
agree
like
I
yeah
that
gets
great
but
like
when
I
go
there
and
I'm.
You
know
reading
about
Sarah
and
her
storybook
adventures
and
stuff
like
if
you
really
want
to
start
building
this
out
in
your
lab
and
you're
coming
in
from
you
know,
just
ground
zero.
B
F
B
Or
writing
outside
of
a
cluster,
because
network
service
fish
can
actually
interoperate
with
physical
networks
right.
So
you
can
extra
client
to
CNN
for
V
enough
and
then,
but
it
can
also
collect
connect
to
Sienna
for
V
enough
as
part
of
a
chain
to
another
CNF
re
enough,
and
so
you
can
build
out
sort
of
a
cloud
native
policy
based
way.
Super
complicated
graphs
of
CNF
SIF
you
want
a
network
service
mesh
will
dutifully
provide
the
virtual
wires
to
connect
them
point-to-point
to
each
other
and
then
inside
those
CNX
that
you're
building.
B
A
Question
to
erased
so
something
that
I
would
like
to
do
as
well
in
regards
to
the
architecture
is,
if
you
think
about
this
documentation
and
this
room
in
this
room
in
this
way
the
documentation
it's
the
spec
itself
is
also
documentation
and
vice-versa.
So
what
I
would
suggest
is
we
start
to
specify
far
more
formally
specified
what
network
service
mesh
is
through
the
specifications
board
as
well?
A
What
is
the
relationship
between
a
net
and
the
sea
and
a
MSC
and
I
think
if
we
start
off
with
those
and
progressively
work
our
way
up
the
chain
we'll
get
down
to
like
what
does
it
mean
to
be
a
remote
mechanism?
What
does
it
mean
to
be
to
what
it
was?
It
mean
to
be
a
data
plane
and
then,
and
so
on
so
I
think
we
can.
A
We
can
derive
that
documentation
through
this
through
the
same
board
and
the
end
result
is
marked
down,
as
this
is
exactly
the
same
process,
and
now
we
give
there
also
give
us
the
ability
to
to
discuss
like
there's.
This
wording
makes
sense,
so
we
need
to
specify
this
out
more
or
is
this
too
complex
or
so
on
so
well
so
I
think
I
think
this
would
be
a
good
way.
So
would
you
be
willing
to
to
help
participate
in
that
in
that
style
of
discussion?
I.
E
Would
I
would
say
the
only,
and
you
know
other
piece
of
that
Fredrik
is
you
know
both
the
overarching
I
expect
for
this,
and
those
individual
components
right
like
I,
think
when
I
go
to
the
get
right
now
and
I'm
reading
through
things,
I
mean
I
kind
of
piece
it
together.
In
my
mind,
you
know,
reading
through
Sarah's
adventure,
but
I
think
you
know,
especially
when
you
guys
move
away
from
the
technical
community
and
all
these
different
vendors
start
trying
to
product
by
this
productize.
E
You
know
where
it
builds
upon
in
layers
right,
I
mean
once
again,
we
like
the
OSI
model
right,
like
I
mean
when
we
start
talking
about
like
making
this
consumable
by
the
network
people,
because
once
this
gets
more
mature,
you
know
less
and
less
code
focused
people
are
going
to.
You
know,
use
this
and
more
and
more,
you
know
like
network
consumers
are
going
to
take
this
and
they're
gonna
go
and
just
grab
a
network
service
in
point.
E
So
there
needs
to
be
like
very
clear,
like
representation
of
all
the
individual
components
and
then
how
all
those
LEGO
pieces
fit
together
as
far
as
the
documentation
goes
so
wants
really
smart
people,
like
you,
have
like
written
80%
of
the
code
and
it's
like
in
some
usable
format,
some
in
consumer
can
go
and
say:
I'm
gonna
grab
these
LEGO
pieces.
This
is
how
they
fit
together.
Gonna
put
this
into
my
network.
B
A
So
yeah
for
the
documentation
of
specifying
what's
there,
perhaps
what
we
should
do,
then,
is
not
start
off
with
an
individual
component,
but
we
start
off
with
a
high
level.
These
are
the
major
components
and
we
can
we
can
set
them
up.
As,
like
your
overarching
document
that
you
can
say
what
is
what
is
network
service
mesh
I,
think
you
want
to
drill
down,
then
you
can
click
on
the
link
and
then
it'll
say
well.
What
is
that
in
this?
What
is
it
a
network
service
clients
really
do
you?
E
It's
not
something
we
have
to
employ,
but
just
something
like
that
concept.
Right
like
like
what
you
were
just
describing
is
we
basically
need,
like
a
high-level
architecture,
document
right
that
you
can
put
in
front
of
people
to
say
you
know
this
is
what
an
SM
is
like,
like
I
said,
I'm
gonna
keep
picking
on
Sarah,
because
I've
read
her
like
four
times
trying
to
puzzle
this
out,
in
my
mind
and
like
Sarah
is
really
good
at
explaining
to
me.
E
You
know
how
NSM
might
make
my
life
easier,
but
then
like,
when
I'm
ready
to
take
that
next
step.
Like
you
know,
I've
only
just
got
these
very
vaguely
abstract
associations,
in
my
mind,
on
like
what
an
SM
is
versus
what
the
demon
is
versus.
What
the
end
points
are.
You
know
when
you
start
looking
at
CR,
DS
and
kubernetes
like
if
we
don't
have
like
a
sound
explanation
right
off
the
bat
there's
like
lots
of
people
who,
like
little
either
tummy
drops
to
their.
You
know
knees
when
you
you're
CRD.
E
B
Right
this
was
my
attempt
to
try
and
do
that.
Clearly,
it's
not
sufficient
for
purpose
right,
but
you
know
it's
you.
It
gives
you
some
pictures
you
can
start
with
and
then
we
can
hash
this
out
and
get
something
that's
actually
going
to
be
comprehensible
to
more
people
and
and
that's
something
that
is
super
easier.
When
you
come
at
the
problem
for
beginner's
mind.
F
Or
I'm
also
interested
in
that
part,
so
maybe
I
can
help
try
to
a
figure
it
out.
Since
is
this
also
I
mean
during
the
practice
of
learning,
fully
understand
or
what
it
has
and
really
kind
of
grows.
What
has
been
really
played.
H
B
And
it's
gonna
be
a
progressive
thing,
because
the
thing
with
network
service
meshes,
it
is
unbelievably
powerful
right,
and
so
it's
the
kind
of
thing
that
you
wrap
your
head
around
in
layers
right.
You
know
so,
like
layer,
one
is
sort
of
the
Sarah
thing.
How
do
you
make
my
life
easy
right
and
then
layer
two
is
okay.
So
how
do
you
understand
how
the
pieces
fit
together?
B
Because
among
the
things
we
could
do,
network
service
mesh
even
beyond
what
we've
sort
of
specified
for
Sarah
is
network
service
mesh
could
be
used
to
request
network
services
from
physical
networks,
so
it
provides
a
super
easy
way
for
somebody
who's.
Just
writing
a
workload
to
ask
for
something
that
may
be
a
super
complicated
thing
that
happens
in
the
physical
network.
It
can
be
used
to
give
you
a
way
to
get
back
to
Vin's
as
well.
B
It
can
be
used
with
PN
SMS
to
do
all
kinds
of
interesting
hinting
in
your
physical
network
or
even
like
we've
got
a
spec
here
for
create
PSM
right,
we've
sort
of
come
to
realize.
Do
you
guys
remember
the
create
verb
that
we
talked
about
in
some
of
the
talks
in
passing
I.
B
Do
yeah
imagine
yeah,
imagine
being
able
to
have
a
network
service
endpoint
created
on
demand
at
the
scope
that
you
want
it?
Maybe
you
would
prefer
to
be
on
the
same
node
as
your
client,
the
the
PN
assemblance.
You
do
that
in
a
very
simple
way,
and
we
have
done
just
a
terrible
job
of
documenting
the
space
of
possibilities
and
they're
super
exciting.
C
Pay
me
I
can
also
pitch
and
also
am
preparing
few
slides
for
the
container
world.
Talk
where,
in
my
intent,
is
essentially
to
sort
of
give
or
no
one,
so
that
if
people
are
not
familiar
with
service
measure
network
service
mesh
and
how
all
of
it
plays
together,
so
I
can
probably
work
with
Jeffrey
to
define
it.
In
fact,
when
we
started,
we
in
fact
created
a
use
case
document
just
to
showcase
how
networks
of
this
mesh
would
play
out
with
the
use
case.
A
D
So
I
will
propose
at
this
point
that
we
define
start
kind
of
converging
into
at
least
the
first
deadline
or
first
milestone
so
I'm
not
sure
what
would
be
the
right
format
to
actually
fix
this,
but
we
should
figure
out
the
milestone
name,
and
we
had
some
comments
here
about
same
verse,
same
things
like
that,
at
least
the
proposal
that
that
is
here
and
whoever
we
seem
to
agree
that
it
would
be
announced
that
keep
coming.
You
click
on
you.
D
I
mean
somehow
to
demonstrate
that
this
is
a
project
even
with
the
very
basic
feature
that
we
can
offer
in
the
beginning.
It's
something
that
is
a
reliable
something
that
can
play
well
in
a
larger
environment,
and
you
know
it
started
also
working
calm.
This
proof
of
concept
for
Google
Cloud,
so
things
along
those
lines.
B
B
You
know
meaning
and
when
I
say
resiliency,
what
I
mean
is
I
had
a
slide
where
I
was
in
the
in
one
of
my
comics
for
the
whole
slide
was
just
pods
die
because
they
do
write.
The
entire
designing
cloud
native
is
that
it
is
that's
where's,
the
pod
side,
pods,
restart,
right
and
I'm
being
to
show
that
we
can
kill
off
various
pods
in
the
system
and
stuff
keeps
working
with
at
most
a
small
blip
in
network
connectivity
right.
B
If
you
tell
off
the
data
playing,
you
will
get
a
small
network
connectivity
blip
right,
but
if
you
kill
off
other
components,
then
you
shouldn't
get
a
network
connectivity
blip,
but
you
should
be
able
to
have
the
system
recover
to
the
correct
sort
of
state
and
move
forward.
So
I.
That
seems
like
the
most
important
thing
in
my
mind
and
then
after
that
there's
a
lot
of
other
cool
stuff
that
we
could
do
and
part
of
that
is
going
to
depend
on
people's
interest.
But
I
would
almost
say
after
resiliency.
B
D
B
And
say
it
would
be
to
put
together
a
project
board
with
things
that
are
on
the
critic
after
that
release,
because
I
don't
want
to
give
the
option
that
we
can't
do
other
things
because
we
can.
But
it
also
gives
just
to
sort
of
capture
the
things
that
are
most
important
for
getting
to
that.
The
things
that
have
to
happen.
D
Yeah,
it's
also
probably
worth
this
cuz
think
if
we
want
to
have
a
release,
branch
or
everything
is
done
in
the
master
branch
I
mean
you
know,
this
is
some
kind
of
different
release
strategies,
so
the
the
the
tick.
So
the
question
is:
are
we
fixing
this
within
this
call?
I
mean
this
or
next
week
or
how?
How
are
we
cover?
We
come
to
proceed.
B
B
I
think
your
proposal
of
a
0.10
release,
ember,
is
probably
quite
reasonable.
I
think
we
have
an
open
question
that
I'd
encourage
folks
to
brainstorm
on
sort
of
a
code
name
for
the
release.
I
think
we
have
a
rough
notion
of
sort
of
stability
resiliency,
followed
by
usability
as
being
sort
of
the
primary
critical
path
objectives
for
that
release
and
I.
Think
we've
agreed
that
we
want
to
do
a
project
board
to
track
the
critical
path
for
that
release.
A
Matches
that
much
is
my
view,
so
we
we
focus
on
like
what
are
the
primary
themes
for
the
for
the
release,
and
that
doesn't
mean
that
the
community's
locked
out,
if
there's
something,
that's
very
important
to
you,
that
you
want
in
and
are
willing
to
work
towards
it.
We
won't
say
that's
not
part
of
the
that's
part
of
the
release
as
long
as
it's
as
long
as
the
architecture
is
been
accepted
by
the
community
and
by
the
core
developing
development
team.
A
B
B
Is
that
you,
you
basically
do
all
your
work
on
master
up
until
a
point
where
you
pull
a
single
throttle
branch
and
the
point
of
the
throttle
branch
is
just
hardening
and
bug
fixing,
and
so
and
then
you
you
basically
do
the
rest
of
the
work
for
that
release
for
hurting
and
bug
fixing
on
that
throttle
branch,
and
this
this
has
the
benefit
that
it
always
takes.
Master
koban,
because
you
know
you've
pulled
your
throttle
branch,
so
somebody
wants
to
do
something
a
little
risky.
B
D
B
Metrics-
and
we
have
a
few
on
the
call-
which
is
also
good
because
I
think
he's
the
one
who's
possibly
most
excited
about
them
in
the
immediate
term,
but
this
really
comes
down
shoes
right
now.
We
have
this
monitor,
connections
and
monitor
across
connect,
calls
that
we
make,
and
all
that
really
do
is
basically
say,
give
you
information
about
the
states
of
connections
and
cross
connections,
and
this
is
how
this
guy
died
of
is
building
out
the
topology.
B
But
it's
super
useful
to
have
metrics,
and
so
one
of
the
things
that
I've
been
bouncing
around
and
chatting
a
little
bit
with
you
on
the
board
about
is
sort
of
what
metrics
do
we
want
to
report
and
network
service
mash,
and
my
initial
thought
for
this
was
interface
stats,
because
from
that
you
can
derive
a
lot
of
stuff
right
from
if
you're
getting
the
interface
stats
updates
periodically,
you
can
use
them
to
derive
information
about
yo,
throughput
and
all
kinds
of
other
things.
B
I
I
I
B
Okay,
cool
so
I've
added
that
to
the
spec
board.
If
you
could
pick
that
up
and
start
an
issue
in
a
Google
Doc,
so
you
can
get
that
hashed
out.
Hopefully
we
have
that
hashed
out.
We
can
talk
about
it
next
week,
know
if
not
before,
and
then
we
can
start
executing
on
getting
you
metrics
I,
like
the
latency
metrics
idea.
A
lot.
One
of
the
things
I
think
we
may
have
to
do
is
do
this
in
two
steps,
because
interface
stats
are
relatively
easy
from
latency.
We
would
need
something
like
IO
a.m.
B
I
B
Or
there
may
be
something
simpler.
Actually
we
could
do
that.
Io
am
even
it's
really
a
matter
of
you.
You
know
a
lot
about
the
kinds
of
technology,
your
network,
I.
You
know
these
kinds
of
technologies,
so
I'm
super
open
to
sort
of
things
we
might
do
to
readmission
latency
across
these
different
l2
l3
connections.
B
A
Well,
so
we
have
a
few
more
minutes:
let's
go
where
the
envoy
spec.
This
is
the
first
draft
and
it's
covering
a
proof
of
concept
for
envoy,
not
the
hardened
version
that
we
want
to
settle
on
and
so
the
proof
of
concept.
So
for
those
of
you
who
are
unfamiliar
with
envoy,
it's
a
open
source
edge
and
service
proxy.
So
basically
it's
a
proxy
that
you
then
can
apply
a
configuration
to,
and
then
it
does
whatever
that
configuration
is.
A
A
So
what
we
so
on
delay
is
the
thing
that
actually
ends
up
driving
interesting.
The
the
interesting
networking
part
of
projects
like
sto
and
ambassador
and
so
weird
there's
a
series
of
different
features
that
you
can
perform
inside
of
it
such
as
course,
filtering
fault
injection
you
can
bridge
the
RPC
or
you
can
go
right,
Lua
bridges
or
so
on
or
Lua
scripting.
Rather
so,
there's
a
there's
a
quite
significant
amount
of
interesting
things
that
you
can
do
within
envoy
to
implement
something.
That's
of
interest
to
you.
It's
also
incredibly
fast.
A
It
also
supports
things
like
live,
live
upgrade
of
the
binary
without
dropping
connections.
So
we
so
basically
on
the
way,
though,
falls
straight
into
a
into
being
a
network
service,
so
you
can
ask
for
on
delay
as
a
network
service
and
or
at
least
that's
that's
what
the
world
we
believe
the
world
should
should
look
like
and
so
so
you've
application.
She
could
request
for
an
application
proxy
of
its
type.
Then
it
wants,
and
so
the
initial
version
should
be
entirely
written
with
Carrell
interfaces.
A
So
basically,
you
have
a
kernel
interface
coming
into
envoy
and
then
the
curl
interface
to
get
back
out.
Actually,
in
this
scenario,
we
might
even
just
use
a
kubernetes
Network,
Inc
and
I've
even
bothered
with
the
network
service
client
at
this
point
to
get
back
Adam
on
like
so
so.
In
this
scenario,
there's
a
there's
a
couple
things
that
we'd
have
to
add
in
here
as
well.
A
So
first
one
is
creepy
the
playable
on
the
way
deployment,
so
the
pods
don't
need
to
be
created
on
the
fly,
so,
in
other
words,
assume
that
we
have
an
MSc
that
is
being
created.
We
have
the
create
PN
SM
later
on.
That
will
help
with
playing
this
on
the
fly
at
a
at
a
pod
level
or
node
level.
We're
gonna
run
the
POTUS
privilege
to
start
off
with
and
the
long
run.
A
We
don't
want
it
running
ass
privilege,
but
in
the
short
term
we
need
to
add
in
some
IP
tables,
which
will
require
some
privilege
and
what
the
IP
tables
does.
Very
specifically
is
an
injection.
What's
called
a
redirect
role
into
Ivy
tables,
so
a
redirect
role,
those
two
things:
first:
is
it
redirects
all
incoming
traffic
to
your
localhost
to
a
specific
port?
So
so
all
the
traffic
coming
in
from
one
port
is
now
redirected
to
on
void
and
then
Envoy
then
uses
the
I/o
puddle.
A
One
of
the
aisle
cuttle
commands
to
work
out
for
each
tcp/ip
stream.
Whoa
is
the
original
destination
so
that
it
can
then
do
something
intelligent
with
with
bad
data.
That's
been
lost
from
the
string,
so
once
that's
correlated
and
then
all
the
way
you
can
then
do
whatever
it
needs
to
do
in
order
to
in
order
to
work
so
the
initial
version,
the
pod
will
meet
taxon
privilege,
ideally
we'll,
be
able
to
extract
this
privilege
and
remove
it
and
have
something
else
on
it.
A
Technically,
you
do
l2
as
well,
but
for
the
approval
pots
that
will
focus
on
the
l3
so
once,
and
so
once
we
have
a
network
service
endpoint
created,
then
we
should
verify
that
that
that
that
works,
then
we
need
to
create
a
new
network
service
endpoint.
We
also
need
to
create
a
new
Android
network
service,
not
endpoint
pot
image.
A
So
what
this
means
is
you
take
the
original
Envoy
which
what's
been
released
by
Envoy
and
we
add
a
layer
on
top
of
that
which
injects
the
network
service,
endpoint
binary,
and
so
that
way,
when
you
run
it,
it
doesn't
run
anyway
directly
and
say
the
rims
are
network
service
endpoint,
which
then
it's
then
acts
as
the
Emmitt
for
or
envoy
so
it'll.
So
the
own
network
service
endpoint
will
will
both
receive
requests
coming
in
and
simultaneously
control
the
on
way
process.
A
I'll
I
will
expect
that
out
more
I
was
in
the
middle
of
writing
this
and
didn't
have
time
to
finish
that,
and
the
last
part
was
ignoring
the
network
service
client
side.
So
how
do
packets
get
out?
And
it's
in
this
case
we
just
use
a
default
kubernetes
networking
for
me
to
get
back
out
of
the
initial
proof
of
concept.
In
the
long
run
we
want
to
actually
have
a
network
service,
client
and
there
as
well.
A
They
can
make
the
type
of
request
that
we
want
out
and
fire
them
together
or
we
make
use
of
the
network
service
wiring
that
that
was
coming
down
the
road
later
on.
I
will
add
an
image
on
here.
That's
it's
unclear
has
to
like
some
of
the
ideas.
It
makes
a
lot
more
clear,
but
I
have
an
image,
so
I
won.
D
So
we're
running
out
of
time
and
I
will
add
my
complimentary
here
for
the
last
part,
but
essentially
we
dedicate
it
should
be
pretty
easy
to
just
start
the
connection
down
downstream
or
upstream.
Ok,
but
probably
the
proper
proper
way.
As
you
said,
a
voice,
you
don't
need
to
restarting
voice
updated
configuration.
So
just
you
can
just
update
the
actual
configuration
of
my
with
the
new
IPS
that
you
get
from
the
NSC.
B
B
So
first
the
SEO
community
is
talking
about
how
would
they
move
envoy
from
being
a
sidecar
container
to
being
a
sort
of
a
separate
pod
they're,
currently
calling
it
a
metapod?
The
nice
thing
with
the
reservists
meshes
we're
getting
we're
moving
in
a
direction
where
we
could
do
this
very
cleanly.
The
second
thing
is
your
employee
basically
acts
as
an
ingress
to
your
application
service
mesh.
You
could
imagine
a
situation
where
you
have
a
pod
and
it's
got
the
application
service
mesh
for
the
career
days
cluster.
It
there's
a
separate
application
service
mesh.
D
A
Yeah
the
same
so
again,
thank
you.
Everyone
for
attending
will
be
will
be
available
on
IRC.
Next
meeting
is
same
time
next
week
on
Tuesday.
Definitely
looking
forward
to
continuing
this
conversation
on
on
the
way
where
we'll
have
it
more
spectat
at
the
time,
and
with
that,
let's
close
out
and
you
all
have
a
good
day,
Thanks.