►
From YouTube: Meshery Development Meeting - February 26th, 2020
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
C
A
A
C
A
Nice
yeah,
you
guys,
can
see
the
meeting
minutes
now:
yeah,
okay,
good
sesh,
I,
messaged
pratik
in
email
and
in
slack.
So
he
is
one
of
those
you
know
he's
one
of
those
lazy
graduate
students-
and
so
you
know
he's
probably
sleeping
in
and
he's
probably
but
but
I
was
talking
to
him
late
last
night.
We're
making
preparations
for
pratik
to
head
out
to
Amsterdam
to
present
a
cute
console
and
part
of
that
presentation
is
around
floats
around
performance
overhead
for
a
specific
Network
function.
If
you
will
or
a
specific.
A
A
A
So
actually
do
guess
since
we're
just
while
the
other
guys
are
preparing
their
sections.
One
thing
that
strikes
me
that
well
is
anyone
that
joins
the
community,
but
I
think
also
in
particular.
You
is
to
have-
and
maybe
you've
done
this
already,
but
if
you
haven't
run
through
the
workshop,
that's
whoops!
That's
at
this
address.
E
I'm
lying
I,
don't
really
like
number
I'm,
not
sure,
okay,
yeah.
A
A
That
has
a
set
of
labs
that
will
have
you.
They
walk
you
through
deploying
measure,
E
and
then
deploying
sto
and
then
beginning
to
play
around
with
this.
Do
the
prerequisites
for
the
labs
are
that
you
need
to
just
have
a
machine
that
has
docker
installed,
which
can
just
be
docker
desktop,
so
to
get
to
your
your
laptop
is
that
a
Windows
laptop
or
a
different
operating
system.
E
A
A
A
A
C
B
A
A
I'm
gonna
ask
for
a
couple
of
a
couple
of
tweaks,
which
is
so
I'm
excited
that
you're
doing
this.
This
is
gonna,
make
the
site
a
lot
more
interesting
for
people
and
yeah.
If
you
browse
not
to
your
localhost
but
to
layer
find
out,
is
that
I
mean
maybe
others?
Okay
go
to
the
production
site?
If
you
would
like
you
know
in
the
address
bar,
it
would
be
layer
5
that
io
/
performance
yeah.
Yes,.
C
A
This
is
great.
I'm
excited
that
you're
doing
this,
moreover,
I
think
there's
a
lot
of
opportunity
for
small
SVG
animations
across
the
site,
for
those
of
you
who
browse
other
sites,
and
you
can
always
get
something
of
an
internal
sense
as
to
how
well
cared
for
a
given
site.
Is
that
you're
on
a
given
website
is,
and
it's
it
just
gives
you
a
few.
You
just
get
a
feeling
when
you're
on
in
any
site
by
K
is
this?
Is
this
thing
real?
How
nice
is
it?
Is
there
money
behind
it?
A
A
So
on
your
first
round
of
feedback
here,
when
you
go
to
layer,
five
dot,
io
/
performance
and
you
look
at
the
existing
SVG,
great
okay,
so
one
of
the
design
principles
of
I'm
sorry,
one
of
the
principles
of
design
is
to
is
consistency,
is
is
providing
an
intuitive
interface
and
intuition
it
and
having
a
user
interface,
be
intuitive,
is
in
part
driven
by
consistency,
and
so
by
that
I
mean
like.
If
you
hover
over,
to
learn
more
Lauren,
learn
more
about
the
service
master
specification.
A
When
you
hover
over
that
hyperlink
that
blue
hyperlink,
it
will
I
think
it
will
bold
emboldened
a
little
bit.
Is
that
right?
Yeah?
Yes,
okay,
I'm,
not
sure
if
yours,
if
your
page
is
updating,
but
for
me,
but
anyway
it's
a
very
it's
a
consistent
experience
that
when
you
hover
over
where
the
mailing
list
is-
and
it
says,
subscribe,
I
think
it
says
subscribe
when
you
hover
it
does
this
little
animation.
It
does
this
little
like
3d,
it's
like
you're
depressing.
A
The
button
subscribe
button
and
basically
that's
the
same
style
interaction
that
you
have
on
every
blue
colored
button
across
the
whole
site
and
as
a
user
you
come.
You
come
to
understand
that
hey,
if
you
click
that
a
those
things
are
clickable
and
B
when
you
click
it
like
an
action
is
going
to
be
taken,
sometimes
you're,
taken
to
a
new
page,
sometimes
or
like
almost
all
the
time,
you're
taken
to
a
new
page.
A
So
my
point
of
saying
that
is
I
think
some
of
this
is
common
sense
or
that,
if
you
were
to
think
about
it
long
enough,
you
would
say:
oh
yeah,
that
that
makes
sense
that
that
you
want
to
try
to
keep
you
a
similar
style
and
a
say
that
the
paradigms
of
interaction
that
people
have
you
want
to
keep
them
somewhat
consistent
across
the
site.
So
my
point
is
through
this
SVG:
it
don't
want
the
existing
one.
It
is
of
essentially
the
same
style
as
the
other
SVG's.
A
So
if
you
go
to
layer,
five
dot,
IO,
just
the
main,
the
main
Sun
and
we
take
a
look
across
the
buttons
and
the
small
icons
that
are
used
on
the
buttons
they
are
not
entirely
dissimilar
from
one
or
they
are
they're
done
in
such
a
way
that
actually
for
anyone,
who's
interested
like
the
it's
kind
of
intriguing
that
the
SPG's
that
are
on
these
buttons,
the
the
SVG's
themselves,
there's
only
one
color
used
in
those
in
those
images
but
Gunjan.
If
you
were
to
zoom
in,
if
you
did
a
ctrl
+,
+
+,.
A
Yes,
and
just
so
that
people
can
see
these
icons
a
little
bit
better
that
you
can,
if
you
can
see
it
guys,
you
can
tell
that
like
there's
a
portion
of
it,
that's
white,
but
there's
a
portion,
that's
slightly
gray
or
slightly
less
or
white.
What's
actually
happening
is
there's
actually
only
one
color
in
these
and
that
color
is
white,
but
the
reason
that
you're,
seeing
it
a
slightly
lesser
white,
is
because
the
opacity
for
that
particular
section
of
the
SVG
in
this
case
the
needle
the
gate.
A
You
know
the
needle
part
its
opacity
is
like
sixty
percent
or
something
like
that,
so
that
it
has
the
effect
of
not
just
being
kind
of
a
boring
flat,
white
thing
and
so
I
don't
know
I'm
telling
you
guys
all
this
is
just
as
a
point
of
interest
about
how
you
can
how
SVG's
are
kind
of
neat,
and
you
can
programmatically
control
its
color.
You
can
programmatically
control
the
stroke,
the
width,
the
other
things.
A
So,
actually,
when
you
go
to
the
performance
page,
that
same
SVG
is
used
literally
literally,
the
same
file
is
used,
but
what
we
do
when
we
load
up
the
page
is
that
you
can
use
a
filter,
a
CSS
filter,
to
change
the
fill
color
for
that
SVG
from
white
to
black,
so
gungeon.
This
is
kind
of
a
long
way
of
me
saying
what
would
be
ideal.
A
What
would
be
ideal
is
if
you
were
able
to
take
the
same
SVG
and
and
animate
its
needle
a
little
bit.
Actually,
the
ideal
effect
is
that
it
doesn't
move
by
default
because
actually
I
don't
know
how
many
of
you
have
ever
watched
a
PowerPoint
presentation
from
someone
and
they're
going
through
the
presentation
and
like
they've,
used
every
animation
in
the
book.
A
But
if
someone
either,
if
someone
hovers
well,
maybe
it
may
be
a
different
inert,
maybe
when
the
page
loads,
but
probably
ideally
if
they
hover
over
the
icon,
that
the
needle
would
would
go
from
like
it's.
It
know
70%
where
it
is
now
to
like
100%
and
then
when
they
move
their
mouse
off.
So
when
it
lost
beyond
focus,
it
would
just
go
reset
back
to
where
it
was
or
something
similar
to
that
it
doesn't
exactly
have
to
be
where.
A
C
Yeah
yeah,
hello,
awesome,
one
more
thing:
I
want
to
talk
about
God
in
this
talk
shut
up
so
when
I'm
I
was
visiting
this
site
and
so
I
go
go,
went
to
some
other
proper
application,
talker
application,
so
in
this
Asian
are
too.
We
need
to
mention
all
the
acts
that
will
be
I,
think
that
will
be
convenient
for
other
users
to
use
the
particular.
C
A
A
C
A
I
think
it's
conceivable
to
do
the
thing
you're
right,
that
improvement
needs
to
be
done
here.
The
thing
that
would
actually
be
much
more
helpful
to
our
users,
but
before
we
get
to
what
you
just
suggested,
would
be
would
be
versioning
of
each
of
the
docker
images,
so
we're
the
project
itself.
Mastery
is
in
a
little
bit
of
a
bad
spot.
Right
now,.
A
A
Actually,
we
produce
a
container
image
every
time
that
someone
pushes
a
commit
to
any
of
the
repos
or
we
create
an
image,
but
we
only
ship
it
to
docker
hub
when
PR
that
when
that
commit,
has
been
included
in
a
pull
request
and
that
pull
request
has
been
merged,
then
that
image
is
pushed
to
dr.
hub.
The
problem
with
what
we're
doing
there
is
that
we're
is
that
there
isn't
a
difference
between
the
latest.
A
Is
that
any
time
that
we
make
a
merge
do
emerge?
The
latest
image
goes
up
on
docker
hub.
It's
tagged
with
the
latest
tag
and
immediately
anyone
who's
running
measuring
can
be
the
recipient
of
that
new
container
image,
because,
when
we're
running
missery,
when
you
guys
are
running
mesh
around
your
local
machines,
there's
an
image,
there's
a
docker
container
and
it's
running,
it's
called
watch
tower.
A
It
watches
every
so
often
it
reaches
out
to
doctors
up,
and
it
looks
for
a
new
image
of
fresh
image
if
it
finds
a
fresh
image
of
mesh
reor
any
of
the
measuring
adapter
containers,
it
will
pull
them
down
and
it
will
update
your
local
running
environment
on
one
hand,
that's
really
good
because
you
just
get
continuous
updates,
but
on
the
other
hand,
it's
bad,
because
what
the
project
today
is
not
offering
people
is
version.
Releases
really
anand
knows
this.
A
Well,
because
really,
the
only
thing
best
versioned
in
the
project
is
measure
ECT
and
ma
cherie.
Ctl,
it's
release
the
release
of
measure.
Ctl
is
controlled.
It's
controlled
through
tags
that
are
done
in
github,
but
that
isn't
the
case
for
docker
images.
We
we
publish
a
new
version
of
each
of
the
docker
images
every
time
that
emerge
is
done
on
any
of
the
repos
and
that's
okay,
if
that's
like
an
edge
channel,
if
that's
like
the
beta
channel,
but
it's
not
okay
for
like
a
stable
release
channel.
A
It's
something
that
we
tried
to
get
a
few
different
contributors
to
help
with,
and
people
have
been
interested
but
but
I
think
they
get
I.
You
know
it's
it's
a
hard
one
and
no
one
has
delivered
it
yet
I'm,
not
suggesting
that
guns,
and
you
should
take
it
on
I'm,
just
saying
hey.
Actually,
instead
of
doing
that,
work
that
you
were
describing
look,
there's
some
work
in
advance
of
that.
That
would
be
a
lot
more
important
or
a
lot
more
helpful.
A
A
A
So
I'm
gonna
give
since
we
were
just
talking
about
docker,
desktop
I'm
gonna,
share
an
example
of
what
I'm
talking
about
I.
Think
I.
Think
you
guys
I'll
get
it
I'm.
Just
so
I'm
running
docker,
desktop
I'm
running
a
Mac
I
go
to
the
about
screen
and
actually
the
channel
that
I'm
using
for
docker
desktop.
A
We
use
yeah
a
non
totally
you're
on
the
right
path
about
like
using
dark
container
tags
as
a
way
of
facilitating
different
release
channels
and
to
the
to
the
extent
that
it's
of
interest
to
anyone
there
is
a
design
document
inside
of
the
mesh
receipt.
Eeeh
I'm.
Sorry,
inside
of
the
measure,
II
team
drive
there's
a
design
document
that
talks
about
how
we
would
go
about
creating
you
know,
leveraging
tags
and
some
other
things
to
go
about
creating
different
channels.
A
A
A
D
Okay,
so
I
was
just
assigned
the
issue
today.
So
I
was
looking
into
this
looking
in
the
61
issue.
I
guess
it
was,
it
is,
is
just
changing
the
config
file,
the
path
of
the
config
file
and
we'll
be
fine
in
this
one,
but
regarding
to
the
issue,
number
36,
which
is
replacing
the
metal,
is
tio
installation
with
elm
I
guess
in
the
installation
guides
of
is
tio.
D
A
That
said,
if
they're
not
gonna
be
supporting
home
and
and
those
charts
won't
be
maintained,
then
it's
not
a
good
direction
to
go
and
that's
somewhat
unfortunate
because
it's,
it
means
more
work
as
we
support
in
Mishra
CTL.
What
would
be
of
interest
and
what
might
be
fairly
easy.
But
what
would
be
of
interest
is
is
if
there
is
a
go
client
from
a
sri
CTO
or
if
there's
a
way
to
import
the
measure.
Ect
our
package
and
then
leverage
that
package
or
if
they
have
a
pre-written,
go
client
that
we
can
use.
B
A
E
A
Alright,
sto
client
go
client,
go
repository,
nice
access
to
the
sto
api's
in
a
kubernetes
environment,
the
generated
crew,
Nettie's
and
formers
and
client
setting.
This
repository,
make
it
easy
for
developers
to
create
controllers
and
perform
crud
operations
for
all
sto
customer
resource
CID.
So
that's.
This
is
good,
because
this
is
something
that
we're
gonna,
we're
going
to
generally
want
to
do
anyway.
A
D
A
Universal
well,
as
a
matter
of
fact,
there
are
operations
that
a
mesh
readapt
err
should
perform
and
many
of
those
are
universal.
There
are
some
of
those
that
may
not
be
implemented
that
we
may
not
implement
for
a
certain
service
mesh,
and
that
might
be
because
it's
inappropriate
for
that
service
mesh
or
it
might
be
because
we
didn't
get
to
it
yet
or
you
know.
A
This
is
like
a
template
eyes
thing:
yeah,
yeah,
you're,
right,
okay,
so
so,
but
then
specifically
like
as
you
look
at
the
liquidy
adapter
and
you
look
to
do
the
linker
D
installation
through
helm.
That
is
a
good
example
of
reusable
code
like
installing
a
service
mesh
through
helm
or
installing
a
sample
app
through
helm.
That
happens.
That
happens
a
lot
and
is
potentially
you
know,
that's
highly
reusable
code,
the
what
won't
happen
a
lot
or
rather,
what
might
happen
more
than
once
would
be
the
use
of
a
CTA.
A
A
It
might
have
a
link
or
D
go
client
as
well
like
a
link
or
de
CTL.
If
you
will
go
client
the
it's
a
common,
no
common
pattern.
Unfortunately
I
don't
know
yeah,
you
could
have
a
reusable
interface,
but
the
code
behind
it
is
unfortunately
still
gonna,
be
client,
specific
or
CTL
specific,
so
keep
iterating
on
keep
thinking
about
that.
Maybe
take
a
look
through
this
adapter
template
repo
understand
you
know
which,
which
type
of
you
know
understand
what
this
one
is
all
the
while,
if
you're
able
to
get
traction
on.
A
A
A
A
A
A
Whatever
that
is,
if
that's
running
on
one
of
these
meshes,
then,
ideally,
this
interface
is
ideally
measure.
Ii
is
aware
of
that
mesh.
We
says
well
yeah,
I'm,
aware
that
you're
running
you've
got
sto
deployed
and
linker
D
deployed,
which
I
do
in
my
environment.
I
also
have
other
adapters
connected
the
citrix
adapter
connected
the
network
service,
mesh
adapter
and
the
octarine,
but
I'm
not
running
instances
of
those
service
meshes
I
am
running
an
instance
of
both
sto
and
like
Addie,
and
so
there's
a
an
endpoint.
That's
called
that
will
I
believe
it's
just
scan.
A
A
Kubernetes
uses
the
kubernetes
go
client
to
call
out
to
kubernetes
and
then
look
for
any
deployed
service
match
and
if
it
finds
it
looks
for
the
certain
service
messages
and
if
it
finds
them,
it
tries
to
pull
back
their
version
number
and
so
that
communities
glow
client
the
the
results
of
it.
That
scan
are
exposed
through
this
endpoint
and
then
the
UI
and
then
in
the
UI.
A
D
A
A
Sin
Devine:
this
is
great
man
I'm
excited
that
you're,
looking
at
the
sto,
adapter
and
and
enhancing
naveen
there's
actually
a
specific
issue
that
we're
facing
right
now,
and
it
is
that,
as
you
look
through
the
adapter
I'm,
pretty
sure
you're
gonna
find
that
we
have
hard-coded
or
pinned
the
version
of
sto.
That
measure
e
is
deploying
I,
think
we've
pinned
it
to
like
1.3
dot,
o
and
I.
A
Yes,
it
was,
and
that's
because
I
think
the
file,
the
the
sto
distribution
package
to
some
of
the
llamo
files,
their
their
file
names,
have
changed
and
Shubham
had
actually
run
into
this
issue.
In
the
workshop
that
we
have
and
anyway
the
adapter
hasn't
been
updated
and
so
right
now
it's
pinned
at
an
older
steel
version
which
that's
not
good.
So
as
you're
going
through
there.
That's
something
else
to
take
a
look
at
if
you
were
yeah.
B
Oh,
my.
D
B
A
A
A
A
A
A
Images
and
code
and
things
around
in
this
sites-
and
so
this
is
a
good
reminder
to
me
as
well
to
like
of
how
this
is
controlled
and
for
anyone
else,
who's
contributing
on
the
site.
So
this
is
the
layer
5
that
IO
site
we
go
to
collections,
there's
a
collection
of
like
prominent
pages,
a
page
like
the
community
page
I.
Don't
think
it
actually
has
a
an
image
that
we're
sharing
and
that's
oh
actually,
it
does
it's
this
image
here,
it's
just
the
generic
community.
It's
a
layer,
5
logo
with
the
community
tag,
tagline.
A
A
A
This
thumbnail
is
being
used
differently.
This
is
just
a
custom
piece
of
frontmatter
that
I
had
created,
because
when
you
go
to
layer
five
that
io
/
projects,
instead
it
needed
to
be
a
white
SVG,
not
a
green
SVG.
I'm.
Sorry,
that's
not
that's
not
true.
It
needed
to
be
an
SVG,
whereas
the
social
networks
they
don't
like
SVG's,
they
like
p.m.
geez,
so
anyway,
I
saw
Anand,
yeah,
very
good,
maybe
Anand.
F
A
Okay,
got
it
yeah
this
one.
Isn't
this
one
isn't
horrible
to
be
like
if
someone
would
just
share
link
to
my
Sridhar
il
this
is
a
decent
image
that
sort
of
represents
what
the
project
is.
It's
not
bad.
It
is
slightly
outdated,
but
yeah
equally
as
good
would
be
an
image
of
the
in
one
of
the
logo.
Images.
F
A
I
see,
okay,
sure
it
makes
sense
because
yeah
this
this
one
is
not
like
late
later
on.
If
we
were
to
have
I,
think
yeah,
if
we
were
to
get
the
dots
site
over
and
someone
were
to
go
to
like
slash
management's
or
something
that
might
make
sense
to
show
that
screenshot,
but
to
your
point
in
general,
if
someone's
just
pointing
to
the
root
mass
video
showing
the
logo
is
probably
most
appropriate.
A
That
was
having
us
include
the
use
of
sudo
here,
which
you
know
can
be
scary
for
a
lot
of
people
to
have
sudo
to
just
execute
this
command
on
their
system,
because
they
don't
necessarily
know
what
all
is
gonna
be
run
by
this
script,
and
if
it's
asking
for
root
level,
privileges,
it's
kind
of
scary
and
so
yeah
Anand
I,
think
we're
all
done
what's
happened.
Now.
Is
that
whoops?
Let
me
go
grab
that
script
what's
happened
now.
A
Is
that
the
point
in
time
in
the
installation
script,
where
we
were
running
into
the
permission
issue,
as
you
well
know,
is
when
we
were
moving
mesh
Reis
ETL
into
what
is
sometimes
a
protected
folder
on
Linux
systems,
and
so
I
think
it's
this
check
here.
It's
sort
of.
If
this
move
is
successful,
you
know,
then
you
know,
then
we're
good.
A
If
not,
then
let's
go
ahead
and
attempt
to
sudo
and
and
basically
we'll
be
asking
for
user
permission
to
do
permissions
right
then,
and
in
this
way
we're
only
asking
for
sudo
permission,
root
level
permission
when,
as
and
when
needed,
which
is
sometimes
the
case
and
sometimes
not
the
case
depending
upon
the
user's
system
and
who
the
user
is.
Sometimes
the
users
are
just
running
as
root
themselves,
and
so.