►
From YouTube: Layer5 Community Meeting (Jan 29th, 2021)
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
While
we're
hanging
out
for
a
moment,
I'm
gonna,
oh
shreedhi,
already
has
shruti
just
I'll
put
a
link
to
today's
meeting
minutes
in
the
zoom
chat,
so
we'll
we'll
be
going
over
those
actually
doo
doo.
Since
I'm
speaking
with
you
do
you,
we
want
to
make
sure
that
you
have
access
to
the
meeting
minutes,
so
you
can
take
notes
or
put
topics
in
there
as
well.
So
if
you
don't
have
access
just
just
say
it
so.
B
B
A
C
C
All
right
lee:
do
you
want
to
stop
or
no.
C
Welcome
everybody
look
at
the
f5
community
call.
This
is
the
call
of
the
week
if
you're
here,
value
needed.
C
All
right,
let's
begin
right
away.
C
Before
we
start
out
with
any
of
the
big
reminders
or
the
big
notifications
welcome
to
the
community,
do
give
us
a
short
intro.
I
know
you've
given
many
short
intros
this
piece
on
a
couple
of
calls,
but
do
go
ahead.
B
Thanks
srithi,
hello,
everyone
I
I'm
from
india
and
I'm
currently
in
my
junior
year,
I
like
front-end
development
and
I'm
exploring
graphql
right
now.
I
also
like
photography
and
football
a
lot.
A
The
pigskin
football
right,
not
that
not
that
soccer
stuff.
C
Moving
on
the
next
reminder
can
actually
be
bunched
up
with
the
first
topic
in
the
agenda
today
and
nichola
might
actually
be
a
better
person
to
introduce
it
little
do
you
want.
C
A
There's,
while
we're
waiting
on
the
kill
actually
is,
has
dudu
introduced
before
this
might
be
her
first
time
on
the
call.
D
Yeah,
I
it's
my
first
time,
I'm
doodoo,
I'm
working
at
sky
scanner
and
I
actually
worked
in
bluetooth
mesh
previously
in
iot
research
and
now
we
have
cell
structure
also
and
we
have
service
mesh
with
istio.
D
C
A
E
C
E
E
E
It's
built
upon
gatsby
and
graphql
for
data
manipulation
purposes
and
the
home
page
is
the
most
is
the
only
page
which
I
guess
is
not
ready
as
of
now,
but
almost
all
of
the
other
pages
are
ready.
You
can
see.
There
is
a
bug
right
now,
which
was
spotted
a
few
days
back
by
one
of
our
contributors,
but
it's
it
needs
some
love
from
us
to
fix
it
and
give
it
a
tweak
like
we
have
a
lot
of
new
things.
The
designs
are
ready.
B
A
Nice,
oh
very
good,
so
nikhil
is
there
a
call
to
action,
then
for
for
everyone
here.
E
The
call
to
action
which
basically
is
to
go
through
this
site,
go
to
this
link
and
take
go
through
all
the
pages
and
see
if
everything
looks
fine
in
all
of
your
laptops
or
desktops
you
are
using
and
if,
if
at
all,
you
find
any
issues,
hopefully
you
won't,
because
we
have
very
great
contributors
who
are
working
their
night
to
fix
all
the
issues
so
yeah.
E
Please
do
don't
report
any
issues
for
the
home
page
as
of
now,
because
that's
still
in
progress
other
than
that
all
of
them
have
been
complete,
so
yeah,
that's
the
first
item
and
other
than
that
we
are
open
for
contributions
from
all
of
you.
If
you're
inclined
to
do
something.
A
I
think
it's
without
question
that
this
new
site
is
will
represent
the
works
of
the
community,
much
much
better
and
sets
us
up
on
what
we're
shifting
off
of
jekyll
for
layer
5io
to
gatsby,
which
is
react
based.
Hopefully
this
sets
us
up
for
many.
You
know
to
be
able
to
support
many
many
many
more
pages,
and
by
pages
I
mean
things
like
well,
I
don't
know
if
ruth
is
on.
I
don't
know
that
she
is,
but
ruth
one
of
the
mesh
mates
in
the
community.
A
She
was
just
suggesting
creating
well
expanding
on
the
layer,
five
community
handbook
like
the
contributor
handbook
or
sort
of
creating
one,
because
we've
got
a
lot
of
introductory
guides
for
welcoming
newcomers
into
the
community,
welcoming
new
contributors,
new
users,
helping
folks
get
involved
in
the
various
projects
that
are
going
on.
All
of
you
are
familiar
with
mesherie
as
our
as
the
largest
project,
but
there
are
well.
A
I
have
to
go
count,
there's
a
number
of
other
projects
that
are
also
active,
get
nighthawk
service,
mesh
performance,
s
p,
the
I
shouldn't
have
started
practicing
them
off
because
there's
well
one
that
was
looked
at
yesterday,
tentatively
called
meshworks
there's
mesh
map,
which
is
a
visual
topology,
we're
using
cytoscape
as
a
javascript
framework.
A
There
there's
been
some
exploration
from
dhruv
patel
who's
on
the
call
around
the
ability
to
drag
and
drop,
bring
in
the
drag,
drag,
draggable
components
and
kind
of
a
drag
and
drop
user
experience
into
mesherie
itself,
and
in
the
time
that
drove
that
you've
been
exploring.
There
was
a
new,
drag
and
drop
framework
or
tool.
Kit,
that's
react.
A
Based
that's
been
released,
it
looks
pretty
slick,
and
so
I
know
that
you're
well
speaking
of
projects,
I
know
that
you're
focused
on
service
mesh
interface
conformance,
but
after
that
project,
maybe
there's
some
draggable
things
to
go
explore
again.
I
don't
know
so
anyway.
Point
is:
there's
a
lot
of
projects
going
on.
A
I'm
thankful
that
all
of
you
have
given
time
to
this
next
generation
version
of
the
site,
which
is
really
important
because
it
it's
the
one
of
the
first
places
where
someone
would
come
to
evaluate
whether
or
not
they
want
to
use
a
project.
They'll
take
a
look
at
the
site
and
see
if,
if
it
inspires
confidence
that
these
are
real
real
things
that
there
are,
you
know
all
of
you
behind
it
and
supporting
it.
A
A
Nikhil
thanks
for
this,
this
is
good.
Oh,
the
the
other
thing,
too,
is
there's.
If
you're
familiar
with
figma,
I
suspect,
half
of
us
are
and
half
of
us
aren't
figma
is
a
user
experience
like
a
wireframing
tool,
a
ux
tool
for
doing
graphic
design,
not
not
graphic
design
exactly
but
website
design,
mobile
design,
there's
a
there's,
an
open,
figma
project
that
has
all
the
designs
for
this
next
generation
version
of
the
site,
and
so
no
doubt
nikhil
will
share
that
in
the
chat.
A
C
All
right,
thank
you
to
the
mesh
mirror
for
the
really
good
intro.
G
F
Yeah
so
I'll
be
giving
a
brief
overview
on
the
progress
of
miss
map,
so.
A
As
you
can
see,
oh
sorry,
thank
you
very
much
for
this
video.
I
didn't
mean
to
interrupt.
I
figured
there's
a
number
of
people
here
that
I
that
probably
haven't
heard
of
mesh
map
as
a
project
before,
and
so
I
just
in
case
you
weren't,
going
to
give
an
overview.
I
I
encourage
you
to
give
it
an
overview.
F
So
mesh
map
is
a
visual
topology
for
your
services
and
I
mean
service
measures.
So
you
can.
You
can
monitor
all
these
services
and
all
that
through
this
yeah,
so
it
offers
you
many
to
what
you
can
say.
Many
services
like
you
can
invoke
a
performance
test
through
this
and
all
that
and
you
can
analyze
it
draw
our
traffic
through
a
view,
graphene
charts
and
all
that.
A
There's
a
long
road
map
for
this
visual
topology,
this
project
called
mesh
map.
The
javascript
framework
that's
being
used
is
is
cytoscape
right
now.
The
the
focus
of
feature
development
for
it
is
is
from
a
with
a
view
or
a
redone.
Well,
not
a
read.
Only
it's
it's
about
visualizing
and
better
understanding.
A
What's
going
on
with
your
services
on
the
mesh
and
and
eventually
there
will
be
probably
something
of
a
designer
interface
such
that
you
can
well,
we
were
just
talking
about,
drag
and
drop
a
second
ago,
but
a
designer
mode
that
you
could
switch
into
visually
design,
some
changes
or
or
effect
changes
in
the
configuration
of
the
mesh,
maybe
onboard
services
or
remove
a
side
card
proxy
or
you're,
seeing
down
below
there's
a
a
terminal
which
is
intended
to
for
the
most
part
it's
intended
to
show
logs.
A
A
Beneath
if
you
go
back
into
the
details,
pane
on
the
right
hand,
side
one
one
of
the
things
between
the
the
details
pane
and
the
actions
that
people
are
able
that
a
user
is
able
to
invoke
the
some
of
the
actions
that
you're
seeing
now
have
to
do
with
managing
the
performance
of
of
your
services
of
your
mesh.
A
One
of
them
has
to
deal
with
well
service,
mesh
interface,
conformance
or
smi
conformance,
and
some
of
you
are
aware
that
mesherie
is
the
tool
that
assess
is
the
tool
that
the
smi
project
uses
to
assess
whether
or
not
every
participating
service
mesh
is
conformant
to
smi
specifications.
A
So
you
as
a
user,
you
can
come
here
and
invoke
that
test
for
your
mesh.
If
you're
running
multiple
meshes,
there's
consideration
in
the
design
of
mesh
map
to
show
you,
you
know
your
different
meshes
to
show
you
your
different
workloads
running
on
those
meshes,
the
ability
to
sort
of
zoom
in
and
out
and
beneath.
I
don't
know
if,
if
that's
something
you're
able
to
demo
right
now,
the
ability
to
zoom
to
different
levels.
F
Actually
that
I
removed
that
code
for
now,
okay,
actually
yeah,
so
okay,
it
makes
sense.
So
you
know
we
kind
of.
I
think
I
changed.
A
Who's
kept
watching,
veneth
move
through
here,
something
for
people
to
think
about
a
little
bit
is
well
well,
it's
actually
happening
right
here,
so
venetha
is
using
his
scroll
wheel
on
his
mouse
to
zoom
in
and
out
to
you
know,
these
are
just
vector
graphics,
so
they
scale
well
at
some
point,
though,
the
the
behavior
that
beneath
and
the
others
that
are
working
on
this,
the
behavior
that
they're
they're
implementing-
is
that
at
some
point,
if
you
scroll
out
far
enough
as
a
user
or
if
you
scroll
in
far
enough
that
you
will
the
your
view,
will
change
that
and
you're
seeing
that
happening
a
little
bit
as
vanity
scrolls
all
the
way
out.
A
At
some
point,
you
just
can't
see
all
the
details
like
it's
too
small
to
make
out
so
at
some
point
it
will
automatically
switch
view
from
like
a
detailed
service
level
view
to
more
of
a
service
mesh
view.
Overall,
if
you
keep
scrolling
out
to
maybe
like
a
multi-service
mesh
view,
if
you
keep
scrolling
out,
maybe
a
like
a
multi-cluster
view
or
whatever
those
there's,
a
few
different
views
that
are
kind
of
planned
for,
but
that's
part
of
the
interaction,
the
the
ability
to
change
these
layouts
is
I'm
assuming
beneath.
A
Yeah,
so
these
rearrange
the
structure
of
the
how
the
nodes
and
edges
are
interacting,
so
so
pretty
early
in
this
project,
I
don't
know
I
think
soon,
if
you're
subscribed
to
the
measuri
edge
release,
soon
you'll
probably
be
able
to
get
begin
to
play
with
sort
of
beta
releases
of
measurement
map,
and
all
of
you
are
welcome
to
engage
and
contribute
if
you
want
or
just
or
provide
feedback
or
just
use
it
or
or
not,
use
it.
Maybe
it's
junk,
but
we'll
see
one
of
the
things.
A
G
A
Is
probably
back
now
he
had
in
one
of
his
initial
prototypes
brought
demonstrated
the
ability
to
have
a
context
menu
which
is
pretty
cool,
and
so
that's
that's
what
you're
seeing
right
now.
So
if
you,
what
is
it
you
right?
Click
on
a
node
and
you'll
get.
A
A
So
yeah
a
lot
of
a
lot
of
possibilities.
That
is,
I
think,
I
think
it
was
beneath
that
was
able
to
give
an
update
on
mesh
map,
maybe
a
month
or
six
weeks
or
so
ago,
and
so
it's
nice
to
bring
it
back
out
and
get
some
feedback
from
folks
yeah.
A
But
I
think
one
of
the
things
we
might
be
seeing
is
the
so
if
you've
got
grafana
or
prometheus
deployed
in
your
service
mesh
environment,
mastery
will
discover
that
those
instances
allow
you
to
connect
to
those
instances
and
beneath
I
think
you
are
showing.
If
you,
if
measuring,
is
connected
to
grafana,
for
example,
it
will
pull
in
it
will
allow
you
to
choose
what
grafana
boards
you
want
to
display
and
you'll
see.
Metrics.
F
So
it's
not
fully
featured
right
now,
it's
just
it
just
isn't
per
node.
It
is
it's
just
there.
A
And
so,
but
at
some
point,
you'll
end
up
seeing
a
few
panels,
a
few
charts
of
metrics
that
that
the
user
will
be
able
to
kind
of
fill
like
into
like,
based
on
the
fact
that
they've
highlighted
a
given
service
or
given
node,
or
maybe
a
collection
of
services
that
they
would
then
just
see.
Statistics
metrics
specific
to
those
nodes,
and
so-
and
so
in
this
case,
like
mesh
map
or
nashery
at
large,
is
not
it's
not
trying
to
displace
grafana,
rather
just
trying
to
augment
and
support
people.
A
You
know,
users
give
it
if
they
have
an
investment
into
grafana
and
into
the
the
charts
that
they've,
maybe
curated
the
panels
and
boards
that
they've
curated
there
meshri
will
re,
read
those
and
pull
them
in
and
let
people
use
their
existing
boards
and
charts
in
an
ongoing
fashion,
so
so
kind
of
need
a
little
bit
of
a
different
approach
than
than
just
trying
to
you
know
expect
give
you
a
link
to
grafana,
like
that's.
That's
only
so
helpful.
F
So
other
few
updates
should
we
go
into
that?
Please
yeah,
so
we're
gonna
this.
This
is
currently
this
state
static
data,
we'll
be
finishing
with
the
resolvers
and
then
we'll
have
the
real
data
where
you
can
view
your
services
under
that,
and
we
also
have
something
called
user
preferences
through
which
you
can
provide.
F
You
can
have
preferences
set
on
the
mesh
map,
like
you,
can
higher,
show
or
hide
edges,
and
all
that
so
zoom
in
level
all
that
and
then
there
needs
to
be
a
work
done
on
displaying
traffic,
which
is
which,
which
you
don't
have
right
now
so
that
need
to
that
needs
to
be
done
and
also,
if
you,
if
you,
if
you
want
to
contribute
it,
is
not
available
right
now.
F
The
branch,
the
mesmer
branch,
is
currently
moved
to
a
private
repose
and
you
won't
be
able
to
see
mesh
map
branch
right
now.
So
this
is
this
will
be
fetched
from
the
remote
component
which
which
it
will
be
faced
from
a
remote
component
which
a
mystery
server
provides.
You.
G
A
Well,
at
some
point
at
some
point,
kyali
will
have
to
eat
its
heart
out.
I
guess,
but
that's
not
that's
not.
Today,.
A
H
C
By
the
time,
android
shares
the
screen
I'll
give
some
context
in
the
issue,
so
the
mastery
docs
currently
does
not
support
versioning,
which
simply
means
that
you
can't
switch
from
one
version
of
the
docks
to
another
and
the
team
that
the
documentation
site
is
hosted
on
jekyll
with
a
team
called
doxy.
That
inherently
does
not
support
versioning,
but
we're
trying
to
use
a
converted,
plugin
or
implementing
this
and
android
will
be
explaining.
C
J
Okay,
so
basically,
this
is
documentation
site.
Currently
it
doesn't
support
versioning
and
that's
been
a
huge
pain
in
the
ass,
and
the
main
part
here
is
because
currently
we
are
moving
towards
adding
more
features
towards
two
meshri
and
we
might
deprecate
the
previous
stuff
and
you
still
need
documentation
for
that,
and
so,
as
many
companies
they
have
a
versioning.
J
We
are
trying
to
introduce
versioning,
but
what
we
are
using
now
we
are
using
jk
right
so
because
we
are
using
j
kyle
and
in
even
in
j
kyle
we're
using
a
theme
or
flavor.
You
could
say
that
doesn't
support
versioning.
J
So
there
are
a
few
ways
you
could
enable
versioning
and
after
some
research
like
I
finally
found
out
two
ways,
one
is
you
could
build
the
site
and
store
the
build,
build
into
different
folders
for
each
release,
and
that
is
the
first
approach
so
build
during
leads
and
store
versions
as
folder.
J
This
like
for
every
app
there's
a
pros
and
con,
and
the
good
thing
is
good.
The
only
good
thing
here
is,
you
have
a
understandable
structure
so
for
you
could
actually
actually
check
if,
for
at
least
0.4.2,
which
documentation
structure
is
and
how
everything
is
how
every
file
is,
and
you
don't
have
to
look
too
much
into
it,
because
you
have
all
the
files
that
you
didn't.
You
don't
need
to
switch
branches
or
need
to
know
how
github
actually
works
and
yeah
versioning
can
be
achieved.
J
But
the
cons
here
is
you
have
like
just
because
you're
creating
so
many
replicas
of
for
each
versions.
The
storage
size
would
put
a
dent
in
the
budget
or
at
least
it
wouldn't
be
efficient
and
that's
the
main
con
here
and
then
there
is
a
second
approach.
J
That's
we
use
preview
deployments
or
branch
deployments,
that's
provided
by
both
virtual
network
and
currently,
I
think
either
say
the
dock
mesh
we
currently,
we
we
use
netlify
and
netlife
also
supports
deploy
context,
and
what
this
would
do
is
it
would
use
specific
branches
to
deploy
and
you
can
actually
preview
what
that
particular
branch
is
using
branch,
name,
dot,
netify
dot,
whichever
tag
mystery,
has
dot
com
or
anything
that
it
uses.
J
So
the
best
part
here
is
the
structure
here
you,
you
won't,
have
multiple
folders,
that's
how
you
had
in
the
first
approach
and
each
link
is
independent
right.
So
there
is
a
base
latest,
which
is
which
is
the
stable,
stable
version
of
measuring,
and
then
you
have
for
each
different
release.
J
You
would
have
a
different
url
and
what
we
would
do
is
we
would
be
switching
urls
as
as
you
change
versions,
and
you
won't
actually
feel
it
just
because
everything
would
be
done
in
line
and
in
case
you
need
a
more
context.
So
this
is
the
link
for
versions
decal,
which
was
just
right
click
and
he
actually
created
a
plugin.
J
That
does
the
folder
approach
and
it's
currently
like
he
has
some
research
material
in
case
you
want
to
delve
into
this
further,
and
the
final
approach
you
could
come
up
come
up
with
is
having
each
of
these
deployments
in
a
folder,
which
is
the
best
we
can
achieve.
But
we
don't
want
this.
It's
too
many.
It's
too
many
files,
too
many
duplicate,
folders
and
the
second
thing
is
deploy
context.
J
The
second
approach-
and
that's
basically
you
have
what
we'll
do
is
we'll
have
branch
deploys
each
we
could
have
each
branch
or
we
could
like
we'll
figure
out
a
way
if
we
can
actually
have
releases
instead
and
do
a
release
deploy
every
time
so
on
every
push
you
get
something
like,
let's
just
currently
this
demo
right,
and
so
this
main
dock
this.
This
is
my
stable
one
and
I
had
a
different.
I
had
many
deploys,
so
let's
just
go
to
the
first
one.
So
this
is
this
and
what
is
that?
J
I
didn't
yeah
preview
it.
So
I
have
this.
This
is
my
first,
the
first
initial
comment
and
this
was
and
the
other
one
was
the
latest,
which
is
this,
and
this
is
the
current
version
right,
so
we
could
actually
find
a
way
to
process
this
and
not
get
these
deleted
and
have
versioning
switch
between
these.
This
will
help
us
further,
but
currently
we
need
to
look
into
pricing
on
how
this
works.
J
So
that's.
These
are
two
approaches
I
came
up
with.
If
you
have
any
more
approaches
in
mind
or
anything
else,
do
let
me
know
or
I'll
just
share
this,
so
you
can
add
it
to
the
dock.
H
A
Good
research,
that's
nice!
That's
good!
Any
feedback
for
aniruddh
any
other,
either
approaches
or
reflections
on
pros
and
cons.
E
J
If
you
want
to
implement
like
if
you're
using
approach
2
the
work
is
literally
almost
nothing
and
that's
the
main
reason
here
is
you
just
need
to
have
a
toolbar
or
we
just
need
to
program
a
toolbar
that
has
the
map
of
each
version
with
which
preview
deployment
it
has
right
and
just
putting
the
entire
website
in
line
is
not
a
big
thing,
because
you
simply
do
an
inline
deployment
and
approach
to
it.
J
It
wouldn't
take
much
time
it
would
be
simpler,
but
research
it'll
take
time
to
research
and
for
switching
themes.
I
did
look
into
it,
but
each
theme
has
its
own
benefit.
So
if
we
switch
we'll
have
to
look
into
what
we
are
losing
on
this
end
and
gaining
on
the
other
theme.
So
if
you
are
gaining
versioning
there,
we
might
have
to
shift
some
things
and
it
will
take
time
to
research
that
either.
H
J
Let's
just
changing,
can
you
can
you
just
name
the
theme
which
you
were,
which
has
version
like
yeah.
C
Could
you
point
some
of
those
I'll
set
an
example
in
the
chat
I'll
look
one
up.
J
Actually
read,
the
docs
is
on
python
as
as
far
as
I
remember
so
there
are
a
few
examples
we
have
sphinx.
So
spanx
is
again.
I
guess
it's
on
python
and
read
the
docs.
It's
also
on
python
and
it's
so
it
supports
j
kyle,
but
it's
not
jakai.
J
It
would
be
better.
My
initial
suggestion
was
that
switch
to
read
the
docs
is
something
that
supports
out
of
the
box
versioning,
because
in
if
we
do
anything
with
versioning
it'll
it'll
take
us
time
as
well
as
it
would
not
be
as
perfect
as
the
inbuilt
ones
right.
A
The
other
thing
is
the
the
inbuilt
ones
they're
going
to
be
done
really.
Similarly,
to
approach
number
one
and
approach
number
one
is
what
gets
built
into
the
other
themes
that
do
support
versioning
kind
of
out
of
the
box.
For
that
version,
approach,
number
one
is
is
probably,
is
probably
the
way
to
go
like
hey.
We
don't
in
number
two.
We
don't
have
long
lived
release
branches
today
we
could.
We
need
to
go
back
in
time
to
to
create
those
and
that's
possible,
because
we've
got
release
tags.
A
We'll
need
to
change
the
way
that
we
do
release
channels
and
the
way
that
where
people
are
trying
to
merge
their
pr's,
those
wouldn't
go
to
master.
Those
would
go
to
a
long-lived
release
branch
that
we
would
maintain
and
then
and
that's
that's
possible.
Those
things
are
feasible.
A
It
is
a
somewhat
decent
sized
change.
The
you
know.
Another
approach
here
is
to
move
to
move
to
a
different
jekyll
theme
that
has
versioning
built
in,
like,
like
you
guys
were
just
talking
about
in
to
answering
the
kills
question.
A
In
my
estimation,
it's
about
the
same
amount
of
effort
to
move
to
a
theme
that
has
versioning
as
it
is
to
retrofit
the
existing
content
with
a
version
and
that's
just
to
take
the
content
itself,
make
it
into
a
collection
like
create
the
folders
the
so
it's
about
you
know
it's
about
the
same
either
way.
Given
that
I'm,
I
would,
I
would
vote
for
just
sticking
with
the
same
theme,
because
you
end
up
working
out
a
bunch
of
kinks
and
css
and
etc.
A
The
theme
that
we
have
now
is
fairly
fairly
customized
to
be
to
incorporate
some
some
things
that
people
have
brought
along
about
like
copying
to
clipboard
or
a
special
footer
different
considerations,
considerations
for
multi-language,
that's
kind
of
coming
through
and
so
yeah
the
the
store
that
one
of
the
cons
on
the
first
one
like
the
storage.
It's
a
zero
concern
to
us
like
we,
it's
very
small
footprint
and
we
don't
the
community
doesn't
have
to
pay
for
it.
It's
just
there
on
github.
A
It
becomes
pretty
clear
to
understand
as
well
when
you've
got
when
you
go
into
the
docs
and
there's
versioned
folders,
okay.
So
anyway,
the
this
is
a
good
great
research
on
monterey.
I
think
either
of
the
two
that
you've
suggested
will
will
get
the
job
done,
and
I
don't
neither.
The
two
cons
are
necessarily
insurmountable
or,
like
aren't
thing,
you
know
a
little
bit
of
work
that
can't
be
done
to
overcome
them.
A
J
For
the
second
approach,
the
issue
like,
if
you
want
to
use
custom
domains
as
base,
so
this
is
an
optimization,
but
not
I
wouldn't
say
it
would
it
would
be
the
best
thing.
So
if
you
use
a
custom
domain
as
base,
you
could
purchase
the
deployments,
but
that
would
take
time
because
senior
records
take
at
least
a
minute
to
get
updated,
and
these
has
to
be
done
manual.
These
have
to
be
done
manually.
A
A
A
Something
to
check
into
I
I'll
bet
you
five
bucks
that
you've
got
to
pay
for
custom
domain
management
like
that.
J
H
J
J
C
I
added
a
couple
of
links
in
the
immediate
minutes
for
anybody
who
wants
to
go
either
figure
out
the
first
method
that
anarch
was
talking
about
or
go
check
out
an
example
shackle
site.
That's
a
post
version.
A
You
know
on
a
related
note,
we
we've
alonso
is
on,
and
so
I'm
compelled
to
ask
this
do
we
did
we
end
up
with
a
plan
of
action
for
multi-language.
K
Hey
yeah
hi
good
morning
good
afternoon.
I
mean
well
kind
of
because
we
are
experimenting
with
the
multi-language
plugin
and
the
order
or
the
structure
of
the
folders.
That
has
to
be
done
to
make
it
work
and
because
we
are
in
in
that
process.
Yet
we
haven't
been
able
to
get
into
the
whole
message:
redux
files
and
folders
and
stuff.
K
I
don't
know
if
we
could
say
we
may
say
that
we
have
a
plan
that
they
have
a
timeline
and
such,
but
I
I
think
is
it's
not
going
to
be
a
lot
of
of
days
to
to
make
it
work,
but
I'm
still
experimenting
with
it.
A
You
know
yeah,
I
agree.
There
is
some
mystery
there
with
the
with
the
ruby
and
what
are
those
called
those
macros?
What
are
they
called.
A
Anyway,
I'm
actually
curious
dude,
since
you
asked
the
question
about
the
docs:
do
you
are
you
familiar
with
jekyll.
D
D
Yeah
and
I
I
will
review
if
I
can
make
any
comment,
I
I'll
do.
A
Nice
so
good
dude
yeah,
that's!
I
kind
of
I
kind
of
thought
that
that's
what
you
meant
like
hey.
Are
you
gonna
and
that's
such
a
perfect
question
to
ask
dude
that
it's
like
well
hey.
Maybe
we
should
consider
that
a
service
match.
A
Can
you
know
help
here
specifically
around
like
sort
of
an
approach
toward
version
two
or
really
either
one
like
yeah,
there's
a
as
a
matter
as
a
matter
of
fact,
what
a
kind
of
a
nice
way
to
cheat
a
little
bit
and
just
you
know,
have
that
redirection
happening
within
the
mesh.
D
The
map
previously
showed
was
really
interesting,
and
just
this
idea
popped
up.
If
we,
if
there
will
be
any
opportunity
to
combine
experiments
with
visualization
that
the
map
you
have
and,
for
instance,
create
scenarios
with
the
map
yeah
to,
for
instance,
to
do
some
experiments
on
reliability
and
connectivity,
it
would
be
quite
useful
for
owning
services.
I
think.
A
Yeah,
it's
totally
because.
D
Vietnamese
topology
visualization
and
I
think,
for
instance,
if
we
could
be
able
to
create
scenarios
with
topologies
like
with
topology
mapping,
just
visual
and
be
able
to
run
that
experiments
would
be.
I
think,
quite
easy
and
oh
yeah.
A
Oh
yeah
yeah
totally
due
to
drew
beneath
anirud
other
folks
that
are
working
on
it
like
you
guys
that
should
be
affirming
of
some
of
the
discussions
that
that
you've
had
in
the
past.
A
Like
do
I,
for
my
part,
part,
I
totally,
I
totally
agree,
and
I
would
expand
on
that
a
little
bit
to
say
like
one
a
number
of
different
types
of
experiments
like
for
some
depending
upon
context,
experiments
people
immediately
go
to.
Oh,
you
mean
chaos,
engineering
and
experimenting
there,
and
it's
like
yes
and
for
sure,
and
also
some
other
types
of
experiments
like
some
experiments
around
chaos,
engineering
and
really
focused
toward
resiliency
and
understanding.
What
happens
when
and
if
this
scenario
occurs
and
et
cetera.
A
You
know
great
beautiful,
there's
a
whole
like
like
that's
a
you
know
that
that's
there's
a
whole
project
within
there
right
right
in
terms
of
like
how
much
support
you'd
want
to
offer
for
that,
and
the
nice
thing
is
that
same
sort
of
feature
functionality
is
extremely
applicable
to
a
whole
set
of
other
experiments
and
dude.
I
don't
know
if
you
can.
D
A
So
the
other
types
of
experiments
are
well,
they
can
be
like
multiple
other
categories.
A
Actually,
things
toward
answering
questions
around
well,
optimization
or
like
things
like
well,
depending
upon
a
few
parameters
that
the
the
user
would
identify
and
set
things
like
well,
things
like
the
anticipated
load
or
the
the
sla
or
the
slo,
the
the
error
budget
or
the
the
number
of
nines
that
that
they
desire
for
a
given
deployment
either
in
the
entire
mesh
or
just
for
that
application
like
what's
the
can,
can
we
run
an
experiment
to
help
us
identify
what
that
is
like
hey,
how
many?
A
How
quickly
should
the
circuit
breaker,
how
sensitive,
should
it
be?
How
long
should
it
stay
out?
What
are
the
so.
A
Totally
there
there's
we'll
have
to
send
you
a
link
to
get
nighthawk
there's
a
couple
of
well
there's
a
few
contributors
that
are
not
on
the
call
today,
who
I
think
would
be
all
over
your
questions
because
they
are
well
they're.
They're
doing
research
on
this
topic.
We're
I'll
try
to
I'll
try
to
be
well.
We
have
five
minutes
left,
so
I
guess
this
is
treaty.
We
don't
have
any
other
topics
right.
Do
we
or.
H
J
Is
great
just
a
quick
point.
A
A
This
is
good
research,
iron,
so
doo
doo,
there's
a
project
called
get
nighthawk
that
well
intends
to
help
or
intends
to
have
a
sharp
focus
on
envoy's
load
generator,
which
is
which
you
might
be
familiar
with,
but
it's
called
nighthawk
and
it
has
an
upcoming
capability
around
adaptive
load
control
which
we're
just
on
the
forefront
of
engaging
with
a
university
and
a
professor,
a
few
of
his
students
to
do
some
academic
research
around
those
sorts
of
optimizations,
like
around
saying,
hey,
I'd
like
to
run
an
I'd
like
to
run
an
experiment
so
that
the
mesh
map
or
meshri
or
nighthawk.
A
This
combination
of
things
can
tell
me:
hey
here's
my
here's,
the
inflection
point
by
which,
like
here's,
your
desirable
configuration
based
on
your
environment
and
your
particulars
that
you're
probably
going
to
want
to
we're
gonna.
You
know
some
of
the
the
questions
that
we're
trying
to
answer,
and
actually
some
of
your
feedback
on
what
questions
to
answer
would
be
really
helpful.
A
We're
gonna
try
to
look
at
service
mesh,
specific
optimizations.
You
could
look
at
an
optimization
or
an
experiment
to
say
well
how
many,
based
on
my
resiliency
criteria
and
the
load
that
that
our
production
systems
undergo
what
should
we
have
for
a
replica
set
count
or
a
scale
set
like
okay.
That's
that's
interesting,
but
it's
also,
you
know,
sort
of
kubernetes
generic
or
kubernetes
general,
like
some
more
service
mesh.
A
Specific
things
are
like
hey:
what's
the
cost
of
a
hundred
percent
distributed,
trace
sampling
of
distributed
traces
gonna
cost
me
here,
given
the
profile
that
I
need
to
maintain
for
latency.
For
our
users,
we
don't
we're
not
gonna,
go
we're
gonna,
we're
not
gonna
dip
under
400
milliseconds
per
response
or
or
whatever.
Those
things
are.
A
I
think
there's
a
lot
of
exciting.
Those
are
the
types
of
questions
we're
going
to
look
to
try
to
empower
users
with
to
go
answer
in
their
own
environment.
A
But
very
good
yeah
I'll
try
to
follow
up
to
point
out
to
you
some
of
the
some
of
the
questions
that
we
think
we're
going
to
answer,
but
it'd
be
really.
I
think
it
would
be.
Your
input
would
be
really
warmly
received
like
any
amount
of
input,
just
especially
if
you
think
that
ants
look
studying
or
trying
to
answer
some
of
the
questions
isn't
really
helpful.
A
A
Nice,
okay
good
deal
did
all
right.
A
That
was
just
that.
I
wanted
to
send
out
a
thought
plant
a
seed
in
people's
minds.
I
wanted
to
raise
up
utkar
shrivastava
as
a
potential
measuring
maintainer,
and
so
I
just
want
to
so
so.
A
No
no
decision
or
no
voting
or
what
have
you,
but
I
wanted
to
for
the
community
here
if
you
have
comments
or
plus
ones
or
negative
ones,
or
what
have
you
to
give
we'll,
provide
a
mechanism
for
that
and
for
the
existing
maintainers,
we'll
be
sending
around
a
nomination
and
we'll
all
be
talking
about
utkarsh
behind
his
back.
So.
A
I
just
want
to
call
that
out
here
cool
all
right.
You
didn't
that
that
would
literally
like
there's
like
one
other
thing
nivendu,
who
he
who
had
said
he's
not
able
to
make
this
call
he's
he's
gonna
be
he's
physically
moving
at
the
moment
he
has
called
out
he's
he's
working
on
a
set
of
new
features
in
measuring
ctl
and
meshes
cli
to
support
using
measuring
ctl
to
provision
measuring
in
into
a
kubernetes
cluster.
A
So
if
you've
used
mastery
ctl
today,
you'll
notice
that
it
will
provision
meshrie
to
your
local
docker
machine
docker
host,
but
we
need
to
give
users
choice
because
some
users
want
to
deploy
mastery
into
kubernetes
cluster
and
so
he's
laid
out
a
number
of
issues
to
build
that
support.
And
so,
if
you're
and
he's
he's
looking
for
other
contributors,
if
people
are
interested
there,
so.
C
C
Apart
from
that,
that's
it
thank
you.
Everybody
have
a
good
weekend
same
time
next
week,
bye.