►
From YouTube: Meshery Development Meeting - March 4th, 2020
Description
Welcome Sanjay Mishra!
Learn more about Meshery at https://meshery.io.
A
A
A
Okay,
all
right
good,
good,
good
good
in
Chandra
is
with
us.
So
that's
great.
We
should
be
able
to
talk
about
some
gke.
Actually,
it's
good
that
hey,
Zeus
and
Chandra
are
here
because,
because
part
of
that
work,
I
think
the
two
of
you
and
looked
at
a
little
bit
in
the
past,
but
I
don't
know
that
you've
spoken
to
each
other
about
it.
So
this
is
going
to
be
gonna,
be
good.
A
C
C
Like
background
of
zippered
products,
so
I
don't
have
much
idea
very
time
and
about
me
you
have
more
than
5
years
of
experience
in
ITN
the
mainland
in
exercise,
so
I'm
working
with
Walmart
labs,
where
I'm
managing
their
food
kubernetes
platform,
where
we
are
you
using
ECU
another
product
as
part
of
the
government
platform,
and
we
have
more
than
300
k,
CP
universi
pupils
working
at
the
dubonnet's
cluster,
which
is
self
management.
So
we
are
mainly
filter
for
our
Intel
investments.
That
would
be
about
me.
That's.
A
Awesome
say
it
again
so
Sanjay
in
now,
I
remember
where
we
met
Sanjay.
So
this
is,
this
is
great.
I
think
I
want
to
try
to
catch
up
with
you
after
the
call.
If
we
can,
but
the.
Let
me
ask
you
this
and
I'm
gonna
when
a
grimace
want
to
ask
this
question,
but
our
is
OpenStack
part
of
your
world.
Still
Walmart
labs,
listen
to
OpenStack,
quite
a
bit
in
the
past.
A
A
A
A
Okay,
alright,
so
here's
the
link
to
the
issue.
This
particular
issue
is
about
the
notion
that
you
can
run
messy
and
point
it
at
the
Google's
kubernetes
engine
gke
and
when
you
do
I
believe
you'll
need
to
create
a
cube
config
file
such
that
Mestre
is
able
to
remotely
interact
with
gke
there's
a
script
that
we
have
used
to
help
us
to
do
that.
But
it's
kind
of
a
pain
for
many
users,
because
those
those
users
might
be
running
messy
locally.
A
They
go
to
connect
to
gke
and
when
they
do,
they
need
to
authenticate
in
the
right
way.
In
order
to
do
that,
they
need
to
run
a
short
script.
What
would
be
a
lot
more
convenient
is
if
mesh
race
ETL
was
able
to
if
they
were
able
to
run
a
mesh
tree
CTL
command
to
produce
the
cute
config
file
that
they
needed
for
to
interface
with
gke.
A
A
And
then
just
some
some,
while
back
I
think
that
issue
is
probably
an
interesting
one
to
you,
but
I
know
you've
been
helping
in
a
few
different
areas
and
probably
I,
don't
know
that
how
much
time
you'd
actually
gotten
to
advance
this.
So
since
you're
both
here
I
thought
it
would
be
good
to
talk
about
this.
One.
A
A
E
D
A
A
But
hey
soos,
if
you're
not
seeing
it
I
was
just
gonna
pull
I
can
see
you,
okay
cool,
so
so
you
know
so
a
measure
user
as
they
go
on.
They
identify
that
the
platform
that
they're
running
is
gke.
You
know
just
like
I
think
a
number
of
you
have
run
measuring
against
a
mini
cube
environment,
and
when
you
do,
you
need
to
generate
this
mini
cube,
cube
config.
This
mini
cube,
yamo
file,
it's
very
similar
to
what
users
need
to
do
for
gke
and
to
do
that.
A
A
Installation-
okay,
here
it
is
generate
cube
config
for
G
key,
so
in
some
respects,
maybe
ignoring
a
little
bit
of
what
it
actually
does,
but
just
understanding
that
what
we're
trying
to
do
like
the
goal
of
this
issue
is
to
is
to
understand
that
measure.
E
users
are
already
interfacing
with
massery
CTL,
and
so
they
use
mastery
CTL
like
start
to
run
ma
cherie,
and
if
it
so
happens
that
they're
using
a
GK
environment,
it
would
be
nice
if
they
were
able
to
do
something
like
mastery.
Ctl
start.
A
A
A
E
A
A
D
A
Believe,
like
it's
gonna,
be
this
service.
It
can
like
yeah,
it's
interesting,
actually,
maybe
to
actually
write
it
out.
You
don't
need
access
to
GK,
certainly
to
test
it.
You
do
but
I
believe
like
what
you'll
end
up
doing
is
like
the
bash
script.
Today
it
looks
like
it
accepts
at
least
two
parameters,
the
first
parameter
being
yeah,
the
name
of
your
account
within
gke,
or
rather
maybe
it's
not
the
name
of
your
account
as
much
as
it's
the
name
of
the
the
kubernetes
account
that's
been
created
as
your
service
account.
A
And
might
as
a
good
practice
I
would
expect
that
they
probably
create
the
servant.
A
kubernetes
service
account
for
applications
like
measure
e,
where
they
acknowledge
that,
like
measure
e
they'll
use
against
they'll
use,
mesh
reinstate
closer
on
going
and
measuring
needs
permissions
to
execute
some
of
the
things
that
it
does,
because
measure
e
will
create
namespaces,
it
will
deploy
surface
meshes
and
it
needs
permission
to
do
that.
It
would
do
that
under
this
account
name.
A
Whatever
that
happens
to
be,
and
so
I
think
you
has
the
best
practice
as
a
kubernetes
administrator,
you
would,
as
you
decide
that
you're
gonna
use
a
tool
like
ma
cherie.
You'd
go
over
and
you'd
make
an
account
for
mastery,
and
so
people
who
run
this
script
that
they've
created
an
account
they'll
run
the
script,
though
they'll
include
that
name
as
a
first
parameter
as
the
second
parameter,
the
assignment
namespace
here
but
I'm.
Not
so
I,
don't
know
if
this
namespace
is
for
where
they
want
to
install
measure
e.
Yes,
that
would
be
the
case.
A
Yes,
because
in
this
case
it's
it's
identifying
where
they,
what
namespace
they
want
to
install
memory
into
a
lot
of
times.
Kubernetes
administrators
might
don't
end
up
doing
I.
Think
a
couple
of
different
things
generally
they'll
either
say
they'll
either
deploy
one
cluster
one
kubernetes
cluster
and
saying
that's
my
management
cluster,
that
one
is
gonna,
run,
ma
cherie
and
Prometheus
and
grifone
a--,
and
a
bunch
of
other
kind
of
management
tools
and
things,
and
then
they'll
run
different
clusters
that
have
all
of
their
workloads.
That's
one
way
that
people
will
do
this.
A
Another
way
that
people
will
set
up
their
kubernetes
clusters
is
that
they'll
set
up.
You
know
different
clusters,
they'll
have
different
workloads
on
them
and
and
each
cluster
they
might
identify
a
namespace
called
management
or
monitoring.
Or
you
know,
I
was
like
a
generic
name
space
to
put
measure
E
and
other
tools
into
or
they'll
simply
do
something
like
create
a
namespace
per
application
that
they
want
to
deploy
so
one
namespace
for
gravano,
one
namespace
for
Prometheus,
one
namespace
for
ma
cherie
so
anyway.
A
D
A
Okay,
yeah,
that's
great,
that's
really
very
relevant
experience.
Part
of
what
ma
cherie
CTL
does
is
to
use
the
kubernetes
go
client
to
go.
You
know,
pull
back
some
information
about
different
objects
that
are
deployed
in
kubernetes.
This
particular
this
particular
script
is
similar
in
concept.
It's
slightly
different
in
that
I.
Don't
know
that
you'll
need
to
use
the
kubernetes
go
client
to
perform.
Some
of
these.
You
know
to
do
what
the
script
is
doing,
but
but
that's
very
relevant
experience.
I.
Think!
Oh,
that's!
That's
great!.
A
A
A
A
Thanks
good
and
then
hate
chunk
rest,
just
as
a
on
a
related
note,
there's
an
sto
go
client
as
well
that
that
will
be
in
this
project
will
be
needing
to
interface
with
that
one
in
a
very
similar
way
to
kind
of
retrieve
objects
about
steel.
So
it
could
be
a
good
area
and
good
next
area.
For
you,
nice,
okay,
well
nip
or
you
want.
F
Hi
everyone
I
just
wanted
to
like
recently
I,
was
going
through
the
website
of
like
layer,
5,
dot,
IO
and
I
realized
that
there
is
a
part
other
like
at
the
bottom,
which
Lee
is
presenting
right
now.
I
was
like
feeling
that
the
color
of
it
is
faded.
I,
don't
know
it's
just
me.
Oh
you
totally
kind
of
like
this
is
like
very
my
nude,
but
like
was
bothering
me
so
I
thought
I
should
put
up
in
the
meeting
today
like.
Is
this
how
it
is
or
is
it
some
bug,
activity
yeah.
A
No,
that's
a
great,
it
was
actually
believe
it
or
not.
It
was
intentional
so
that
what
we
did
was
each
of
those
icons.
Each
of
those
flags
are
SVG's
and
what
we
did
was
using
CSS
just
change
the
opacity
like
the
transparency
Oh
each
of
the
SVG's
to
be
lighter
and-
and
this
is
this
is
just
an
opinion,
so
dip
or
maybe
we
want
to
change
this.
The
I
think
my
perspective
at
the
time
was.
A
It
is
something
that
people
will
use
not
very
frequently
and
so
having
the
full
100%,
color
and
opacity
was
I
was
like
almost
calling
too
much
attention
to
the
flags
because
they're,
very
colorful
and
they're.
You
know
red
and
green
and
yellow
and
blue
and
yeah,
and
now
to
your
point,
there's
a
couple
of
there's
like
one
of
the
issues
with
having
that
opacity
lower
is
I
actually
had
to
go
back
into
the
Japanese
flag,
a
little
after
of
fact
and
I
ended
up,
putting
I
think
I
put
a
board
here.
A
Let
me
blow
it
up.
Cuz
I
know
it's
kind
of
I.
Think
I
put
a
literally
put
a
border
around
the
Japanese
flag
because
you,
otherwise
all
you
saw-
was
a
loop
that
yeah
and
so
yeah.
If
we
take
a
look
real
quick
if
we
sort
of
inspect
that
code
here
is
the
Japanese.
If
we
can
see
the
word
Japanese
the
Japanese
flag.
Well,
we
ended
up.
Doing
was
I
used
well,
I,
guess
now:
I'm
gonna
prove
myself
wrong.
A
F
A
F
A
So,
while
I'm
browsing
here,
I'm,
just
oh,
this
is
interesting
actually
in
the
poor
hears
that
a
bug
is
that
I
set
the
language
to
Spanish
just
to
please
hey
soos,
on
the
call
and
and
and
then
actually
as
I
browse
to
something
like
this
performance
page.
This
talks
about
the
fact
that
measure
e
is
defining
new
specification.
That's
setting
a
new
standard
for
how
you
measure
ServiceMaster
horns.
A
The
problem
is,
it
went
back
to
English
and
the
reason
it
went
back
to
English
is
because
we
have
two
types
of
footers
on
the
site:
this
one
it
is
one
and
and
the
other
one.
Is
this
one?
It's
kind
of
got
the
curve
down
here
and
then
all
this
and
the
other
one
it's
missing
the
languages
I
think
so
a
little
bit
of
a
bug
like
it's
to
your
point.
We
would
do
well
to
open
up
a
bug
to
ensure
that
that
other
style
of
footer
has
the
languages
feature.
F
A
A
A
And
then
on
the
topic
that
Anand
was
referring
to
about
mastery
CTL
Perth.
That's
a
great
discussion.
I'll
note
that
as
we're
talking
about
it,
there
are
two
other
individuals
who've
given
some
thought
to
it.
My
ink
Shah
gave
some
thought
to
it:
I,
don't
think
that
he's
done
any
work
on
it,
though,
and
the
other
one
is
Kanishka
J.
G
H
G
G
G
A
You're
right,
yeah
yeah,
that's
it's
bad
it!
So,
hopefully
everyone
else
can
see
what
you
kill
is
talking
about,
which
is
the
fact
that,
oh,
that,
if
you
don't,
if
you
have
a
not
such
a
tall
browser,
you
will
end
up
with
what
look
like
fairly
ugly
logos,
I.
Think
in
so
that's
a
bug.
I
think
from
a
left-to-right
perspective,
we're
doing
okay
but
yeah.
It
needs
improvement
to
kill
that
needs
fixed
you're,
right
I
had
so
guys,
I,
don't
know.
If
you
know
I
mean
I'm
gonna.
A
Try
to
give
a
big
community
update
on
Friday,
but
I
just
I
had
the
Hashi
Corp
logo
up
here
earlier.
I
took
it
down
briefly
yesterday
to
finish
up
conversations
with
them:
they've
been
pulling
layer
five
and
mystery
into
some
of
their
accounts,
because
some
of
those
accounts-
I,
won't
I,
won't
say
the
name
of
the
customers.
A
These
are
very
large
accounts
where,
where
people
already
know
about
measure
e
they're
already
they're
very
excited
to
talk
to
layer,
five
representatives,
because
because
they're
excited
about
massery
and
they're
going
off
and
trying
to
decide,
should
they
use
Sto
or
should
they
use
console
from
Hoshi
Corp,
and
that's
why
Hoshi
Corp
is
pulling
in
layer
five
to
help
with
that
evaluation,
because,
from
our
perspective,
we
just
want
people
to
be
successful
with
adopting
service
meshes
and
whether
that's
any
of
the
meshes.
It
doesn't
really
matter
it's
just
whatever
is
the
best
fit
for
them.
A
So
that's
why
I'm
Hoshi
core
pulls
us
in
because
they
want
us
to
help
represent
console
and
anyway,
so
that
what
my
point
is
that
Hoshi
Corp
logo
was
right
here,
it'll
be
back
up.
I
expect
later
this
week,
I'm
meeting
with
docker
tomorrow,
I
met
with
him
last
week
and
meeting
with
them
again
tomorrow
about
two
things:
one
is
about
partnership
with
docker
I'm
gonna.
You
know
what
maybe
I
should
give
all
this
update
on
Friday
just
suffice
to
say
a
lot
of
exciting
activity.
A
So
if
you
look
under
partners,
yeah
you'll
see
that
a
couple
of
new
ones
have
been
added,
so
this
is.
This
is
good.
These
are
people
that
have
used
ma
cherie.
These
are
people
that,
like
Citrix,
is
a
well.
It
has
become
a
small
customer
of
layer
5,
and
so,
while
it's
not
much
funds,
those
funds
I
anticipate
that
we
will
use
to
hopefully
engage
a
few
of
you
in
what
would
be
you
know:
Skype
ended
or
paid
internships.
A
A
You
kill
you're
right.
We
should
probably
file
a
bug
about
the
logos,
partner
logos
and
then
the
keel.
There
was
one
other
item
that
I'm
gonna
try
to
bring
up
now
on
the
provider.
Ui
I
think
it
would
be
helpful,
maybe
for
everyone
to
weigh
in
on
their
perspective,
about
the
provider.
Ui
I,
don't
know
if
Santa
Sanjay
is
still
on
the
call.
A
So
this
is
good
so
Sanjay,
maybe
what
we're
gonna
bring
measuring
briefly,
so
one
you'll
get
to
see
a
little
bit
of
what
it
looks
like
and
then
there's
a
initial
interface
that
people
see
and
they
need
to
choose
a
how
they
want
to
log
in
to
measure
e,
and
it
would
be
good
for
to
hear
from
each
of
you
about
what
you
think
about
this
interaction
so
because
Makela
has
been
doing
a
good
job
of
updating
it
and
he's
got.
Hopefully
one
more
update
to
do
so.
A
A
So
my
ID
had
the
font
on
my
terminal
window
really
large,
because
I
was
on
with
Kanishka
the
other
day
and
we
could
only
get
connected
on
his
mobile
phone,
so
he
was
having
to
look
at
my
screen
on
his
mobile
phone
so
anyway,
so
on
my
machine
and
I
was
running
mini
cube,
I'm
gonna,
stop
running
mini
cubics.
That
me
too,
on
on
my
machine
I'm,
also
running
docker
desktop
so
I'm
running
a
single
node
cluster
of
kubernetes
under
docker
desktop
I.
A
One
thing
that
I
would
note
like
Chandra
SH.
This
could
be
a
or
hey-zeus
like
an
easy
first
win
with
contributing
to
mastery
CTL.
This
description
of
what
mesh
Ria's
was
previously
accurate
anymore.
It
doesn't
disservice
to
how
people
think
of
massery
the
term
playground.
Is
it's
accurate
because
people
can
go
off
and
play
with
a
bunch
of
service
meshes,
but
it
also
has
the
connotation
of
Mary
just
being
a
toy
and
not
really
a
tool
that
people
would
use
in
production
or
on
there
or
to
ongoing,
operate
their
environments.
A
A
All
right
so
then,
back
to
so
Sanjay
and
others.
Just
I
want
to
talk
about
the
provider
UI.
But
I
was
just
going
to
say.
There's
this
mystery,
CTL,
client
and
to
start
you
can
run
massery
in
kubernetes
in
a
cluster
or
outside
the
cluster
and
that
kubernetes
can
be
running
in
gke
or
uks,
or
you
know,
or
locally
and
in
my
case
is
running
locally.
What
I've
done
is
I've
deployed,
I
started
Mary
and
Mary
is
now
running
on
my
docker
host.
A
A
You
were
accustomed
to
using
software
that
when
you
you
go
to
use
it,
you
sign
in
as
a
user,
because
you
have
a
profile
and
preferences
and
an
account
and
so
seeing
the
fact
that
you
might
need
to
sign
in
or
select
in
this
case
select
the
provider
is
not
the
weirdest
thing
in
the
world.
I
would
expect,
but
this
term,
where
it
says
in
where
it
says,
please
choose
a
provider
to
continue.
I,
don't
know
that
provider
is.
C
A
A
People
associate
the
term
provider
what
they
associate
that
to
Sanjay
I
would
say:
hey.
You
know,
I
hear
that
term
I.
You
know
first
thought
that
pops
into
my
head
is
what
infrastructure
provider
do
I
want
to
connect
to,
and
that
is
in
fact
what
we're
doing
here.
We're
we're
saying:
hey
my
local
nursery
instance.
I
would
like
to
connect
you
to
this
infrastructure
provider
that
the
problem
is.
Is
this
infrastructure
provider?
It's
a
it's
a
measure,
II
persistence
layer,
it's
not
a
provider
like
UK
s,
gke
local,
kubernetes,
local
docker.
It's
not
a
platform.
A
It's
not
like
it's
not
like
you're
choosing
between
mini
Cube,
blah,
blah
blah,
and
so
my
concern
is.
Is
that
the
way
that
we're
using
provider
here
while
it's
appropriate
it
without
further
explanation?
The
first
time
a
user
comes
to
this
interface,
they
may
not
understand
which
one
of
these
two
they
shouldn't
necessarily
choose,
there's
not
a
wrong
choice.
A
It's
just
that
one
of
the
choices
they
have
a
lesser
experience
with
it's,
not
as
good
of
an
experience
and
that's
the
that's
the
default
local
provider
that
that
basically
lets
someone
sign
into
Missouri
without
as
a
generic
user.
So
they
really
don't
have
an
account
and
really
can't
set
and
remember
preferences.
A
A
Service
meshes
that
I'm
using
docker
desktop
that
I
have
in
the
past.
Well,
hopefully,
it
would
remember
in
the
past
that
I've
run
some
performance
test
results.
It
may
be
that
my
session
yeah,
so
this
is
funny.
Okay,
something
really
funny
just
happened,
and
we
gonna
explain
this
anyway.
It
would
normally
return
all
of
my
performance
test
results
and
I'm
gonna
show
you
that
it
will
in
a
second.
A
But
do
you
guys?
Did
you
guys
see
it
there?
It
just
logged
me
out
right,
I
just
lost
my
session,
and
now
it's
showing
me
a
new
provider.
Ui.
Do
you
guys
see
a
difference?
Yes
yeah.
You
know
what
happened.
This
is.
This
is
pretty
interesting.
I,
don't
know
if
this
has
ever
happened
to
any
of
you,
but
I
think
that
it's
a
good
I
think
it's
something
something
something
good
to
take
away
and
something
that
you
will
each
find
interesting.
Okay.
A
So
what
just
happened
is
I
got
my
session
got
killed,
I
got
logged
out
and
then
all
of
a
sudden
I'm,
seeing
a
new
version
of
mastery
that
provider
UI.
That
was
a
new,
a
new
release.
Well,
what
happened
was
there's
a
docker
container
that
comes
with
measure
e,
that's
called
watchtower,
and
this
community.
Our
community
didn't
write
watchtower,
but
it's
a
common
purpose
utility
for.
A
Watching
to
see
if
any
new
container
images
land
inside
of
docker
hub,
because
when
when
we
create
mesh,
read
containers,
you
know
one
for
different,
each
for
different
types
of
service,
meshes
sto
and
console
and
linker,
D
and
etc.
As
we
make
enhancements
and
publish
new
releases
if
you've
been
a
measure,
a
user
for
a
while
and
you're
just
running
memory
right
now,
unless
you
do
something
different,
the
default
deployment
of
nursery,
it
runs
an
instance
of
watchtower.
So
if
we
were.
A
Status,
so
this
is
the
status
of
mastery.
Ctl
is
running,
watchtower
and
watchtower
is
carrying
this
label
of
being
enabled.
So
it's
sitting
there
looking
checking
in
with
docker
hub
to
see
if
there's
any
newer
versions.
Any
newer
images
of
any
of
these
containers
point
is
it.
There
is
because
I
hadn't
run
macerated
locally
for
a
few
days,
Watchtower
found
a
new
container
image
for
missery
downloaded
it
in
the
background.
A
While
we
were
all
talking
just
now
and
then
look,
you
know
swapped
out
that
container
loaded
up
that
new
image,
which
is
why
we
were
seeing
a
new
interface.
So
I,
don't
know
how
many
of
you
I,
don't
know
how
many
of
you
actually
were
aware
of
that
before,
but
but
so
it
suddenly
kill
and-
and
myself
had
done
a
little
bit
of
work
on
the
UI
could
try
to
make
it
more
understood
for
users
what
the
difference
between
these
two
choices
are,
but
I
consider
it's
still
not
very
good
description.
A
It's
still
not
easily
understood
that,
okay,
so
as
a
user,
if
I
choose
the
local
as
the
provider,
I
will
get
an
ephemeral
session
and
it's
free
to
use.
Okay,
if
I
choose
my
free
cloud,
I
will
get
a
persistent
session
and
it's
free
to
use.
Okay,
this
just
isn't
a
great
format
and
so
Nikhil
hope.
I'm
keep
looking
at
you.
A
Hopefully
we'll
figure
out
a
way
to
like
over,
on
the
right
hand,
side
or
somewhere
or
down
below,
to
help
explain
a
bit
about
the
difference
in
choice
and
also
my
ask
for
everyone
here
on.
The
call
is:
can
anyone
think
of
a
better
term
other
than
provider?
Maybe
something
else
we
should
be
using
here
something
more
intuitive.
B
A
A
It's
unfortunate
that,
though
I
was
gonna,
do
a
quick
mock-up,
but
it's
unfortunate
that
Google
Hangouts
only
lets
you
share
like
one
application
at
a
time
or
it
shares
the
whole
desktop
and
but
yeah.
My
perspective
is
that
of
the
things
that
we
would
like
to
tell
people
about
my
forgive
the
difference
between
these
two
providers.
I'm
gonna
paste
it
into
the
Mestre
slack
right
now,
so
I
just
pasted
it.
Hopefully
everyone
can
see
it
in
the
mastery
channel.
These
are
brief
descriptions
of
the
differences
between
the
two
choices
and
I
guess.
A
My
thought
was
that
as
a
user
that
that
the
drop-down
would
simply
say
something
like
Mestre
cloud
or
it
would
say
local
and
when
a
user
chose
one.
What
would
happen
when
they
clicked?
Is
that
like
I,
don't
know
if
you
can
see
my
mouse
but
like
right
over
on
the
right
hand,
side
that
that
same
set
of
bullet
points
or
a
small
description
would
be
visible
and
would
change
based
on
their
selection.
A
The
other
thing
that
I
think
would
be
helpful.
Is
this
word
provider
or
if
we
change
it
to
a
different
word,
but
irrespective
this
word
would
be
a
point
of
interaction
with
the
user
that
if
they
that
it
would
have
an
underline
or
a
small
little
like
information
button
next
to
it,
that
if
someone
clicks
on
it
to
that
using
that
same
component,
we
would
describe
like
that.
A
G
A
Mastery
seek
or
for
invoking
a
performance
test
through
measure
ease
rest
based
interface,
and
it
is
right
here
so
this
is
an
example.
Rest
call
to
start
a
performance
test.
This
same
rest,
interface,
you
know
on
on
massery
would
be
the
interface
that
you
you
know
that
the
end
point
you
would
invoke
from
measure
ECT,
l.
A
Part
of
how
you're
describing
the
the
test
itself,
because,
as
you
consider
my
sri
CTL
perth,
it
may
come
with
any
number
of
flags
or
it
might
simply
come
with
the
fact
that
people
can
point
to
a
file
Ashley
CTL
Perth,
and
they
link
to
a
file
that
has
that
describes
what
endpoint
they
want
to
generate
load
against.
How
much
load
to
generate
how
long
to
run
the
test.