►
From YouTube: Layer5 Community Meeting (June 4th, 2021)
Description
Layer5 Community Meeting - June 4th, 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
A
B
B
Who
is
you
know
old
news
at
this
point,
but
the
other
ayush
it'd
be
really
nice
to
hear
from.
C
Hello
hi,
so
I'm
irish,
I
am
a
junior
at
sri
lanka,
washington
university,
I'm
currently
studying
computer
science
and
I
am
really
interested
in
design
actually.
So
I
might
switch
and
I've
been
working
on
my
sweet
measuring
ui
till
now.
B
Yeah,
that's
a
great
that's
great
yeah!
I
hear
what
you're
saying
about
you
know
ui
and
in
particular,
like
ux,
being
being
a
really
a
try,
interesting,
intriguing
thing.
So
much
about
psychology
yeah,
so
many
tools
of
the
trade
too
and
such
an
such
like
a
well
something
of
a
relatively
an
emerging
trade,
human-centered
design,
user-centered
design
things
that
some
some
things
that
are
hard
to
quantify.
B
B
By
the
way,
does
ps
stand
for
postscript
is
that
the
last
name.
C
Yeah
p
stands
for
pradop:
sync,
that's
just
a
cell
phone,
so.
B
Mesh
mate
patel:
has
you
covered
so
yeah
yeah
ishida
welcome
ishida.
Have
you
gotten
a
chance
to
introduce
before.
B
So
if
she
did,
if,
if
you
can,
you
know,
say
hi
and
chat
or
or
it'd
be
nice
to
get
to
know
you
more
and
then
you
know
navindy,
I
suspect
we
have
some
other
new
folks
on
as
well
as
nithish
karthik.
D
Hello,
everyone,
I'm
nitish
kartik
and
I'm
actually
currently
studying
in
my
first
year
at
iit,
bhu
and
like
I'm
very
into
I'm
very
much
into
front-end
engineering,
and
I
am
very
proficient
with
like
I've,
been
exploring
many
technologies
till
now
and
I
have
been
thinking
for
a
long
time
about
contributing
to
open
source
and
I
haven't
gotten
a
chance,
but
I
think
I've
landed
at
the
right
community
right
now.
So
I'm
very
excited
to
be
a
part
of
this
community
and
let's
see
how
it
goes.
A
Nice
to
have
you
here
any
other
newcomers
would
like
to
introduce
very
much.
A
So
the
first
item
on
the
agenda
is
from
michael
michael:
do
you
want
to
share
your
screen
to
talk
about
error
codes.
B
B
Perfect
good,
that's
what
you
took
the
words
out
of
my
mouth
very
nice
to
meet
you!
Thank
you
for
noticing
the
name
good
to
have
you
and
thanks
for
putting
up
with
all
of
the
sarcasm
as
well.
So
it's
nice,
it's
nice!
Yeah!
E
No
problem,
my
favorite
color
is
also
blue.
So
that's
good.
It's
a
good
color!
Yes,
thank
you!
So
just
a
quick
demo
of
this
new
tool.
It
has
been
discussed
a
couple
of
times.
Oh
now,
I
have
a
gallery
view.
That's
really
nice
and
lots
of
people
having
their
happy
shack
for
the
first
time,
hello,
I'm
good!
E
Well,
anyway,
so
this
this
tool
is
actually
a
static
analysis
tool
of
the
code
to
to
check
that
we're
following
the
convention
about
errors
that
we
have
established
and
to
extract
error
information
or
in
the
details
of
errors.
E
From
from
the
code
from
each
repository
each
component
and
as
a
basis
for
publishing
that
error
reference
on
the
measuring
website.
I
think
it's
easiest
just
to
share
the
screen
and
show
what
the.
E
So
do
you
see
my
screen
with
intellij
yeah,
it's
readable,
okay,
so
this
tool
is
intended
to
be
used
on
the
command
line
locally,
to
check
the
errors
and
to
extract
information
and
to
do
some
sanity
check
and
get
some
analysis
summaries,
but
mainly,
I
guess
it
will
be
used
in
the
cicd
workflow
to
extract
this
error
information
and
then
to
publish
it
or
to
upload
it
to
commonplace
or
maybe
to
create
the
pr
in
the
measuring
repository
so
that
the
error
reference
can
be
updated
for
the
time
being.
E
It's
just
called
error,
u2,
nothing
special
it
has
it
uses
cobra
as
command
line
utility
command,
then
parameter
and
flag
utility.
There
are
a
couple
of
available
commands,
analyze,
talk
and
update
the
most
important
one,
maybe
stock,
where
there
is
a
more
or
less
detailed
description.
E
What
the
tool
is
doing
and
what's
the
contract
is
between
the
tool
and
the
code
error
names
I
I
won't
go
through
all
of
this.
You
can
read
that
a
pleasure
when
you're
using
it
or
when
you're,
just
interested
how
the
the
convention
is
built
up.
So
this
tool
is
relying
on
conventions
on
naming
conventions.
So,
for
instance,
all
error
codes
have
to
start
with
err
and
then
a
capital
letter
they're
recognized
by
this
and
that
they're,
constants
or
variables
in
the
code.
E
The
tool
uses
ast
to
travel
the
code
and
finds
the
entities
we're
interested
in
using
these
conventions,
and
it
makes
it
much
better
than
just
using
a
regular
expressions,
for
instance,
and
it's
more
reliable,
what
we're
not
doing
yet
using
typing
library.
So
it's
still,
for
instance,
when
we're
looking
going
through
the
code
and
finding
method
signatures
function
signatures.
E
What
this
one
does-
and
it
has
a
couple
of
mountain
parameters,
so
you
can,
for
instance,
specify
which
is
the
starting
directory.
It's
if
you
don't
specify
anything
it's
the
current
directory.
You
can
add
directories
that
need
to
be
skipped.
For
instance,
if
we're
running
on
in
the
measure
repository,
we
would
like
to
have
an
error
reference
in
a
separate
file
for
measuring,
but
also
one
for
measure
ectl.
E
So
when
you
run
it
on
measurey,
you
would
excel
measure
ctl
and
there's
a
default
list
of
directries
that
are
skipped
like
github
and
the
the
git
and
the
hub
directory,
and
you
can
also
have
a
verbose
flag.
Then
it
prints
out
and
debug
information
as
well
running
analyze,
we'll
just
go
through
the
code.
Do
not
change
it
at
all,
get
detailed
information,
and
then
you
get
outputs
output
files.
E
Here
you
have
a
an
analysis
that
just
lists
the
raw
data
that
it
finds,
so
it
has
entries
and
then
for
all
the
error
codes
it
finds
in
the
code.
It
gives
information
details
the
old
code
when
the
code
is
going
to
be
changed.
So
this
is
the
same
result
that
you
see
here
when
you
actually
update
codes
I'll
speak
about
that
in
a
minute,
then
there
is
an
information
whether
this
code,
that
is
encountered
in
the
code
is
actually
a
literal
and
not
a
call
expression.
E
For
instance,
it
can
be
not
an
integer,
it
can
be
a
string
and
then
the
convention
is
that
the
tool
when
you
update
it
replaces
this
string
with
an
integer.
So
that's
one
of
the
main
purposes
of
this
tool
to
assign
error
codes
to
all
these
errors
that
are
here.
I
am
going
to
demonstrate
that
in
a
bit,
so
everything
that
is
found
in
this
library
is-
and
here
the
raw
entries.
If
you
like
and
then
get
the
summary
in
this
summary,
you
get
information.
What
is
the
minimal
code
used?
E
Minimum
code
used
the
max
code
used,
and
you
see
it's
a
big
range
here,
but
it
doesn't
mean
that
all
the
numbers
are
taken
because
here
they're
big
holes,
so
the
first
implementation
was
of
handling
error
codes
or
the
design
was
that
we
have
certain
error
code
ranges
assigned
to
like
adapters,
different
adapters
and
so
on,
and
that's
why
the
range
is
so
big,
but
we
have
not.
E
We
have
decided
not
to
to
have
error
code
ranges,
but
anyway,
so
this
code,
this
tool,
this
cool
tool
analyzes
and
makes
sure
that
we
do
not
have
duplicate
codes.
It
doesn't
do
anything
with
duplicate
codes,
but
just
reports
it
so
here
this
code
is
duplicate
code,
so
this
has
to
be
fixed
manually,
for
instance,
by
changing
the
number
or
by
adding
a
text
instead
of
the
integer
saying
you
know,
replace
me
or
anything,
there's
no
convention
there.
E
You
can
also
see
that
one
of
these
text
that
should
then
be
replaced
by
an
integer
is
kit
1025,
and
they
will
also
show
up
as
duplicates
then
mesh
kit
test
code
and
so
on.
So
this
is
all
our
old
duplicates
that
will
be
replaced
when
the
tool
is
used
to
update
the
codes
and
replace
the
text
with
integers.
E
You
can
also
see
that
there
is
an
entry
for
duplicate
names,
so
the
convention
is
that
the
error
names
need
to
be
unique
across
this
component,
a
specific
component,
so
it's
not
allowed
to
have.
For
instance,
this
one
twice
declared
twice
in
the
code.
So,
if
that's
the
case,
it
would
show
up
here
and
so
the
two
uniqueness
constraints,
if
you
like,
the
name,
has
to
be
the
same
across
one
component,
which
is
usually
a
repository
assumed
to
be
unique
and
then
also
the
the
integer.
E
The
error
code
itself
is
also
unique
across
a
for
a
component.
Then
what
I
said
a
minute
ago
is
that
the
error
codes,
so
these
codes
here
right.
They
need
to
be
literals
so
and
they
need
to
be
integers
in
the
end
and
but
they
need
religious
strings.
E
So
they
can't
be
call
expressions
like
this
one.
Let's
see
whether
you
can
find
it
this
one
right.
This
is
a
call
expression
and
that
is
used
in
the
code
and
that's
not
allowed
because
then
the
utility
won't
work.
So
this
has
to
be
replaced.
This
is
a
manual
process,
so
someone
has
to
go
in
and
actually
do
this
and
put
in
a
literal
string.
That
is
then
replaced.
E
It
is
marked
as
deprecated
in
the
code,
so
the
static
interval
issue
an
error
if
it
finds
it,
but
also
this
tool
will
actually
will
actually
list
it
here,
but
not
do
anything
with
it.
So
you
can
have
just
a
quick
look
to
see
what
I'm
talking
about
so
broker.
C
E
Go
back
to
the
summary,
and
then
it
also
creates
a
another
utility
export
json
file,
which
is
the
file
that
is
then
used
in
the
error
code.
Reference
on
the
measuring
website
to
create
this
website.
So
you
can
see
here
right
now
in
meshkit
component
they're,
just
three
areas
that
actually
correspond
to
the
convention
and
that
is
then
extracted.
You
see
that
this
component
is
called
meshkit.
E
It
has
a
type
called
library
and
the
three
areas
that
show
up
that
are
ready
and
that
actually,
where
actually
the
the
new
the
error
new
with
all
the
details
about
the
errors,
are
used
right.
So
these
are
show
up
here
and
you
can
see
here.
I
replaced
two
of
them
and
one
is
error
in
them.
E
The
other
one
is
a
reviver,
and
then
you
can
see
that
the
information
which
is
in
the
code
where
this
is
declared
is
extracted
and
can
then
be
imported
into
the
measuring
airport
reference
on
the
website,
and
so
this
is
one
thing
and
then
the
other
thing
that
it
does
is
actually
update
the
codes,
the
error
code.
So
all
these
strings
replace
me
test
code.
Everything
are
replaced.
E
E
E
Appear
in
exit
presentation
mode,
so
you
see
what
has
been
changed
now
f5.
Can
you
still
read
it?
If
not,
I
know
you
see
here.
If
I
look
at
the
difference,
what
the
tool
now
did
it
replaced
the
string
literals,
which
were
a
string
and
not
an
integer
called
test
here
with
the
next
codes
in
that
list,
so
it
just
automates
incrementally
and
then
it
also
implements
component
information.
E
E
E
You
see
it's
replaced
to
1039.
C
E
A
Thank
you,
michael.
That
was
a
really
good
demo.
One
question
I
had
like:
while
using
this,
do
we
define
errors
for
one
particular
package
like
within
a
component
or
what
is
the
convention
that
you
so.
E
The
convention,
if
you
understand
your
question
correctly,
is
the
errors
are
for
one
component,
so
the
name
spaced
to
the
component,
so
which
means
which
means,
if
you
want
to
you,
can
use
the
same
error
code
like
1050
in
different
components.
You
can
also
use
the
same
error
name.
You
know
if
it
makes
sense
in
different
components,
but
the
convention
is
also
that
we're
not
reusing
errors
or
error
codes
across
components,
meaning
what
I'm
trying
to
say.
With
this
we
don't
import.
E
A
Other
comments,
other
suggestions,
so
to
give
a
bit
of
context
like
if
it
wasn't
clear
before
what
we
can
use
is
we
can.
A
We
can
use
this
cli
tool
that
might
be
presented
right
now
and
use
it
for
other
measuring
other
layer,
5
projects,
other
measuring
components
like
measuring
ctr
and
measuring
server,
so
that
we
can
have
a
consistent
error
code,
consistent
error
record
across
all
of
these
projects,
and
we
also
have
a
error
called
documentation
that
would
be
automatically
generated
with
this
with
this
file
so
so
yeah
and-
and
it
should,
it
can
automatically
be
published
on
the
layer
file
on
the
measuring
website.
E
Links
in
the
comments
here
you
can
see.
This
is
how
it
looks
like
who
has
been
working
on
on
that.
I
can't
remember
his
names
yeah.
B
E
E
So
this
is
the
result
of
I
mean
that
there
were
no
like
error
codes
and
no
no
details
defined.
So
that's
why
it's
empty
everywhere
here,
so
they
will
show
up
here
with
all
the
information
about
the
error
codes
it
can
be
used
to
for
the
users
to
troubleshoot
and
find
out
what
is
going
on
what
went
wrong
if
anything.
B
Cool
good,
well,
hey,
there's,
a
reason
why
we're
so
so
this
is
the
the
layer.
Five
community
call.
We
end
up
talking
about
all
the
projects
that
we
talk
about
measuring
more
than
the
other
projects.
It's
the
biggest
project,
there's
a
reason
why
we've
dedicated
you
know
as
much
time
to
michael's
effort
here
on
the
community
call
as
we
as
we
all
there's
a
few
reasons,
and
we
would
have
dedicated
the
time
irrespective
of
this
next
reason.
But
this
is
a
really
good
reason,
and
that
is
that
what
michael
is
showing
here?
B
It's
well,
it's
it's
not
embedded
in
mesh
receipt.
He
showed
a
command
line
utility
and
it's
not
inside
of
mastery
ctl,
it's
like
well,
it
could
be,
but
it's
broader
than
that,
it's
higher
level
than
that
and
it's
as
a
matter
of
fact,
it's
you
know
it's
so
so!
B
Hence
it
it
sits
at
this.
Mesh
kit
level,
like
michael
in
particular,
has
been
a
steward
of
not
repeating
ourselves
and
of
working,
smarter,
not
harder
and
some
other
cliched
statements.
But
of,
and
so
I
wouldn't
propose
that
this
go
into
measuring
ctl,
not
that
it
wouldn't
conceptually
fit
there
necessarily,
but
it's
broader
than
that
anyway.
I
kind
of
just
said
that,
but
the
reason
that
we're
covering
it
in
part
on
the
community
call
is
because
this
is
intended
to
be
used
across
all
components
of
measuring.
B
B
I
mean
there'll,
be
examples
and
things
if
you
work
on
the
operator
or
on
mesh
sync
on
the
broker,
so
abhishek
abu
shaykh
apashek,
then
you
have
to
I
guess,
and
hopefully
avashek
is
wooing
others
to
work
with
him
in
those
components,
but
those
are
example,
components
that
will
also
directly
benefit
from
this
framework.
This
system
mesh
reserver.
B
If
you've
worked
with
in
there,
eventually
one
of
you
and
maybe
you're,
not
on
the
call
at
the
moment,
call
it
out
if
you
are,
but
one
of
you
was
just
taking
on
a
measuring
ui
issue
in
which
you
were
beginning
to
expose
some
documentation
in
a
like
a
help,
a
help
icon.
B
So
if
you're
going
to
run
a
performance
test,
there's
three
different
types
of
load
generators
to
pick
from
what's
the
difference,
and
so
the
contributor
was
putting
in
a
help
icon
to
explain
what
the
difference
is
to
the
user.
It's
like
well,
that's
sort
of
like
inline
help,
that's
like
inline
documentation
and
well
it
does.
It's
no
stretch
of
the
imagination
that
the
json
that
michael
was
just
showing
one.
That's
like
he
was
showing
where
that
directly
maps
right
into
the
documentation.
B
So
so
for
those
of
you
who
don't
like
documentation,
you
can
thank
michael
again
and
again
because
this
is
an
auto
documenting
thing.
Literally,
the
there's
an
open
call,
there's
a
couple
of
calls
for
help
here,
a
few
things
for
all
of
you
to
digest.
B
So
I'm
looking
across
all
the
names
here,
it's
like
almost
almost
all
of
you.
This
will
touch
because
you're
either
contributing
in
golang
to
any
of
the
measuring
components
or
mesh
for
ctl,
and
the
nice
thing
is
that
you
know
two
months
from
now
when
this
is
just
commonplay.
You
know
a
month
from
now,
hopefully
when
this
is
just
commonplace
and
just
common
use,
a
lot
of
the
contributors
may
not
even
you
know
really.
B
You'll
you'll
just
copy
and
paste
what
other
people
you'll
just
go:
mesh
kit,
dot,
new
error
and
you'll
fill
in
the
stuff,
and
you
won't
really
think
about
the
fact
that
all
of
this
other
all
the
rest
of
the
publishing
and
the
conventions
and
the
assignment
of
error
codes
and
all
this
is
kind
of
being
taken
care
of
for
us.
But
the
users
will
well
eventually.
You
will
be
very
thankful
again,
because
users
will
have
error
codes
specific
to
your
error.
B
Whatever
you
made
up
right,
there
like
and
it'll,
be
unique,
it'll
be
referenceable,
but
we
need
a
couple
of
volume
and
a
couple
volunteers
to
see
the
effort
through
so
one
is
michael,
does
in
general
a
fantastic
job
of
writing
stuff
down
and
he
wrote
down
a
lot
of
things
about
the
behavior
of
this
utility
and
he
just
explained
not
all
of
them,
but
but
most
of
them,
and
he
spared
us
some
of
the
rest
of
them
because
it's
like
because
it's
a
lot
for
anyone
to
digest
and
he's
written
them
all
down
here.
B
We
really
need
to
take
this
and
upgrade
our
contributor
docs,
and
the
contributor
docs
have
recently
been
upgraded
on
docs.measure.io
they're
being
broken
down
by
individual
component.
This
definitely
needs
to
go
there
in
some
examples.
So
between
some
of
the
other
maintainers
here
on
navindu
adidi
abhishek
of
the
areas
that
you're
focused
on,
I
called
out
three
names
of
individuals.
That
would
be
great
to
like
pick
this
up
and
begin
to
show
demonstrate
examples
of
its
use.
B
So
there's
a
need
for
someone
to
document
the
from
a
developer's
perspective,
how
this
is
used
and
what
it's
done
and
michael
is
here
to
review
and
help
ensure
that
that's
you
know
correct.
Another
open
need
is
that
that
utility
it
is
such
a
prime
or
we
can.
We
can
integrate
that
into
the
github
workflows.
There
are
any
number
of
github
workflows
that
run
ever
like
anytime,
you
open
a
pr
anytime.
You
commit
to
an
existing
pr
anytime.
You
merge
a
pr
anytime.
B
Here
across
our
projects-
and
one
of
those
should
be
the
invocation
of
this
utility,
I'm
impressed
that
it
puts
out
three
different
json
files,
one
of
which,
like
we
were
saying,
directly,
goes
into
docs,
the
other
of
which
we
should
highlight,
and
I
try
to
highlight
it
just
in
some
note
taking
here
that
I
think
it's
the
I
I
think
it's
the
air,
the
errors,
export
json.
B
No,
I'm
sorry!
It's
the
analyze
summary
json
that
we
might
we
either
will
need
a
little
bit
of
human
review
or
just
in
the
github
workflow
that
execute
invokes
this.
It
should
probably
look
to
see
if
the
json
objects,
duplicate
codes
and
duplicate
names,
if
there's
any
entries
in
there,
and
if
so,
those
are
not
auto
remediated
from
what
I
understood,
and
so
you
know
we'll
need
to
go
back,
and
basically
you
repeat,
you
either
repeated
yourself
as
a
contributor
or
you.
You
use
the
same
name
that
someone
else
did
within
that
particular
component.
B
The
cool
thing
about
like
I
know
this
takes
a
little
while
for
everyone
to
digest,
but
the
cool
thing
about
what
michael
had
done
in
a
few
ways
is
is
that
he
had
he'd
work
to
make
sure
that
all
the
error
codes
are
unique.
There's
some
auto
incrementing
going
on
and
there's
some
persistence
that
happens
for
that
component.
B
B
Tracked
and
kept
kept
track
of
each
individual
component.
He
showed
the
component
underscore
info.json,
so
the
last
the
last
of
the
three
bullets
that
for
any
every
component,
every
component
measuring
component.
That's
here,
it
uses
mesh
kit
to
some
certain
extent.
B
Today,
not
a
meshry
server
in
particular
uses
it
maybe
the
least,
and
that's
an
outstanding
work
item
for
all
of
us
to
have
mesh
reserver
use
mesh
kit
more
and
more
and
more
specifically,
around
the
instructions
that
are
going
on
here,
and
so
every
component
needs
to
fill
in
their
component
name
into
that
json
file,
michael.
They
need
to
fill
in
their
component
name
and
and
start
the.
E
And
we
have
not
yet
an
authority
authority
authority
sort
of
list
of
types
right,
so
so
adapter
is
an
obvious
type.
Client
is
another
one,
maybe
library,
so
we
have
to
sort
of
define
that
somewhere
as
well.
B
Oh
yeah,
I
can't
do
it,
do
it
once
yeah.
That
makes
sense.
I
took
all
well
I'll
admit
to
taking
liberty
in
advance
of
all
this
being
automated
and
making
up
one.
B
Actually,
I
think,
michael
just
said
it
so
like
there
was
some
mystery
ctl
errors
that
people
just
kept
running
into
kept
running
into,
and
I
think
I
was
going
to
shoot
myself
if
I
had
to
repeat
to
tell
people
how
to
troubleshoot
it
for
the
9000th
time
and
it's
not
because
it's
because
of
our
it's
not
not
because
of
the
person,
the
people,
it's
because
it's
just
it's,
because
it's
a
natural
pitfall
that
people
fall
into
and
it's,
and
that
means
that
the
more
that
people
run
into
a
given
error,
the
more
that
a
we
should
consider
how
to
help
them
avoid
that
in
the
first
place,
but
so
to
michael's
point
michael,
I
I
use
the
word
client
or
the
type
of
client.
B
So
maybe
that
makes
sense.
Maybe
it
doesn't
good
so
there's
this
is
there's
more
to
do
more
to.
B
B
B
What
what
we
need
really
next
is
I'm
going
to
stick
my
elbow
into
abhishek's
side
and
probably
the
vendor's
side,
and
if
other
people
want
elbows
in
their
side
too,
they
just
say
you
know
volunteer,
but
for
these
two
individuals,
I'm
going
to
look
for
some
examples
that
they
will
help
promote
and
propagate
the
use
of
error
utils
so
between
between
some
of
our
more
popular
components,
mesh
reservoir
and
measuring
cto,
and
then
we'll
start
to
get
these
into
the
adapters
as
well
extrude.
B
By
the
way,
this
is
for
those
that
work
on
ui
you'll
inevitably
have
this
sweet
moment
where
you're
like
you're,
looking
you're
looking
over
someone's
shoulder
you're
walking
by
their
desk
you're,
seeing
them
share
a
tweet
or
something
and
they're.
Looking
at
your
ui
they're,
like
looking
at
your
docs
that
you
wrote
or
they're
looking
at
the
ui
that
you
designed
and
it's
such
a
sweet,
such
a
such
a
you
know
satisfying
thing
to
see
and.
B
Now
I
lost
my
train
of
thought
from
what
my
point
was
going
to
be
something
about.
Oh
you
know
like
if
you
ever
cease-
and
I
saw
it
the
last
a
week
ago
was
someone
was
in
slack
and
they're
like
yeah.
B
And
these
are
kind
of
circling
around
some
of
pyusha's
contributions
by
the
way
now
yeah,
you
know,
hey
I'm
getting
this
thing
and
it
wasn't
just
that
they're
like
saying
I'm
getting
this
thing.
What
do
I
do?
They're,
like
hey,
I'm
getting
this
thing,
and
I
see
that
there.
This
is
what
I'm
supposed
to
do.
They
were
like
they
were
saying
they
had
already
googled
it
it
had
already.
Actually
I
didn't
think
that
they
had
to
google
it.
You
know
why,
because
in
mastery
ctl,
when
they
got
this
error,
it
said
to
them.
B
It
gave
them
this
url
pound,
sign,
measuring
ctl,
client
or
whatever,
or
whatever
the
pound
sign
like
it,
brought
them
right
to
this
thinking,
air,
which
is
awesome,
but
hence
what
I
was
trying
to
say
earlier
about
measuring
ui
when
you're
in
the
ui,
and
if
you,
if
you
let
your
brain,
think
about
this
for
a
little
while
there's
a
lot
of
power
in
like
inline
documentation,
specifically
around
errors
that
people
might
face,
let
me
go,
let
me
go,
do
something
that
may
or
may
not
work.
I
don't
know.
B
B
B
Obviously
this
was
written
by
engineers
who
thought
that
other
engineers
were
going
to
use
it
because,
but
not
regular
humans,
because
no
you
know
because
nobody
can
read
this,
but
if
it
had
an
error
code
like
yeah,
so
so
this
area
we're
going
to
be
talking
about
this
area
more
and
more,
it's
really
cool.
Anyway.
B
We
have
so
much
more
to
talk
about
in
this
meeting,
and
I
just
I'm
super
pumped
about
how
dramatic
of
an
impact
that
foundational
capability
will
be
to
a
lot
of
what
goes
on
so
it'll
be
a
different
user
experience
than
a
lot
of
people
have
seen
from
other
utilities
in
the
cloud
native
ecosystems
be
good.
A
Awesome
yeah
that
was
great
and
let's
move
on
to
the
next
agenda.
So
utkarsh,
do
you
want
to
talk
about
your
upcoming
sessions
to
help
contributors.
F
Yeah
yeah,
it's
basically
a
short
announcement.
I
guess
so
we'd
be.
F
Very
soon
posting
some
sessions
for
contributors,
and
these
would
be
ideally
there
would
be
only
two
sessions.
One
would
be
dedicated
to
basically
letting
the
contributors
know
that
how
they
can
add
a
pattern
support
to
other
adapters.
Also
so
right
now
we
have
support
in
istio
adapter,
but
not
in
other
adapters,
so
that
supports
c2
appropriated
to
other
doctors,
and
that's
something
that
one
other
decision
would
be
dedicated
to.
The
second
decision
would
be
dedicated
to.
F
It
would
be
basically
incorporating
dynamic,
ui
components,
dynamic,
react
components
into
the
measuring
ui,
and
this
has
actually
also
coming
from
the
patent
sync.
I
don't
have
the
bad
patterns
register
some
of
this
stuff.
This
is
something
that
we'll
be
probably
talking
about
in
the
sessions
that
how
exactly
we
can
do
the
things
that
are
important
files
in
the
ui,
also,
while
rendering
a
very
dynamic
and
very
flexible
components.
So
these
are
the
two
sessions.
F
We
would
basically
discuss
the
basic
architecture
and
then
how
the
contributors
can
directly
start
contributing
onto
it.
I
think
yeah,
that's
it
I'll,
be
actually
probably
I'll.
I'll
probably
share
the
more
more
details
about
the
sessions
on
mystery
channel.
A
Awesome,
thank
you
utkar.
So
if
you
are
interested
in
working
on
that,
like
look
out
for
rudkash
messages
in
the
slack
channel
moving
on,
we
have
a
new
discus
dot.
A
A
So
it's
still
pretty
empty,
like
there
aren't
much
many
topics
of
discussion
going
around
in
here.
So
what
I
would
like
to
do
is
like
it
would
be
helpful
for
others
and
for
people
in
the
community.
A
If,
if
we
start
discussions
in
in
in
here
other
than
the
slack
channel,
because
the
slack
channel,
we
only
have
up
to
like
one
month,
one
month,
storage
for
messages,
so
we
end
up
repeating
most
of
the
stuff,
and
that
goes
for
like
contributors
and
maintainers
like
there
are
a
lot
of
repeated
questions,
and
this
can
save
us
a
lot
of
time
and
effort.
So
if
you
have
a
question,
you
can
create
a
new
topic
on
the
on
the
appropriate,
appropriate
channel.
A
I
guess
it's
called
so
this
will
help,
so
we
can
we
can
in
the
future
like
reference
this,
these
two
others,
so
we
don't
have
to
repeat
ourselves.
So
if
you
haven't
already
like
you
can
create
an
account-
and
there
is
a
invitation
link
here
which
you
can
use
to
join,
join
this
community.
E
B
Yeah,
the
more
act
and
the
more
the
more
active
we
are
in
slack
the
it
starts
to
get
to
13
13
days
above
this
now,
it's.
A
Okay,
all
the
more
reasons
to
use
the
discourse
forum,
so
I
have
shared
the
link
in
the
chat.
You
can
join,
join
it
right
away.
Okay,
moving
on
to
the
next
item,
so
we
are
approaching.
We
we
1.6
release
and
there
is
a
public
roadmap
for
measuring
and
measuring
ctl.
It's
on
the
measuring
github
rapper.
A
So
what
we
have
here,
what
we
have
been
working
on
like
we
it
it
will
be
released
in
a
couple
of
weeks,
so
we
have
updated
the
performance
profiles
in
measuri
server
and
we
also
have
support
for
measuring
patterns
and
a
coming
back
to
measure
ctl.
We
we.
We
have
been
working
on
kubernetes
support
and
we
also
have
support
for
measuring
patterns
and
we
have
refactored
the
measuring
perf
command
and
we
have
we
have.
A
We
are
working
on
working
on
supporting
multiple
kubernetes
platforms,
so
yeah
the
that's
the
roadmap
and
we
are
in
need
of
a
release
manager
early.
Do
you
want
to
talk
about
the
responsibilities
of
release
management.
B
I
do
actually
I
want
to
point
people
to
a
doc.
I
can't
I
I
only
had
30
seconds
before
the
call,
and
I
couldn't
immediately
find
the
doc
so
I'll
find
the
link
to
the
doc
rather
than
talking
everyone's
head
off
about
what
it
would
mean
to
wear
the
release
manager
hat
it's
in
short,
herding
of
cats,
is
sort
of
the
byline,
the
byline
of
that.
B
If
you're
interested
come,
see
me
and
also
I'll,
just
I'll
paste,
the
link
and.
B
But
so
oh
okay
and
you're
gonna
are
you
going
through
the
rest
of
the
their
names?
Next
to
some
of
these
items,
we
can
reserve
some
of
this,
maybe
just
given
the
time
for
some
of
the
mastery
call
potentially
but
but
the
names
that
are
next
to
each
of
the
items.
B
Some
of
you
have
some
extenuating
circumstances
surrounding
your
commitments
here,
which
is,
which
is
entirely
understandable.
Some
of
you
have
stretched
to
pull
in
a
couple
of
items
that
were
otherwise
slated
for
v070.
Just
that's
something.
That's
awesome
and
some
of
the
rest
of
us
just
have
are
still
crunching
away
on
v060
items
and
so
for
the
names
that
are
associated
here.
B
It'd
be
really
nice.
If,
if
this
next
wednesday,
if
we
could
try
to
arctic
cap,
articulate
capture
how
far
along
we
think
you
know
how
far
along
each
of
you
think
that
those
respective
areas
are
not
that
you're
entirely,
you
know
not
that
you're,
the
only
one
working
on
those
areas
by
by
far
rather
just
being
the
representative
of
the
area,
the
feature
lead
so
to
speak.
A
Yeah,
I
guess
we
are
almost
there
like
most
of
the
stuff
has
been
tried
and
tested
and
yeah
awesome.
So
we
will
be
talking
about
that
on
the
next
week's
missionary
development
call
on
wednesdays,
okay.
A
So
the
current
proposal
is
to
conduct
it
on
24th
june,
so
it
will
be
an
on-demand
webinar,
so
you
can
watch
it
later
on
the
on
the
cncf
website
or
on
their
youtube
channel,
and
it
will
be
about
measuring,
measuring
and
how
it
can
be
used
to
manage
service
measures.
So
look
out
for
that.
A
Okay,
back
to
julie,
you
know:
do
you
want
to
talk
about
ui
contributions
and
figma
contribution.
B
Yeah
so
on
vendor's
last
note:
there
are
more
opportunities
available
to
us
collectively
about
things
like
the
cncf
webinar
than
we
have
people
to
give
them,
and
so
there's
a
content
calendar
a
spreadsheet.
That's
in
the
google
drive
that
generally
tracks
well,
they
think
where,
where
we
are
intentionally
being
social
and
where
we're
intentionally
marketing
the
projects,
so
if
you're
into
any
of
these
things,
you
know,
say
something
and
there's
different
venues
and
forums.
You
know
we
often
give
talks
at
conferences.
We
do
webinars
like
this.
B
We
can
do
panels,
we
can
do
blog
posts
on
various
venues,
so
it
was
a
lot
so
but
we're
not
doing
a
whole
lot
of
it.
You
can
see
that
there's
any
number
of
newcomers
to
the
community
and
all
newcomers
are
welcome
of
any
shape
and
size
and
form.
Most
of
them
are
of
the
shape
and
size
of
form
of
contributors.
B
Beautiful,
what's
not
happening
nearly
as
much,
though,
is
users-
and
I
don't
know
if
you're
like
me,
but
I
like
to
get
feedback
on
work
that
I
do
it's
even
if
it's
you
know
harsh
feedback
like
that's
better
than
none.
At
all
point
is:
we
are
a
little
starved
for
users,
so
getting
some
of
the
word
out.
That's
the
goal
is
to
get
some
cranky
you
some
patient,
but
maybe
cranky
users.
B
You
know
people
who
want
to
go
along
on
the
journey
of
discovering
bugs
with
us,
but
but
putting
some
things
to
use,
giving
some
feedback
so
so
anyway,
if
you're
interested
in
this
stuff,
you
know
please
please
say
something
and
on
the
next
topic
I
figured
we
had
a
number
of
newcomer
of
people
who
are
here
who
are
saying
things
like
josh
had
said
the
first
time
that
he
joined
and
he
got
pounced
on
right
away,
and
that
was
something
like
I'm
into
front
end.
I
do
visual
things.
B
I
do
javascript
things
and
now
josh
is
a
mesh
mate
and,
and
now
josh
was
the
almost
singularly
not
not
entirely,
but
almost
singularly
the
guy
behind
disqus.layer5
so
really
really
cool.
But
there's
a
number
of
us
who
are
here
and
there's
all
kinds
of
things
to
design
in
measure
ui
in
particular
it
it's.
It
needs
an
overhaul.
It's
not
terrible.
B
To
improve
upon
it
means
that
we
need
to
give
it
relatively
deep
thought
we
need
to
be
relatively
conscientious
and
thoughtful
about
the
user
flows
and
what
people
are
trying
to
accomplish
and
how
consistent
we
can
make
the
interactions,
how
we
can
set
a
paradigm
and
reuse
that
same
paradigm.
How
we
can
you
know,
I
won't
anyway,
as
multiple
contributors
come
in
well
and
the
tool
that
you
know
we're
using
for
better
or
worse
generally,
it's
probably
for
better.
B
It's
lacking
a
few
things
like
every
tool,
but
it
is
figma
and
it's
like
writing
code.
The
unstructured
in
some
sense
that
you
can
do
designs
in
a
bunch
of
different
ways,
and-
and
so
we
figured
it
would
be
helpful
to
establish
just
a
handful
of
conventions
about
how
it
is
that
multiple
fingers
can
be
in
the
same
pie
and
not
not
collide,
so
we
don't
need
to
so.
B
B
You
should
adhere
to
these
conventions
where
you're
not
bumping
into
other
people
where
you're
anyway,
here's
an
initial
set
of
conventions
intended
for
us
to
be
able
to
have
many
people,
many
fingers
in
the
pie
and
still
have
something
that
tastes
good
when
we're
done
so.
The
the
next
item
on
the
agenda
was.
B
Oh
so
we
we
mentioned
a
little
bit
earlier.
If
you
go
to
that
link,
there's
we've
had
in
google
docs
and
in
markdown
files
in
the
repositories,
the
contributor.md,
the
proverbial
well
the
literal
and
proverbial
how
to
contribute
to
the
projects.
B
And
you
know
each
of
the
components
require
some
some
common
contribution
conventions,
but
but
each
of
them
also
are
deep
enough
that
they
they're
their
own
areas,
they're
their
own
components,
and
so
so
we've
been
expanding
and
public
publishing
out
of
those
google
docs,
I'm
here,
and
some
of
that
is
around
convention.
B
B
B
B
Do
we
have
there's
one
other
on
the
agenda,
and
this
is
this
is
certainly
community
oriented,
so
I
didn't
put
a
hyperlink
because
you
know
just
toss
it
on
real
quick.
We
use
a
number
you
know
so
so
when,
when
people
join
the
community
you're
most
of
you
filled
in
the
same
form
and
the
form
will
ask
you
things
like
what
are
you
into?
What
are
you
passionate
about?
B
What
are
you
here
to
to
do,
and
and
some
of
that
one
of
the
check
boxes
is
like
here
to
be
a
community
manager
and
there's
like
at
one
point:
there
was
like
30
of
you
that
are
sitting
in
a
community
manager
channel
there's,
like
20
of
you,
now
they're
fairly,
actively
helping
steward
the
community,
helping
make
sure
that
everyone
has
a
you
know
a
good
time,
and
so
we
use
a
couple
of
tools
to
help
with
the
automating
of
community
management
and
one
of
the
ones
we
were
just
exploring
around
with
is
well
it's
it's
a
young,
it's
a
startup
young
tool,
it's
called
orbit
and
I
think
that
their
site
is
orbit.love
and
it's
about
community
management.
B
They've
automated
a
few
things
that
we
already
automated,
and
so
we
were
just
exploring
the
tool-
and
I
figured
I'd-
call
it
out,
there's
a
few
more
seats
in
the
tool
for
people
to
come
and
take
now
that
I'm
saying
this
out
loud.
That
would
be
other
community
managers
could
have
those
one
of
the
things
that
have
been
suggested
recently.
B
That's
why
the
channel
is
private.
That's
where
the
community
managers
hang
out
and
we'll
talk
all
about
all
the
rest
of
you
in
the
best
of
ways,
trying
to
make
sure
that
people
are
having
a
good
experience.
People
are
unblocked,
people
are,
and
so,
if
you,
I
guess,
my
point
is,
if
you
want
to
tr,
so
I
just
want
to
be
transparent,
hey
we're,
trying
out
orbit,
there's
no
there's
nothing
to
really
be
transparent
about
is
just
sort
of
generally
sharing
we
use
tools
and
community
management.
Here
is
a
thing.
B
Hopefully
it
feels
like
it's
a
thing
and
orbit
itself:
it's
kind
of
like
a
segment
if
you're
familiar
with
that,
it's
just
it
helps
you
intertwine
connect
to
your
slack,
connect
to
your
twitter
connect
to
your
github
and
then
help
show
you
activities
that
are
going
on
across
them.
It'll
connect
your
discourse,
so
the
disqus.layer5.oh
that's
an
instance
of
discourse,
so
it'll
connect
to
those
four
things
and-
and
maybe
we'll
have
to
share
some
reports
about
activities,
and
you
know
some
data
from
orbit.
If,
if
we
continue
to
use
it.
A
A
Yeah
thanks
everybody
like
it,
was
a
productive
meeting.
I
guess
and
yeah
we
will
see
you
next
week
have
a
great
weekend.