►
From YouTube: Meshery CI Meeting (April 8th, 2021)
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Sorry
for
everyone-
and
there
was
a
conclusion
in
the
in
the
meeting
hour
in
the
in
the
meeting
time
and
justly-
and
I
discuss
about
these
topics
that
I
will
share,
but
I
will
do
a
quick
review
about
we,
the
things
that
we
were
talking
about,
and
the
first
thing
we
talked
about
was
the
iso
cook.
Work
for
automatic
releases
on
hand,
charts
that
is.
A
A
A
A
He
is
doing
a
github
action
workflow,
and
for
this
every
time
we
publish
a
new
release
of
mercury,
there
will
be
a
new
version
of
helm
chart.
Also,
so
isuku
is
not
here
because
he's
at
a
different
time
zone,
but
with
this
pull
request,
we
will
have
autumn
automatic
helm,
charts,
and
that
is
for
the
first
topic
that
we
talked
before
the
second
one
is
something
that
we
do
about
windows
support,
and
this
is.
A
Talking
well,
the
second
item
was
about
supporting
windows
in
memory.
B
Yes,
meshi
doesn't
work
on
windows,
mention
ctr
or
measuring.
Think
is
mercury
ctl
okay,
because
there's
some
things
that
work
on
windows
say,
for
instance,
you
know
the.
B
It's
not
that
it
doesn't.
Maybe
there
are
one
or
two
things
that
don't
work,
but
a
lot
of
the
commands
in
mishri
ctl
do
work
on
windows
and
okay.
Somebody
tested
this
a
few
weeks
ago.
Actually,
maybe
I
don't
remember
but
lee
didn't
you
ask
that
and
you
you
had
a
spreadsheet,
we
had
a
spreadsheet,
you
know
and
then
you
know
they
and
then
somebody.
I
don't
remember
who
it
was.
A
C
B
I
think
maybe
in
that
case
you
know
if
you
can
specify
if
you
can
take
out
the
windows
or
you
know,
make
make
a
comment
there-
that
it's
it's
not
completely
windows
and
make
sure
that
it's
arm
you
know
I
mean
so
that
you
know
that
statement
becomes
more
specific.
It
points
to
arm
otherwise.
You
know
people
may
be
looking
at
the
problem
in
the
wrong
place.
Right.
A
B
Or
maybe
we
should
ask
lee
why
he
maybe
he's
he's
aware
of
something
else
that
we
are
not
if
he
put
that.
D
A
So,
for
the
next
topic
that
we
were
talking
about
is
about
the
auto
publishing
the
release
notes.
Currently,
if
we
release
a
new
version
of
mercury.
A
Yeah
before
a
there
was
an
error
trying
to
running
the
the
the
release
raptor
workflow,
so
I
I
tried
to
to
fix
that
workflow
to
only
be
triggered
on
github
releases,
but
we
have
another
problem.
The
other
problem
is
that
we
need
to
update
a
temporal
draft,
put
request
here
on
every
merged
master,
for
example
a
li
merged
something
in
the
morning
today,
and
this
change,
for
example,
this
404
error,
the
exchange
should
be
in
the
draft.
Pull
request
of
measuring,
so
edition
should
be
posted
here.
A
So
this
is
for
in
the
time
that
we
want
to
release
a
new
version
of
measuring
all
the
changes
should
be
listed
here,
and
we
only
like
edit
this
draft
and
publish
this
release
so
that
that
is
not
working
right
now.
I
will
create
another
workflow
for
this
one,
but
the
thing
that
is
working
now
is
that
we
publish
some
notes
in
in
the
in
the
docs
folder.
I
I
think
so.
A
A
A
And
also
another
thing
is
that
we
should
use
the
release
drafter
patch
token,
instead
of
the
github
token,
because
if
you
see
every
merchant
release,
it
says
that
it
was
made
by
by
lee,
but
this
is
not
necessarily
correct,
because
this
measuring
release
was
made
by
guma.
I
think
so
it
was
made
by
abhishek.
A
So
I
will
change
that
in
order
to
to
replace
this
name
into
layer
5,
for
example,.
A
E
E
Yeah
like
where
we
could
relate
the
documentation
files
that
are
where
supposed
to
be
on
the
maturity
version
at
zero
0.2,
for
example,
and
not
to
have
the
documents
mixed
up
between
versions.
C
E
C
E
C
But
I
think
it
it
is.
It
is
already
started
this.
We
presented
this
last
place
last
week
or
this
week
last
week,
actually
yeah
this
week.
I
I
don't
think
I
have
it,
but
on
the
development
meeting,
so
the
you
can
there
is
this
there
is.
There
are
tasks
in
place,
for
this
help
is
needed.
C
So
maybe
rodolfo
can
being
a
bingo
link
to
give
us
the
information
for
this.
C
E
Okay,
thank
you
very
much
and
also
yeah.
I
I
was
having
the
idea
that
about
the
mature
documentation
versions
on
the
site
on
the
documentation
site,
it
will
have
only
the
documents
about
the
the
version
miner
number,
if
I,
if
I
not
mistaken
so,
for
example,
no,
it
will
be
like
the
documents
for
version
0.1
and
then
the
document
version
for
for
maturity,
0.2
and
such.
E
But
if
we
start
this
way,
I
think
that
we,
we
will
have
only
the
documents
for
version
0.5,
which
is
the
the
current
version
in
the
in
developing
correct.
A
Yeah,
I
I
think
so
I
have
not
a
dig
in
more
into
the
version
of
the
docs
but
yeah.
I
think
that
that
will
be
need
to
be
fixed.
Do
you
mean
this?
These
dogs
right.
E
Yeah
that
right,
that
ones,
so
this
document
has
been
been
updated
with
some
changes
that
have
been
making
the
the
contributors
to
every
one
of
the
measure
is
versions
yeah.
So
so,
but
what
we
we
don't
have
like
a
map
of
which
content
is
about
which
version
and.
A
A
E
F
E
A
He
knows
how
or
knows
more
about
the
dogs,
so
I
think
an
option
should
can
be
asked
today
for
how
to
manage
that
mean
operations
problem
and
and
how
to
tack
every
every
document
for
to
know
which
version
is
related
to
to
that
talk,
but
yeah.
I
think
we
we
should
ask
them
because
I'm
not
sure
how
how
it
works.
E
A
Yeah
sure
this
is:
is
this
one
right
version
of
a
concept:
okay,
yeah!
Let
me
take
a
look
of
this
one
and
try
to
help
asking
the
right
people.
E
B
Right,
I
think
you
know,
when
you
make
a
change
to
the
software
that
affects
the
functioning.
I
think
you
change
the
the
documentation,
and
so
when
this,
when
the
software
that
you
made
a
change
to
goes
into
a
release
version,
then
the
documentation
of
that
functionality
should
also
go
into
the
documentation.
Right
I
mean
I,
I
think
the
two
should
always
be
in
sync
together.
B
Right,
I
think
so,
if
you
have
a
pr,
I
think
that
pr
should
include
not
only
the
functionality
of
the
fix
that
you're
going
to
put
in,
but
also
a
change
to
the
documentation.
So
if
you,
if
you
change,
if,
if
you,
if
you
merge
something,
then
you
should
also
have
a
merge
for
the
documentation
right.
B
If
you
don't,
then
that
means
it's
going
out
of
sync
right,
I
mean
I'm
just
talking
I'm
not.
This
is
not
it's
not
the
policy
or
anything
like
that.
You
know
these
are
just
my
ideas.
I'm
just
talking
about.
E
Yeah
yeah,
of
course
I
thought
I
I
am.
I
think
that
some
of
the
contribution
of
documentation
has
been
made
out
of
sync
for
for
that
words
for
out
of
things
about
the
the
functionality
contributions,
so
some
people
make
the
contributions
and
some
people
make
the
documents,
and
that
has
been
done
in
different
times,
so
maybe
that
maybe
that
that
could
be
done
in
at
the
same
time
and
be
synchronized
to
be
automated
in
the
future.
I
don't.
E
B
A
F
Yeah
yeah,
you
guys
are
right.
I
mean
vj
is
correct
that
like
hey,
if
you,
if
you
change
something
and
release
the
software,
if
there's
documentation
about
it,
you
know
it
should
probably
you
know
people
should
be
able
to
that's.
That's
like
yeah.
That
would
be
the
point
of
versioning.
That's
the
point
of
having
the
docs
is
to
have
something
written
about
the
thing
and
then
yeah.
If
you're
gonna
change
it
and
release
and
version
the
version,
the
functionality
version,
the
software
then
yeah.
The
same
thing
goes
for
the
docs.
F
We
ended
up
choosing
some
docs
the
jekyll
theme
that
doesn't
actually
have
versioning
as
a
native
capability,
nor
does
it
have
multi-language
support
as
a
native
capability,
which
means
that
folks,
like
alonso,
are
having
to
build
it
out
as
a
one-off.
There
are
there's
a
lot
of
things
that
can
be
automated.
F
One
of
the
topics
that
we
were
discussing
before
is
the
automated
the
automated
drafting
of
release,
notes
that
happens
today
as
we
go
to
make
a
release.
There's
a
very
nerdy
looking
like
explicit,
set
of
release,
notes
that
are
drafted
automatically
to
include
what
pr's
were
merged
and
links
to
those
those
pull
requests.
F
F
Some
of
you
spend
time
to
write
blog
posts
that
fully
describe
the
feature
and
kind
of
you
know,
take
people
through
what
that
feature.
That
functionality
is,
which
is
a
blessing
when
you
do
that.
It's
really
helpful
to
the
project
to
vijay's
point
like
you
should
yeah
hey
if
like
for
most
features,
not
all
but,
like
you
know,
your
stereotypical
feature
that
comes
through
in
a
piece
of
software
oftentimes,
there's
a
back-end
capability,
there's
a
front-end
capability,
maybe
there's
an
api
to
it
and
maybe
there's
some
documentation
to
it.
F
Maybe
there's
a
user
experience
that
goes
with
it
and
some
mock-ups
like
actually
for
any,
given
even
a
small
feature,
there's
a
lot
of
times
any
number
of
engineering,
artifacts
or
product
like
there's.
There's
road
map,
there's
product
management,
artifacts,
there's
ux,
artifacts,
there's
user
acceptance,
tests,
there's,
there's
testing
artifacts,
there's
lots
of
artifacts
that
comprise
like
a
single
user
story.
If
you
would
or
a
single
issue
in
github's
nomenclature,
and
you
wouldn't
believe
the
barrier
to
getting
people
to
try
to
contribute
or
be
successful
contributing.
F
F
Elizabeth
is
here
she's
just
to
use
her
as
an
example
she's
making
waves
already
she's,
building
building
jekyll
sites,
she's
fixing
issues,
you
know,
she's,
being
she's
being
productive
and
anyway
I
guess
I
I
was
going
to
use
her
as
an
example
to
say
and
of
the
thing
that
she's
documented
the
documentation
that
she's
updating
or
like
it's
just
a
lot
of
coordinate
like
doing
all
that
and
coordinating
like
that,
the
docs
would
land
in
the
same
pr
that
does
the
back-end
functionality.
F
Does
the
front
end
like
that
documents,
the
ui
that
the
documents
the
the
rest
api?
What
have
you
like?
There's
just
no
way
in
hell,
that's
happening
or,
like
I
mean
we
just
hopefully
someday
that's
what
happens
when
you
start
paying
people
to
be
here
to
do
all
the
things
like
there's
a
reason
why
we
don't
have
we
have
a.
F
We
have
a
desire
for
a
master
test
strategy
and
a
document
that
says
there
should
be
words
inside
this
master
test
strategy
and
until
someone
intrigued
enough
like
like
stephen,
comes
along
and
says:
well,
hey,
you
know,
I've
got
some
additional
fodder
that
can
be
used
here.
Here's
a
some
test
plan
examples
that
gets
fairly
specific.
Maybe
this
structure
would
work
well,
let
me
start
building
that
out
until
people
start
getting
interested
in
having
these
like.
It's
an
awesome
thing
to
see.
F
12
15
of
you
on
the
call
all
on
a
ci
call
which
to
some
people
is
like
super
boring
to
other
people.
It's
really
interesting,
like
there's
a
lot
of
very
interesting
things
in
here.
This
thing
that
alonso
is
doing.
It's
like
one
of
the
challenges
alonzo
is
going
to
have
here
and
to
vijay's
point
is
like
well
vijay.
If
we
first
of
all,
if
we
publish
a
release
of
the
an
entire
documentation
set
for
each
of
the
each
version
of
software
that
measure
releases
yes,
technically,
it
should
be
true.
F
If
you're
changing
code,
you
should
probably
change
docs
if
there's
a
doc
change
affiliated
and
okay.
Well,
where
are
we
publishing
this
we're
publishing
on
jekyll
using
github
pages,
rather
we're
using
jekyll
to
publish
on
github
pages?
How
much
store
that's
free
right?
Yes,
it
is
that's
why
we're
using
it.
F
How
much
storage
do
they
give
you?
Oh
it's
500
megs
a
month,
something
not
a
month
like
500,
megs
total,
or
something
like
that.
I
I
forget,
you
can
look
it
up.
A
quick
search
will
actually
tell
you
okay.
Well,
how
big
is
the
site
all
right?
How
big
is
the
docs
site?
That's
like
54,
megs
or
something
I
forget.
I
don't
know
if
you
multiply
it
out
times
the
number
of
releases
we
have.
We've
got
like
a
hundred
and
something
releases.
F
I
think
because
we,
you
know,
we
have
a
lot
of
very
minor
bug,
fixed
releases.
Okay,
well,
does
that
add
up
and
does
it
fit?
Yes,
if
you
only
publish
them
to
use
semantic
versioning
nomenclature,
if
you
only
publish
the
minor
releases,
v0.1
v0.2.3.4,
if
you
don't
publish
the
got,
the
bug
fix
releases,
the
dot,
the
v0.4.22.
F
If
you
start
to
publish
those
you'll
run
out
of
free
space-
okay!
Well,
it's
not
like.
We
can't
go
pay
for
a
small
s3
bucket
or
something
like
that
to
host
the
other
somewhere
else.
Oh
okay,
anyway,
there's
just
like
a
thread,
a
long
thread
of
things
that
unravel
and
hence
hence
why
a
lot
of
things
kind
of
are
the
way
that
they
are
there's
a
lot
of
fantastic
things
we
have
going
on.
This
thing
that
alonso
is
doing
is
when
he's
done.
I
think
he'll
turn
around
and
he'll
probably
say
that.
F
Wasn't
that
hard
right
now,
it
feels
pretty
hard,
though,
when
you're
done
like
actually
that
wasn't
that
hard
and
we
auto
publish
the
really
the
new
docs
every
time
we
make
a
release
like
actually,
I
feel
really
confident
with
that.
Now
there's
a
workflow
that
auto
publishes
the
docs
and
version
and
then
and
auto
versions
them
too,
and
it's
going
to
be
when
he
goes
to
do
that
somewhere
else,
it'll
be
pretty
fast,
not
having
done
it
before
here.
F
It's
a
lot
of
things
to
consider.
Hence
someone
like
any
rude
jain,
who
is
kind
of
on
on
point
on
this
before
gave
up,
and
despite
a
lot
of
support
and
like
many
hours
of
conversation
with
me,
one
of
the
things
that
I
made
a
mistake
and
continue
to
make
mistakes
on
is
not
writing
stuff
down.
It's
like
talking
to
folks
and
not
writing
our
conversations
down
it's
one
of
the
pains
in
the
rear
about
slack.
F
I
can't
tell
you
how
many
things
we've,
sometimes
we've
architected
or
designed
something
in
slack
and
it's
gone.
So
I'm
pleased
that
you
all
have
access
to
google
docs
you're
all
entitled
to
edit
the
docs
create
new
docs.
I
think
by
default
only
those
that
have
been
here
long
enough
and
have
come
into
a
community
manager
role
or
a
maintainer
role
can
delete
docs,
but
don't
let
that
stop
you
from
creating
them
steven
just
created
quite
a
few.
F
So
anyway,
I
was
sorry.
I
was
off
and,
like
that's
such
a
distraction,
such
kind
of
a
a
question
that
doesn't
like
anyway,
the
the
thing
that
I
only
heard
the
last
little
bit
of
what
you
guys
were
all
talking
about.
So
I
hopefully
that
those
thoughts
helped.
I
didn't
want
to
pretend
to
provide
all
the
answers
or
what
have
you,
but
I
think
you
well
we're
on
the
right
path.
What
I
had
heard
folks
talking
about.
A
B
Let
me
ask
a
question:
the
question
is:
when
you
put
dogs
into
github,
it
is
what
is
being
put
there
from
one
version
to
another.
Is
it
the
difference
or
is
the?
Is
it
the
whole
documentation
all
over
again
for
a
new
version?
F
F
Oh
man,
I
was
trying
to
be
explicit
elizabeth
and,
like
give
you
all
the
details-
and
I
think
in
trying
to
do
that,
I
like
didn't,
really
just
give
a
clear
answer,
but
one
but
elizabeth's
question
on
how
docs
work,
how
jekyll
works
relates
to
what
vj
is
asking
which
is
which
relates
to
how
git
works.
F
Git
is
a
versioning
system,
it's
worn
out
over
subversion
and
cvs
and
other
versioning
systems
for
some
reasons
that
I
won't
digress
into.
But
we'll
say
that.
F
Like
why
would
if
your
public,
if
the
docs
are
stored
in
git,
and
this
there's
this
versioning
system,
that
does
this
these
intelligence
things
around
deduplicating
text-based,
you
know
text
which
is
all
of
what
our
source
code
and
things
are,
and
even
images
as
well.
Why
would
having
multiple
versions
of
the
docs
published?
F
That's
really
just
source
code,
underneath?
Why
would
each
of
those
independent
versions
count
against
us,
like
in
terms
of
the
storage
space,
because
if
it's
stored
in
git-
and
it's
just
if
there's
if
git
is
just
using
pointers
to
point
back
your
if
get
it?
If
I'm
trying
not
to
make
this.
F
Any
anyway
vijay
the
reason
that
it
costs
that
that
it
that
publishing
a
version
of
the
docs
takes
up
the
full
space
of
the
docs
is
because
is
because
each
of
those
are
statically
compiled,
go
off
generate
that
site.
The
docs
store
that
output
do
the
next
one
to
the
next
one
so
that
they
can
be
referenced
because
the
docs
themselves
they're
not
dynamically
built.
F
Yeah
just
explicitly
it's
just
measuring
ctl
something
that
had
happened
when
we
made
the
the
v
this
last
release
in
v05.1,
which
was
just
like
six
days
ago,
or
something
ish
that
the
to
to
build
mastery
ctl
that
go
program
to
help
us
make
a
build
of
the
mesri
ctl
binary
and
to
have
a
version
of
that
binary
produced
that
runs
on
linux
runs
on
mac,
runs
on
windows
and
runs
on
windows
on
intel
windows
on
amd
windows,
on
arm
linux.
F
On
arm
linux,
on
intel
linux,
you
know,
produces
all
those
versions
of
that
support,
different
os's
and
different
architectures,
so
os
being
linux,
mac
windows,
architecture
being
like
32-bit,
64-bit
arm
versus
86
x86,
we
use
a
a
go
based
utility
called
go
releaser
to
help,
take
care
of
all
that,
like
cross-compilation
of
golang
to
different
architectures
different
systems
and
and
it
unbeknownst
to
us
at
the
time
that
we
made
the
v051
release,
there's
a
package
in
golang
that
I'm
going
to
explicitly
state.
But
no
one
should
remember
it's
not
it's
not!
F
There's
a
there's.
A
package
dependency
error
in
a
common
golang
package
called
memory
map
map.
That's
having
problems
with
compatibility
of
windows
and
arm
systems,
so
it's
not
related
to
us.
It's
nothing
that
we
wrote,
but
the
ci
system
go
releaser
that
we
use
this
project.
We
collectively
all
of
you
use
to
build.
Mastery
cto
is
having
a
problem
is
broken,
and
so
the
thing
is
is
the
workflow
that
creates
all
those
individual
versions
of
measuring
cpl.
It's
not
like
it
intelligently,
says:
okay!
F
F
So
today,
if
you're
any
of
you
are
running
measuring
you're
running
the
latest
stuff,
the
latest
version
out
there
is
measuring
server
5.1.
The
latest
client
is
only
5.0.
There
is
no
5.1
because
of
that
because
of
this
issue.
So
what
we
did
is
because
that
memory
map
package
dependency,
issue
on
windows
and
arm
is
like
such
a
massive
distraction
for
us
to
try
to
go
figure
out.
It's
not
even
like
we're
going
to
go
figure
it
out
and
fix
it.
F
It
will
be
fixed
in
time
and
since
we
don't
have
users
here,
beating
down
our
doors,
saying
we
run
windows
and
we're
on
arm
and
we
love
measuring.
We
we
want
measure,
ctl
and
windows
and
arm
we
just
like
temporarily.
Let's
just
add
this
architecture
and
operating
system
combination
to
an
ignore
list,
so
then
that
way,
next
time
we
make
a
release
which
we
need
to
do
today.
F
I'll
tell
you
why
that
it
would
work,
it
would
just
ignore
those
that
type.
F
F
There's
this
bash
script
that
you
can
run
when
the
bash
script
runs
it
reaches
out
to
github
identifies
the
latest
release,
pulls
down
the
latest
meshery
ctl.
F
F
Oh
okay,
yeah
rudolph.
I
don't
know
if
you
like
two
things:
if
one,
if
you
can
go
to
the
releases
section
of
mystery
and
we
go
to
our
latest
release,
5.1
is
there
any?
Are
there
any
artifacts
in
there
to
download?
There's?
There's
draft
really,
there's
release
notes,
that's
great,
but
then
these
artifacts,
no
there's
no
pre-compiled
mesh
receipt.
So
when
someone
tries
to
install
mesh
for
using
the
bash
script
right
now,
they.
F
So
normally
you
have
all
these
choices,
but
so
that's
why
we
need
to
make
a
release,
because
the
bash
script
fails
to
install
it's,
not
that
you
have
to
be
running
the
same
version
of
mesri
ctl
that
client
and
the
same
version
of
mesh
reserver.
They
don't
have
to
be
the
same
necessarily
I'm
running
mesh
renow
and
I'm
using
s3,
ctl,
5.0
and
measuring
server
5.1.
That's
just
fine!
F
But
if
you're
a
brand
new
user-
and
you
run
that
that
bash
script
that
rudolph,
if
you
go
to
measuring.io
and
you
click
on
getting
started,
this
script
won't
be
able
to
pull
the
yeah.
If
people
say.
Oh,
the
they've
got
choices
here,
but
if
they
choose
this
first
one
it'll
error
out
on
you,
because
it
will
look
for
that
latest
release
and
won't
find
any
files.
F
Hence
rudolfo:
do
you
mind
making
a
release
while
we're
all
here,
so
that
all
of
us
can
know
how
to
do
a
measuring
release.
A
F
And
normally
that
would
be
like
this
is
it
this
is
a
glitchy
release,
but
that
normally
it
would
be
filled
in
and
the
only
other
thing
for
everyone
else
to
hear
to
take
away,
there's
kind
of
a
couple
of
interesting
things.
One
is
that
all
right?
Well,
that's
just
the
name
of
the
release.
It's
just
there's
actually
nothing
meaningful
about
it,
we're
just
being
consistent.
That's
all
just
convention.
F
What
is
meaningful
is
that
other
field
that
tag
that
that
that
will
be
a
specific
string
and
we're
following
semantic
versioning
that
there's
a
hyperlink
on
that
right
hand,
side
for
people
to
go,
read
about
that,
but
we're
just
following
a
specific
way
that
you
version
your
software,
we're
assigning
a
new
tag
and
aligning
it
to
the
master
branch,
the
main
the
main
line
branch
and
that's
actually
it
to
make
the
release.
Rudolph
will
end
up
just
clicking,
publish
this
release.
A
Yeah,
I
will
do
it
sounds
good,
and
this
is
this
will
be
trigger.
Some
actions
right
should
be
yeah
sure.
Do.
F
Oh,
you
know
what
oh
my
gosh,
there's
like
three
bugs
here
all
based
on
that
that
last
change.
Actually,
the
the
other
bug
is
that
this
is
a
bad
example
for
everyone,
because
none
of
this
does
anyone
need
to
remember.
Normally
you
just
go
in,
and
click
publish
you
serious.
You
should
literally
just
press
edit
and
publish
that's
it
there's
a
bug
here
because
of
that
personal
access,
token
that's
being
used,
it
doesn't
have
the
correct
permission.
F
A
F
A
F
F
A
I
think
yeah,
okay,
great
hey,
save
draft
prayers.
I
think
so
publish
directly.
F
F
It
might
be
that
so
you
did
make
a
release
that
were
the
the
reason
that
the
workflow
isn't
kicking
off.
If
you
do
it
again,
I
refresh
again
like.
F
Might
be
that
we
have
to
delete
the
tag
and
start
it
with
a
new
tag.
I
don't
know
maybe.
F
But
yeah
if
yeah.
A
F
F
F
F
Yeah
there
will
be
at
least
two
kubecon
talks
about
measuring
and
that's
there
will
be
a
third,
that's
will
have
s
service,
mesh
performance
and
get
nighthawk
mentioned,
and
then
there
might
be
more
that
I'm
just
not
tracking.
F
But
that's
a
great
question
or
something
to
talk
about
on
tomorrow's
community
call
and
something
that
rudelpho
and
I
were
talking
about.
Novendou
gave
a
talk
on
meshary
today
at
a
virtual
conference.
A
Yeah,
do
you
want
to
do
you
want
me
to
do
it
right
now?
Yeah,
if
you
don't
mind
just
like,
because
it'll
just
be
really
quick:
okay,
yeah.
F
A
F
E
Yes,
of
course,
nice
all.
F
F
Anyway,
I
yeah
yeah,
it
seems
he's
working.
F
C
F
Yeah,
there's
there's
a
so
a
friend
of
mine
had
put
on.
He
teaches
people
about
kubernetes
he
put
on
a
podcast,
his
podcast
was.
It
was
like
this,
this
pub
quiz
where
we
talked
for
a
little
while
kind
of
like
a
kind
of
like
a
podcast,
and
he
interrupts
and
gives
people
like
he
had
quit
these
these
questions
on
kubernetes
and
he
had
made
them
up
about
26
of
them
and
some
some
of
the
questions
were
like
hey
of
the
kubernetes
logo,
the
blue,
that's
in
the
kubernetes
logo.
F
A
F
Hey
since
we
have
like
four
or
five
minutes,
I
know
that
you
know
joy
and
elizabeth
are
top
of
mind
for
me
and
just
in
making
sure
that
they
are
that
each
of
you
are
warmly
welcomed
that
you're
getting
your
questions
answered,
enjoy
your
topic
of
conversation.
Earlier
this
morning
we
were
talking
about
you
behind
your
back
in
the
best
of
ways,
there's
a
meshry
ctl
command.
F
That
is
collecting
dust,
and
I
don't
know
that
it
works.
I'm
going
to
slack
you
and
a
couple.
I
don't
know
if
I
should
use
that
as
a
verb,
slack
you,
but
I'm
we're
gonna
exchange
some
chats
and
slack
and
see
if
we
can
introduce
you
to
this
lesser
known,
mastery,
ctl
command
and
see.
If
you
can
help
us
popularize,
it.
A
I
think
it's
everything
about
this
two-hour
meeting
ca,
but
yeah
it
it
went
well.
So
I
think
we
can
adjourn
thanks
everyone
for
attending
and
if
you
have
more
questions,
please
do
it
in
slack.