►
From YouTube: Layer5 Community Meeting (Dec 10th, 2021)
Description
Layer5 Community Meeting - December 10th, 2021
Join the community at https://layer5.io/community
Find Layer5 on:
GitHub: https://github.com/layer5io
Twitter: https://twitter.com/layer5
LinkedIn: https://www.linkedin.com/company/layer5
Docker Hub: https://hub.docker.com/u/layer5/
A
Welcome
everyone
to
the
layer,
5
community
meeting
today
is
the
10th
of
december,
and
we
had
a
lot
of
folks
join
in
on
the
community
this
week,
so
welcome
to
all
the
newcomers,
and
I
think
we
have
some
some
some
new
folks
in
with
us
on
the
call
today
so
patel.
Oh,
I
think
khardik
is
connecting
to
the
audio,
but
let
me
see
shivayan.
I
think
you
are
in
this
call
for
the
first
time.
So
would
you
like
to
introduce
yourself
to
the
community.
B
A
Awesome
thank
you
for
introducing
it's
really
great.
C
Can
you
add
your
last
name
in
slack
or
in
zoom
if
you
would
yeah.
A
Awesome
and
going
down
the
list.
A
Sudya
suryaganta
sahoo.
Would
you
like
to
introduce
yourself.
B
E
E
I've
contributed
not
into
so
much
into
open
source.
Still
a
beginner.
A
Awesome
did
you
by
chance
get
an
opportunity
to
join
in
on
the
yesterday's
newcomers,
call.
A
One
is
my
first
ball:
okay,
so
yep
we
had
some
sessions
on
yesterday's
newcomers
call
but
don't
worry
like
we
do
have
recordings.
We
do
record
all
of
our
community
meetings
and
they
are
made
public
on
youtube
and
yeah.
Hopefully
we
can
help
you
with
some
resources
so
that
you
can,
you
can
contribute
yeah.
I
looked
up
the
resources.
A
Nice
to
hear
that
awesome
welcome
to
the
community,
and
it's
really
great
to
have
you
here:
yeah
thanks
thanks
a
lot
for.
F
Yeah
hi
yeah
sure
hi
everyone.
I
am
ankita
pre-finally
a
student
at
iit,
roorkee,
so
yeah.
I
was
looking
for
some
open
source
projects
in
the
devops
site
because,
like
I
really
wanted
to
explore
cloud
and
stuff,
I
had
a
previous
experience
in
backend
and
front-end
and
mobile
development
majorly,
but
yeah.
This
was
one
side
that
I
really
wanted
to
explore
and
then
I
got
to
I
got
to
that
ripple
of
misery.
So
yeah
I'm
looking
forward
to
contributing
to
that.
A
That
sounds
really
great,
ankita
nice
to
have
you
here
and
did
you
get
the
chance
to
go
through
some
of
the
projects
and
did
you
see
anything
that
interests
you.
F
A
Sure
sure
sure
also
we'll
try
to,
I
think,
like
you,
will
get
an
onboarding
set
of
documents
when
you
join
the
community
and
sign
up
on
the
newcomers
form.
So
did
you
get
a
chance
to
yeah.
D
A
Awesome
so
yep
you
can
check
some
of
those
resources
and
hopefully
we
will.
We
will
cover
some
areas
where
people
can
get
involved
on
today's
call
so
yep.
So
thanks
again
for
introducing
and
welcome
to
the
community.
G
Yeah,
hello,
everyone,
I'm
hadik,
I'm
currently
studying
in
triplet
delhi
and
doing
computer
science
there,
and
it's
my
currently
second
year
like
and
the
way
the
reason
I
signed
up
for
layer,
five.
It's
just.
I
keep
hunting
for
different
opportunities
and
I
don't
know
if
I
have
the
skill
set
or
not.
I
just
first
enrolled
myself
with
opportunity
and
then
I
just
later
see
if
I
can
develop
in
it
or
not.
Yeah.
C
Nice,
hey
I'm
going
to
mention
a
couple
of
things:
real,
quick
just
because
there's
a
few
newcomers
on
the
call
and
some
of
us
that
might
need
a
reminder
or
a
reminder
might
serve
well.
So
by
the
way,
my
name
is
lee
so
nice
to
meet
some
of
you
that
are
on
for
the
first
time,
there's
two
resources
that
I
would
point
out.
Actually
three
resources
that
I
would
point
out.
One
is
the
this
discussion
forum.
It's
disqus.layer5.io
is
the
link.
C
C
There's
a
few
community
managers
that
are
on
the
call
now
and
they
keep
abreast
of
all
the
going
zones
they're,
also
quite
familiar
with
the
spread
of
projects
that
we
have
and
the
technologies
that
are
used
in
them
and
can
really
help
orient
you
very
well,
as
a
matter
of
fact,
there's
a
number
of
people
who
dedicate
their
time
every
thursday
for
an
hour.
There's
a
continuous
newcomers
call
the
one
that
novendi
was
just
mentioning
that
you're
all
welcome
to
attend
multiple
times
like.
C
I
would
encourage
you
not
to
attend
just
once,
but
many
times
and
if
you're
around
here
long
enough,
like
actually
it's
a
great
call
to
present
on
to
like
share
a
few
things
that
you've
learned
on,
believe
it
or
not.
There's
a
lot
of
people
out
there
that
have
the
same
questions
as
you.
C
How
do
I
sign
off
on
a
commit
that
I've
already
made
like
those
might
sound
like
easy
questions?
I
don't
know
they
might
sound
like
really
difficult
questions
either
way.
I
guarantee
you
they're
difficult
for
a
lot
of
people,
so
just
random
examples
of
like
sharing
with
others,
but
if
you
can't
tell
in
this
particular
community,
everybody
here
is
learning.
Some
of
us
are
learning
at
different
rates.
Some
of
us
are
learning
different
at
different
levels,
but
we're
all
learning,
which
generally
means
that
we're
all
making
some
mistakes.
C
I
would
tell
you
if
you're
not
making
some
mistakes,
then
you're
not
doing
it
right
and
you're,
not
learning
correctly
or
you're,
not
learning
fast
enough,
so
go
make
a
mesh
of
things
and
learn
some
stuff,
while
you
do
like
contribute
and
help
out
your
fellow
community
member
share
what
you
know.
This
is
a
great
place
to
share
what
you
know
and
to
ask
some
questions.
There's
a
one
of
the
significant
reasons
why
we're
using
this
particular
forum
is
because
the
slack
history
lots
of
activity
goes
on
in
slack.
C
It's
very
compelling
to
ask
your
question
in
slack.
The
problem
is
it
all
gets
lost
over
time?
There's
a
lot
of
people
have
the
same
question.
So
please,
please
make
use
of
this
forum
to
the
extent
that
you
can
so
that's
one
resource.
I
wanted
to
point
out,
as
I
point
that
out
I
want
I'll.
I
will
actually.
I
won't
show
this
next
thing,
but
I'll
put
I'll
ask
balmy
or
deepak
or
delopria
or
someone
else
who's
on
the
call
later
in
the
call.
C
If
someone
could
demonstrate
the
new
custom
slack
command
for
interacting
between
slack
and
this
disqus
forum,
that
would
be
really
great.
I
think
that's
kind
of
a
neat,
a
neat
thing
out
there,
so
for
all
of
you
that
are
relatively
new
and
for
all
of
you
that
are
old-timers,
spend
some
time
on
here.
We've
got
a
few
questions
unanswered,
so
we
award
a
million
dollars
at
the
end
of
the
month
for
those
who've
as
asked
the
most
questions
and
two
million
dollars
for
those
that
have
answered
the
most
questions.
C
Okay,
we
don't
actually
do
that,
but
but
we
will
say
thank
you
with
a
big
smile,
but
we'll
tweet
your
name
out
we'll.
Do
we
do
a
bunch
of
stuff
with
people
we
generally
generally
harass
people
too
so
watch
out
all
right,
the
other
resource
that
I
encourage
you
all
to
go.
Take
a
look
at
there's,
there's
a
bunch
of
resources,
but
one
of
them
is
the
community
handbook
which
is
out
on
layer
five
that
I
owe
this
website
layer.
5.
io
is
open
source.
C
It's
built
by
about
240
different
people,
thus
far
so
you're
welcome
to
come,
contribute
to
it.
One
of
the
recent
additions
has
been
a
community
handbook
which
is
still
in
the
process
of
being
refined
and
still
in
the
process
of
being
linked
to.
Actually,
since
I
had
to
manually
type
in
the
link,
but
there's
a
contributor,
josh
who's
actually
been
extremely
patient
on
and
shown
a
lot
of
perseverance.
C
I'm
trying
to
create
a
few
links
to
this
so
but
anyway,
great
resource
in
here
last
resource
that
I
wanted
to
point
out
was
hardik
and
I
were
just
having
a
quick
conversation
about
ramping
up
an
open
source
and
and
doing
the
open
source
thing
and
getting
some
pointers
where
to
get
going.
You
know,
just
you
know,
like
there's
all
kinds
of
things
to
talk
about.
There's,
there's
pro
mentorship
programs
like
this,
and
I
have
to
say
who
who
here
is
on
the
call
is
and
mesh
mate
anita
with
us.
C
Well
to
my
point:
we
need
more
message:
mates
we
actually
do.
We've
had
a
few
graduate
and
become
mesh
mate
emeritus
and
that's
leaving
us
a
little
bit
of
a
gap.
What
I
was
going
to
say
is
I'm
dumbfounded,
appalled,
horror-stricken.
C
C
I
don't
know
why
it's
not
right
here
anyway,
there's
a
program
for
questions
like
the
the
good
ones
that
hardik
was
bringing
up,
and
that's
the
fact
that
there's
currently
these
four
individuals
in
the
community
who
spend
a
lot
of
time
helping
helping
with
those
types
of
questions-
and
these
are
called
mesh
mates.
C
C
C
Please,
for
the
love
of
of
something,
if
you're
going
to
communicate
in
a
written
fashion,
please
take
a
moment
to
be
kind
to
those
that
you're
communicating
with
and
like
add
some
punctuation,
maybe
a
capital
letter
at
the
start
of
the
sentence,
maybe
a
period
even
an
exclamation
mark.
Sometimes
at
the
end
you
know
put
a
toss
on
a
comma,
a
little
prepositional
phrase.
C
You
know
any
some
of
this
stuff,
you
don't
have
to
be
a
whiz,
but
just
like,
please
don't
be
lazy
and
part
of
that
is
like
being
lazy,
is
one
more
call,
there's
a
little
over
2
000
folks
that
are,
in
the
slack
lots
of
people
super
interested
in
the
work
that
goes
on
here.
C
A
few
of
the
people
that
are
on
the
call
we're
giving
talks
at
kubecon,
four
of
them
this
week
on
the
works
that
we
do
some
really
interesting
stuff
and
so
there,
but
there
are
only
a
few
maintainers
like
there's
a
of
the
ratio
to
of
maintainers
to
contributors
that
have
questions
is
upside
down
or
is
like
is
not
a
great
ratio.
C
We
want
to
change
that
as
a
matter
of
fact,
some
of
you
might
become
maintainers
someday,
and
that
would
be
fantastic
like
I
can
tell
you
the
other
maintainers
that
are
on
the
call.
They'd
love
for
you
to
become
a
maintainer
so
like
go,
do
it
go
get
it,
but
what
I
was
trying
to
say
is
that
please
don't
be
lazy
as
you
ask
questions,
do
your
homework?
Do
some
reading?
Don't
don't
make
people
spoon
feed
you
or
you
might
be
politely
ignored.
C
It's
like.
Don't
don't
make
it
hard
on
the
maintainers.
They
already
have
a
tough
job
and
so
yeah,
but
reach
out
to
a
mesh
mate
grab
one
and
for
those
that
have
been
around
for
a
while.
A
We
do
have
a
to-do
list.
We
try
to
adhere
to
this
to-do
list
where
community
members
can
ask
for
different
tutorial
topics
and
we
try
to
cover
those
on
the
newcomers
call.
So
yesterday.
So
last
week
we
had
gerard
mentioned
mentioned.
He
wanted
to
contribute
to
measure
recital,
and
so
in
this
week
we
we
we
had
a
walkthrough
on
contributing
to
measuring
ctl,
basically
setting
up
the
environment,
running,
building,
measuring
locally
and
running,
building,
measuring
ctl
locally
and
all
those
things.
So
such
sessions
definitely
have
definitely
helped
new
contributors.
A
I
mean
get
get
get
better
quickly
and
get
start
start
start
contributing
quickly,
so
do
check
those
out
and
don't
hesitate
to
add,
add
your
own
topics
here,
but,
as
we
mentioned,
make
sure
that
you,
you
do
your
homework
and
there
are
a
lot
of
resources
that
you
get
access
from
the
layer,
5
community
drive
so
make
sure
you
go
through
that
and
try
answering
your
own
questions.
A
First,
before
you
start
answering
your
questions,
and
also
since
there
are
a
lot
of
you
who
are
new
on
the
call
to
point
out
the
impact
of
the
work
we
do
here,
as
lee
mentioned,
we
had
a.
We
had
a
session
on
cubecon
china
a
little
bit
earlier
today,
so
we
were
just.
We
were
talking
about
all
the
work
that
we
have
been
doing
in
this
community.
Indie
projects
missionary
and
resource
smash
performance,
and
we
were
also
represented
at
at
the
cncf
tag
network
sessions,
so
yeah.
So
there
are.
A
There
is
a
lot
of
representation
of
what
work
we
are
doing
within
cncf
and
within
the
cloud
native
ecosystem,
so
it
is
really
impactful
and
there
is
definitely
a
lot
of
room
for
more
people
to
come
in
and
make
an
impact
here
all
right.
A
So,
let's
so
moving
on
to
our
second
topic,
for
today,
we
we
thought
it
would
be
good
to
look
into
some
areas
where
contributors
can
get
involved.
So
one
of
our
main
focus
main
focus
right
now
is
on
measuring,
so
measuring
measuring
is
the
maturity.
A
Is
one
of
the
one
of
the
projects
that
we
work
on
on
the
community
and
one
of
the
most
popular
ones,
and
what
we
have
been
trying
to
do
right
now
is:
is
we
are
working
towards
the
v0.6
release
and
I'll
just
put
this
on
top,
so
we
have
been
trying
to
focus
on
quality,
so
measuring
is
almost
halfway
there
from
the
one
dot
or
release,
so
we
are
trying
to
drive
towards
making
measuring
more
more
quality,
checked
and
a
more
rugged
product.
A
A
You
find
any
anything
that
any
ux
changes
that
you
suggest
or
any
feature
features
that
are
missing,
so
that
would
be
a
really
great
way
to
contribute,
and
if
you
especially
if
you
are
really
new
to
the
project
and
if
you
haven't
used
it
before
because
most
of
us
who
have
been
here
for
for
a
long
time
are
really
too
close
to
see
some
of
the
issues
that
has
been
creeping
in
and
maybe
like,
we,
we
have
some
prior
beliefs
that
that
that
mine,
it
might
need,
might
need
to
be
changed.
A
So,
yes,
we
would
like
to
so.
This
is
one
of
the
most
impactful
ways
that
you
can
get
involved
and
if
you
go
so
the
the
screen,
I'm
sharing
is
a
test
plan.
So
I
just
share
the
link
in
the
chat,
so
we
have
a
different
test
scenarios
written
on
this
sheet.
A
C
Let
me
I'm
gonna
interrupt
just
for
a
moment
if
I
could,
just
because
I
think
that
people
kind
of
get
lost
as
we
talk
about
a
lot
of
this
stuff
so
like
very
directly,
if
you're
here
and
you
don't
know,
golang
you're
looking
to
contribute
to
measuring
great
talk
to
nuventu,
there's
a
lot
of
things
to
test,
be
a
power
user
and
test
some
things.
C
So
for
mr
sahu,
this
is
a
great
place
to
jump
in
yeah
at
being
a
user
of
the
software
is
a
great
thing
to
do.
You
might
think.
Well,
that's
not
a
good!
C
Cypress
is
today's
the
new
age
j
meter
or
it's
it's
a
modern
tool
for
doing
end-to-end,
functional
tests,
driven
from
web-based
user
interfaces
and
mario
arriga
who's
on
the
call
is
well
seasoned
in
this
regard.
C
So
he's
a
great
individual
to
like
this
is
another
great
way
to
contribute
to
mashri
as
a
project
and
come
come,
get
yourself
an
internship.
If
that's
something
you're
looking
for
or
a
job
or
and
become
a
cypress
expert
like
we
really
really
need
it,
and
you
you'll
definitely
definitely
learned
a
number
of
things
in
doing
so.
C
To
answer
mario's
question:
there's
a
channel
in
the
slack
it
is
meshri
hyphen,
ci
ci
for
continuous
integration.
That's
a
it's
a
fairly
appropriate
channel
to
have
cyprus
oriented
conversations
within,
but
if
it
doesn't
feel
like
a
good
fit,
don't
let
that
stop
you
from
cracking
open.
Another
channel.
C
G
E
I
have
one
I
have
one
question
like
for
these
tests
like
do
we
have
to
create
these
test
scenarios
or
like
these
are
like
some
are
assigned
to
someone
and
like
some
are
open
so
like
we
can
work
on
them
or
like
we
have
to
create
new
ones.
H
C
Yep
yep
mario,
I
missed
the
name
of
the
individual
that
asked
it,
but
I'm
not
gonna
repeat
it
for
him.
If
he
wants
to
know
the
answer,
yeah.
E
H
Oh
okay,
yeah
well!
Well!
If,
if
we
take
a
look,
we
can
see
that
there's
different
kinds
of
tests
both
or
you
know
for
the
different
components,
measuring
ctl,
measuring
server
and
measuring
ui.
If
we
take
a
look
at
let's
say
the
second
column,
we
see
that
the
first,
maybe
100
might
be
for
other
components
that
are
not
like
for
the
ui,
but
once
you
get
to
the
yup
that
one
yeah
you
see
measuring
ui,
it's
ui
measuring
server,
there's
different
tests
and.
H
Totally
you
can
pick
any
up,
there's
a
specific
issue
that,
where
we're,
where
we're
tracking
in
github
and
ms.
H
Measure
is
github
repo
and
that
that
issue
is
to
track
cyprus.
Integration
tests,
yeah,
please,
let's
just
take
a
quick
look.
There's
to
be
honest,
it's
not
like
that.
It's
something
it's
the
way.
It
is
right
now,
there's
not
like
a
great
or
simple
way
to
track,
which
let's
say,
which
you
know
that
that
other
spreadsheet,
like
a
really
easy
way
to
you,
know
see
which
area
is
for
which
issue
or
like
that's
an
umbrella
issue.
H
But
if
you
know
the
idea
is
to
we
can
just
our
coordinate
and
if
you're
interested
in
contributing
for
now,
you
can
feel
free
to
send
up
a
message
in
in
slack's
in
in
the
ci
channel,
and
then
we
can
get
you
started
in.
H
H
But
if,
if
people
interested
in
in
cyprus,
testing
can
just
subscribe
to
the
slack
channel,
I
think
it'd
be
good
so
that
we
can
identify
who
wore
like
a
avid
at
cypress,
coordinate,
work
on
that
and,
and
also
if
there's
people
that
are
really
new
to
cyprus,
but
interested
in
functional
and
integration
testing.
With
this
tool
and
for
measuring.
H
We
can
just
provide
a
set
of
ideas
to
identify
like
if
learning
areas
where,
where
we
can
a
both
share
knowledge,
but
also
build
some
resources,
like
there's
already
some
resources,
like
a
blog
where
a
newcomer
could
just
go
through
and
understand
how
to
run
and
work,
cyprus,
tests,
locally
yeah
and
for
now,
as
lee's,
writing,
there's
this
spreadsheet
to
track
like
overall
testing.
H
But
if
you
see
there
should
be
some
field
where
yeah,
if
it's
automated
or
not-
and
also
there
should
be
like,
is
there
like
an
owner
or
a
signing?
It's
I
think
it's
the
first
like
you
go
to
the
beginning
of
yeah,
the
tester.
H
If
it's
empty
and
you
see
it's
for
ui,
you
can
just
raise
your
hand
on
on
the
slack
channel
or
on
the
github.
The
github
umbrella
issue
for
like
cypress
testing
and
just
very
quickly-
I
I
just
like
just
to
explain
so
so
when
we're
talking
about
an
integration
test,
we're
basically
writing
a
test
where
the
only
thing
running
in
the
agent
is,
or
that's
just
the
ui
and
both
the
only
the
ui
and
the
provider.
H
So
there
are
two
different
ui
projects,
both
written
in
react
and
some
other
tools.
H
So
whenever
there's
a
request
sent
to
the
server,
what
we
do
is
that
we
tell
cyprus
to
to
return
a
canned
response,
and
you
know
you
can
just
take
a
look
at
our
cypress.
H
Basically,
in
measure
project,
you
could
go
to
the
to
the
ui
folder
and
then
to
the
cypress
folder
under
that,
and
you
know
I
suggest
like
if
someone's
very
new,
just
follow
the
the
the
newcomer
documents
but
once
you're
out
of
that
you're
in
the
swerving,
the
ui.
H
What's
this
folder
yeah
ui,
so
cyprus,
uh-huh,
and
so
the
fixtures
folder
is,
is
where
we
have
like
different,
like
mock
responses
or
stop
responses,
and
we,
we
also
define
other
a
test
artifacts
that
we
use
on
the
testing.
H
And
so,
if
we
go
and
also
if
we
go
into
the
integration
folder,
the
difference
between
integration
and
intuit
is
that
that
integration
will
mark
the
server
responses
and
e3
is
something
that
goes
against
both
server
and
ui,
just
like
either
a
couple
or
no
marked
dependencies,
and
the
idea
is
to
run
different
tests.
You
know,
aha,
also,
okay,
so
italy,
I'm
just
looking
at
at
the
chat
so
end
to
end
means
it's
something
from
the
user
perspective.
H
This
could
be
defined
as
user
acceptance
testing
in
the
context
of
the
ui,
we're
simulating
an
actual
end
user
going
through
the
ui
and
doing
different
actions
to
ask,
for
you
know,
acceptance
criterias
or
you
know,
requirements
use
cases
and
so
on,
oh
yeah
and
and
lee
shared
like
community
newcomers
tutorials.
H
So
there
you
can
find
some
cypress
details.
Please
go
ahead
and
if
so
I'll
try
to
click
I'll,
create
a
slack
channel
for
cypress
and
I
think
that'll
be
like
a
very
good
way
to
like.
So
anyone
can
join
and
contribute
and
we
can
coordinate
like
specifics
of
of
these
efforts.
H
I
think
that
be
very
helpful.
A
Yeah,
so
there
are
also
other
areas
that
you
can
get
involved
in
so
nitish
like.
Would
you
like
to
yeah.
C
I
Okay,
I'm
audible,
yep,
okay,
so
recently
we
added
some
end-to-end
tests
in
all
of
our
adapters,
so
what
these
end-to-end
tests
do
is
basically,
so
we
have
adapters
and
adapters
do
a
lot
of
things
and
one
of
those
things
is
to
manage
the
life
cycle
of
a
service
mesh
and
what
an
adapter
should
at
least
do
is
deploy
a
service
mesh
and
not
fail
to
do
that.
So
what
these
end-to-end
tests
do
is
that
whenever
someone
makes
a
pull
request
and
on
each
commit,
we
basically
run
this
core
workflow.
I
What
it
does
is
it
deploys
your
it
deploys.
A
cluster
deploys
a
service
machine
that-
and
this
is
very
configurable,
so,
for
example,
if
we,
if
we
are
using
it
for
steel,
we
can.
We
know
that
the
pods
of
these
two
control
plane
are
like
this
theorem
and
the
issue
ingress
and
is
theory,
so
we
can
make
certain
assertions
through
it
and
these
stats.
These
tests
can
basically
tell
us
that,
yes,
even
after
this
commit,
we
can
the
this
particular
adapter
would
still
be
able
to
deploy
a
service
mesh.
I
So
nothing
is
breaking,
and
this
is
important
because,
because
we
have
faced
this
issue,
a
lot
that
something
breaks
and
it
goes
under
look
so
these
end-to-end
tests
really
help
in
that
and
we
have
integrated
these
workflows
and
almost
all
the
adapters.
Maybe
one
two
are
left
for
some
given
reason,
but
yeah
after
a
while-
and
we
can
extend
these
functionalities
to
basically
make
assertions
on
more
than
just
the
deployment
of
the
control
plane.
So
we
can
make
assertions
on
whether
the
add-ons
are
getting
installed
or
not
so
like
in
stu.
I
We
have
two
workflows.
One
is
one
that
just
makes
the
assertion
that
can
we
deploy
a
service
mesh?
Can
we
deploy
any
service
mesh
after
this
pr
is
merged,
or
will
this
break
something?
So
these
end-to-end
tests
are
really
helpful
in
catching
those
now
now
seguing
and
all
of
these
things
are,
if
you
can
share
this,
the
link
to
this
doc.
All
of
these
things
are
well
documented
in
this
particular
document,
and
now
that
would
be
a
segue
into
extending
the
fun
use
of
these
particular
end-to-end
workflows.
I
So
currently,
what
we
do
is
that
these
workflows
are
kind
of
an
intermediary
workflow,
so
they
require
some
input
and
they
spit
out
some
output.
So
these
inputs
would
be
a
patent
file,
so
I
would
not
go
into
describing
what
path
that
would
take
time,
but
a
patent
file
can
can
basically
helps
you
define
what
your
workloads
are
and
that
workload
can
be.
I
Okay,
deploy
a
service
mesh,
so
we
can
set
a
patent
file,
give
it
to
this
particular
workflow
and
it
would
run
the
it
would
try
to
run
the
deploy
the
patent
file
check
if
all
the
pods
are
up
and
currently
we
stopped
there
so,
but
what
we
would
really
need
is
to
reflect
these
changes
in
a
dock,
so
when
the
pr
is
merged,
so
we
are
on
the
edge
or
when
we
make
a
new
release
in
both
of
these
cases,
we
would
want
to
know
that
whether
the
tests
are
passing
or
failing.
I
So
what
I
have
added
is
that
I
have
added
a
basically
whatever
the
output
comes
out
of
this
particular
workflow.
We
can
push
that
into
a
file
and
then
we
can
use
that
file
to
that
file
is
basically
a
json
file.
I
Then
we
can
convert
js
file
into
particular.
Read
mark
then
show
it
to
the
user
current.
Apparently,
the
structure
of
this
json
file
looks
something
like
this
so
for
this
particular
kubernetes
version
and
for
this
particular
service
mesh
and
for
this
version
of
that
service
mesh,
the
adapter
version
is
edge
when
the
pr
is
merged
and
when
we
make
a
release,
it
would
be
the
name
of
that
particular
release.
I
So
this
is
what
this
particular
thing
does
so
yeah,
that's
pretty
much
it,
and
if
anyone
wants
to
contribute
to
by
start
by
writing
actions,
we
have.
The
thing
is
that
we
have
more
adapters
than
we
have
people
contributing
to
those
adapters,
because
we
we
are
trying
to
support
a
lot
of
service
meshes.
So
we
have
a
bunch
of
adapters
and
we
don't
have
enough
people
to
contribute
to
them.
So
if
you
are
looking
for,
if
you
are
looking
to
start
contributing,
this
can
be
a
good
place
to
go.
I
Also,
they
also
need
some
love
and
sometimes
what
happens.
What
goes
on
to
happen
is
that
we
make
some
change
in
the
in
the
more
active
adapters
and
the
passive
adapters
kind
of
we
forget
about
them,
or
we
may
change
to
them
a
bit
later
like
one
week
later,
so
you
can
go
ahead
and
basically
these
when
I
would
make
even
more
changes
on
these
workflows,
for
example,
for
the
next
topic
that
I'm
going
to
talk
about.
I
Is
I'm
going
to
change
a
little
bit
of
the
core
workflow,
so
these
workflows
would
need
change
and
that
would
kind
of
be
a
copy
based
thing.
So,
if
you're
just
looking
for,
I
will
make
that
change
into
the
html
workflow
and
after
that,
maybe
someone
if
you're
interested
in
writing
it
up
workflows.
You
can
just
copy
paste
that
into
all
of
the
other
adapters
so
and
we
have
a
really
a
lot
of
them.
Even
like
five
of
you.
I
If
five
of
you
sign
up
to
to
basically
do
these
things,
we
have
six
adapters
so
more
adapters
than
people
and
okay.
I
can.
I
can
also
talk
about
the
next
thing,
so
we
support
okay.
So
there
is
this
dynamic.
There
is
a
lot
of
dynamicism
in
adapters,
so
every
time
a
release
is
made,
we
check
for
the
for
the
latest
particular
version
of
that
service
mesh.
What
workloads
of
that
service
measures
up
so
currently
we
don't.
Actually
we
just
push
it
to
the
adapter.
I
Okay,
no
one
else
knows
that
what
particular
components
for
a
service
mesh
we
support
so
a
little
bit
of
change
that
that
I
made,
which
would
be
in
the
already
existing
workflow
that
we
had.
All
I
would
need
is
that
inside
of
the
docs
in
what
directory
we
need
that
to
be
pushed
so
every
time
a
release
would
be
made.
Something
like
this
would
be
displayed
onto
that
particular
version
number,
so
supported
transfer.
I
This
version
is
okay,
so
these-
and
these
are
all
workloads
that
we
support
for
this
particular
version
of
this
service
mesh.
So
we
can
display
all
of
this,
and
that
would
happen
every
time
a
release
is
made
so
and
this
this
would
be
pretty
quick,
and
I
I
would
make
this
change
again
in
this,
and
we
would.
It
would
just
be
a
copy
paste
in
all
of
the
other
adapters.
I
So
if
you
are
looking
to
start
contributing
in
workflows
so
that
that
can
be
a
thing
after
I
after
I
basically
move
to
your
app
so
yeah.
These
are
the
two
things
that
I
want
to
talk
about,
and
these
end-to-end
workflows
we're
going
to
use
them
a
lot
in
a
lot
of
other
places
as
well
and
yeah.
That's
that's
about
it.
C
Yep,
hey
so
a
brief
repeat
of
the
last
set
of
discussions
that
we
just
had.
We
were
talking
about
the
same
thing
we
were
talking
about.
Quality
assurance
in
the
measuring
project
specifically
talked
about
it
from
two
perspectives.
C
One
is
driving
integration
tests
as
mario
informed
us
these
e-2e
these
end-to-end
integration
tests
driven
from
one
of
mesherie's
clients,
mesri
ui.
You
can
imagine
a
mesh
user
they're
using
the
mouse
to
interact
with
meshri.
C
Well,
there
there
are
bugs
around
where
mouse,
mice
click
right.
So
it's
like.
We
need
to
do
testing
from
measure
uis
from
that
vector
from
that
starting
point.
C
We
also
need
to
do
as
a
project
collectively
we
need
to
do
testing
from
other
perspectives.
From
other
vantage
points,
measury
ctl,
it's
the
met.
It's
the
cli
for
measuring
running
tests
from
that
starting
point
is
also
an
excellent
thing
to
be
doing
and
actually
in
part,
what
ashish
was
just
covering.
C
I
believe,
under
the
covers,
is
using
mastery
ctl
in
a
few
places,
and
so
anyway,
between
these
two
topics,
we're
kind
of
covering
how
to
do
functional
testing,
which
is
distinct
from
unit
testing
from
both
front
end
and
back
end.
The
way
that
ashish
is
pushing
the
project
forward
here
on
this
back
end,
testing
is
starting
out
with
a
focus
on
one
of
mescheri's
components:
meshri
adapters,
there's
one
meshri
adapter
for
each
and
every
service
mesh
that
meshry
supports.
C
Someone
makes
a
pull
request,
and
even
after
that
happens
that
the
code
is
the
adapter
code
is
compiled
and
but
really
to
test
whether
or
not
the
adapter
is
doing
what
it's
supposed
to
do,
which
someone
like
ashish
who's
sharing,
might
bring
up
the
mastery
docs
and
tell
you
that
what
a
mesh
readapter
is
supposed
to
do
is
speak
to
and
manage
its
respective
service
mesh.
So
in
this
case,
here's
istio
that
adapter
is
supposed
is
supposed
to
talk.
Istio
speak
because
istio
does
certain
things,
while
other
service
meshes
do
other
certain
things.
C
Well,
really,
in
order
to
flex
the
adapter's
code,
you
you
need,
for
you
need
to
run
integration
tests,
which
is
to
say
that
you
need
a
kubernetes
environment.
You
need
a
meshery
server,
you
need
the
meshre
istio
adapter,
and
then
you
need
to
run
through
some
tests
that,
in
that
provision,
the
service
do
all
the
things
that
she
had
said
so
good.
C
I
just
did
a
brief
recap
just
to
make
sure
that,
like
everyone,
who's
on
the
call,
whether
you're,
new
or
not
like
everyone,
that's
on
the
call
recognizes
the
opportunity
to
engage
here,
especially
if
golang
isn't
your
thing
or
react.js.
Isn't
your
thing
by
the
way,
there's
a
lot
of
other
technologies
that
that
get
used
in
the
community
other
than
those
you
can
find
them
in
the
community
handbook.
C
C
There
should
be
green
all
over
that
thing
at
some
point.
Why?
Because
we're
doing
lots
of
tests
and
the
tests
are
passing
at
some
point.
You
know
we
start
out
manually
tracking
this.
So
mario
was
pointing
people
to
the
spreadsheet.
Saying,
hey,
you
know,
jump
in,
engage,
we
don't
have
all
the
test
cases
listed.
Meshery
does
a
lot
of
things.
They're,
not
all
articulated
in
here.
So
certainly
new
lines
need
to
be
added.
C
If
you
see
someone's
name
like
poke
them
or
bump
them
aside
or
there's
a
lot
of
things
going
on,
so
don't
be
shy
to
ask
people
if
you
can
help
or
if
that's
an
area
of
interest
that
specific
line,
but
what
ashish
was
trying
to
do
is
level
us
up
a
little
bit.
C
Good
the
thing
is,
is
like
he's
he's
helping
the
project.
Take
the
next
step
and
he's
saying
at
any
given
moment.
If
I
were
to
come
over
as
a
user
and
say
hey,
you
know:
what's
the
health
of
the
current
release
like
how
are
its,
how
are
those
tests?
Looking
you
know,
I'm
using
the
older
release,
I'd
like
to
go
to
this
new
one.
C
Should
I
be
confident
in
taking
a
step
forward?
That's
a
question:
they
might
have
another
question
they
might
have
that
ashish
might
show
us
on
the
in
the
mesherie
repo
under
the
issues
section,
there's
a
highlighted
issue
called
compatibility
matrix.
So
another
question
that
people
might
have
outside
of
just
wanting
to
get
a
confidence.
A
quick
confidence
level
read
on
the
quality
of
a
given
measuring
release.
C
Did
you
already
say
a
few
words
on
this,
or
do
you
want
to
say
a
few
words
like
so
there's
a
difference
between
tests,
passing
and
compatibility
with
external
systems,
but
the
two
are
highly
related.
C
So
that
so
that
pause
was
for
ashish
to
say
some
things,
but
ultimately
what
she
is
enabling
the
project
to
have
is
basically
kind
of
what
you
see
in
the
screenshot.
Here
is
just
an
example
of
a
compatibility
matrix
so
part
of
the
integration
test
when
they're
done
and
we
see
those
green
lights
and
we
click
merge.
Well,
those
test
results,
kind
of
get
lost
to
the
annals
of
github.
C
Instead,
what
he's
leaked
champion
championing
an
effort
to
do
is
to
centralize
the
output
of
those
test
results,
so
they
can
be
exposed
in
a
page
on
in
measuring
docs,
so
that
when
someone
wants
to
know
how
healthy
is
a
project,
how
from
how's
the
project
doing
both
from
a
quality
perspective
and
because
these
are
integration
tests,
where
that
a
particular
measure
adapter
for
a
certain
service
mesh
like
linker
d,
if
that
test
pass
against
a
certain
linker
d
version.
C
Good,
so
that's
what
those
two
things
are
about
like.
I
really
encourage
you
to
reach
out
to
in
the
community,
to
the
folks
that
you
just
heard
from
abi
abhishek
kumar
is
also
digging
his
heels
in
go
ahead.
Do
you
have
questions.
A
Comments
yeah,
while
we
are
on
the
subject
of
testing
like
used
like.
Would
you
like
to
quickly
point
out
to
some
of
the
work
you
have
been
doing
in
and
around
measuring
ctl
and
advancing
the
unit
test
coverage
in
measuring
city
and
also
how
people.
B
Ctl
has
its
own
kind
of
an
area
where
we
test
things.
So
let
me
just
demonstrate
where
we
are
running
these
tests.
Okay,
so.
A
A
Ctl
first
is
unit
testing,
which
basically
checks
like
we
kind
of
mock
the
back
end
response
from
the
message
server
and
we
try
to
check
if
the
like
the
behavior
of
the
command
is
exactly
that
it
is
supposed
to
be,
and
that
is
the
one
kind
of
testing
we
do.
A
The
second
is
the
intuition
testing,
where
we
actually
try
to
deploy
measuri
in
docker
and
kubernetes,
and
we
try
to
like
run
different
system
commands
like
start
restart
logs,
almost
all
of
them
to
like
make
sure
that
whenever
someone
is
deploying
a
machine
in
their
system
and
the
commands
are
running
okay.
A
So
how
do
someone
should
engage
into
this
this
area?
There
is
a
whole.
What
should
I
say,
a
guide
written
about
message,
ctl
that
has
the.
A
Guidelines
about
how
you
can
actually
start
writing
these
tests,
how
you
can
improve
them.
There
is
a
high
demand
of
people
to
improve
these
tests.
To
be
honest,
let's
say
I
have
opened
an
issue
just
now
that
actually
wants
to
improve
the
messages.
Integration
testing
and
I
have
written
that
like
right
now
in
intuition
testing.
We
are
deploying
the
machine
messaging
in
a
cluster
on
docker
platform
on
both
of
these,
but
we
are
not
actually
checking
after
each
command
if
measuring
is
running
or
not
like.
A
After
start,
we
should
like
confirm
if
measuring
is
running
now
or
not,
because
sometimes
it
fails
and
we
don't
know
it
might
fail
in
real
case,
but
during
the
testing
we
fail
to
detect
it.
A
So
there
should
be
another
check
to
like
verify
if
measure
is
running
after
executing
system
start,
so
these
kind
of
things
needs
to
be
improved
in
the
testing
areas.
So
how
should
someone
go
here.
E
A
It's
not
like
that,
you
have
to
just
you
will
find
issues
here.
What
I'll
ask
you
to
actually
go
deep
down
into
the
test,
what
we
are
doing
there
and
how
you
think
we
can
improve
them,
because
that
is
how
we
will
improve
the
test
efficiently,
because
as
any,
I
might
have
missed
something.
Another
person
might
miss
something
while
writing
these
tests,
but
if
some
other
guy
or
other
person
will
check
the
code-
and
he
might
like
get
an
idea
how
he
can
improve
that
particular
test
very
efficiently.
A
That
is
one
area,
and
if
you
there
are
many
areas
to
engage
with
mercedes
like
there's
a
whole,
a
very
big
issue
that
we
want,
we
want
to
generate
reports
with
system
check
there
is,
there
is
a
command
in
the
system.
We
want
to
generate
diagnostic
reports,
and
this
is
for
the
case
whenever
a
user
is
user
is
facing
something,
but
he
fails
to
get
the
supposed
behavior.
A
So
we
want
to
create
a
command
that
actually
grabs
all
kind
of
stuff
like
logs
many
kind
of
reports,
so
that
we
can
debug.
How
exactly
was
the
system
behaving
during
that
user
was
trying
to
do
something
with
measuring.
A
There's
a
whole
if
you
are
interested
in
like
writing.
Docs.
There
is
a
really
good
issue,
then
that
we
want
to
do
is
we
want
to
record
the
terminal
sessions?
We
want
to
record
record
how
exactly
a
system
command
works
and
we
want
to
add
all
these
these
videos
or
kind
of
gifs
to
the
message
ctl
docs.
That
is
also
one
area.
If
you
want
to
engage,
you
can
engage,
engage
here
and
yeah.
Some
of
you
might
will
try
just
clicking
on
component
mastery
ctrl
and
see
what
kind
of
issues
are
there.
A
That
is
one
way
to
like
get
started
and
know
where
you
can
improve
messy
ctl,
but
there
are
many
things
that
you
can
actually
do.
If
you
go
to
the
code
like
there
are
commands
like
perf,
there
are
commands
like
mesh
and
some
commands
like
config
and
many
more,
which
are
used
very,
like
very
less
in
comparison
to
other
commands,
and
they
actually
requires
a
lot
of.
They
have
a
lot
of
areas
to
improve
on.
A
So
I
will
ask
you
guys
to
use
those
commands
first,
try
to
see
what
can
be
improved,
how
what
can?
What?
What
can
be
the
additional
features
you
can
come
like
bring
on
the
plate
so
use
them
first
and
then
try
to
go
to
the
code
and
think
of
the
ideas
you
are
going
to
bring
on
the
plate
and
improve
those
messaging
commands.
A
Thank
you
pews,
like
yeah.
I
like
all
of
these
tests,
although
talking
about
all
of
these
tests
and
everything
like
it,
might
be
hard
to
newcomer
hard
for
newcomers
to
digest
all
this
in
one
sitting,
so
yeah
like,
but
still
this
is
the
as
we
mentioned
you
being
a
user,
is
the
best
way
to
start
contributing
and
best
way
to
make
a
lot
of
impact
right
off
the
bat.
A
Any
comments
on
this,
or
so
we
are
almost
at
time,
and
I
think
we
have
one
more
topic
left
so
before
that
like.
If
anyone
has
any
comments
so.
H
I
did
wanted
to
like
raise
a
point
on
that
comment
to
the
right
about
like
be
used
answer,
but
maybe
I
could
just
answer
that,
like
on
the
slack.
A
Okay,
I
okay
good.
A
And
on
the
final
topic,
for
today
the
workflow
to
alert
newcomer
sounds
like
do.
We
need
to
discuss
that
on
the
call
now.
C
Yeah
yeah,
we
should
just
give
a
quick,
let
so
pronounce
you
want
to
give
people
a
quick
overview.
Yeah.
J
C
For
two
dollars,
it
is
by
the
way,
just
there's
a
standing
rule
in
the
community.
If
you
ask
if
your
screen
is
visible
when
you're
in
a
time
crunch,
and
then
you
spend
10
seconds
or
like
when
I
comment
you
spend
another
minute
on
whether
or
not
your
screen
is
visible.
It'll
just
cost
you
two
dollars,
so
that's
all
yeah,
okay,.
J
So
I
I
would
say
that
initially,
like
every
maintainers
had
to
like
explicitly
write
about
a
new
issue
which
was
coming
up
on
different
repositories,
so
this
was
like
very
difficult
for
everyone
to
keep
track
of
so
any
so.
Instead
of
that,
like
there
was
an
initiative
of
writing
workflow
which
can
automate
this
whole
task.
J
So
what
I've
done
is
that
I
have
added
a
workflow
which
this
is
a
workflow
which,
on
every
whenever
a
new
issue
is
open
or
an
issue
is
stable
with
good
first
issue
label
or
first
timers.
Only
then
it
automatically
sends
that
link
to
that
particular
issue
onto
the
our
newcomers
channel
so
like.
We
also
did
a
test
today
and
it
was
working
fine,
and
the
only
thing
is
that
we
need
to
replicate
this
particular
workflow
onto
all
of
the
repositories.
J
So
no
and
the
change
there
shouldn't
be
any
change.
It
should
directly
be
added
to
other
repositories,
so
newcomers
can,
if
anyone
wants
to
contribute
to
it
by
adding
this
workflow,
then
feel
free
to
do
it
and
we
really
need
to.
We
didn't
need
help
on
that,
because
they
are
around
60
repositories.
So
it's
little
difficult
to
directly
add
on
so
if
anybody
wants
to
get
hands-on
on
writing
workflows,
then
this
is
a
really
good
start
and
that's
all
for
this
any
comments.
J
A
The
this
is
like
a
really
great
pronoun,
like
i41.
I
think
people
would
definitely
benefit
from
having
good
first
issues
shorter,
shorter
than
them,
so
newcomers
can
jump
in
and
take
a
look
at
it.
So
yep,
as
per
anna
mentioned,
feel
free
to
get
involved
here.
If
you
just
want
to
start
contributing
and
all
right,
I
guess
that's
all
the
time
we
have
today
and
all
the
agenda
we
have
today.