►
From YouTube: Meshery CI Meeting (May 13th, 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
A
Yeah
yeah
yeah
exactly
it's
a
green
screen
back
there
and
I'm
wearing
a
I'm
wearing
a
hoodie
that
I
got
for
speaking
at
engine
x,
conf.
I
don't
know
four
years
ago
or
so,
and
it's
all
green.
A
A
A
A
A
Good
deal,
did
you,
do
you
want
to
say
hi
to
everybody,
just
tell
them
what
your
last
name
is
and
a
little
bit
about
yourself.
B
A
Perfect
perfect,
very
nice
to
meet
you
mancha.
You've
got
a
couple
of
questions
for
you
in
the
chat
I
think
about
about
how
much
of
a
pain
in
the
butt,
I
suspect
the
required
last
name
field
on
certain
forms
is
for
you.
What
do
you
end
up
putting
in
there
when
they
force
you
to
fill
it
in
and
they
don't
take
a
white
space.
C
C
The
first
topic
for
this
meeting
is
a
possible
flaky
test
that
was
reported
by
lee,
and
this
is
on
dr
hop
and
we
mentioned
that
he
just
kicked
the
test
and
it
passed
again.
So
I
think
maybe
we
can
supervise
that
test
in
in
future
releases
in
order
to
know
if
it
is
a
flaky
test
or
is
a
failing
test
and
for
those
that
didn't
know
or
doesn't
don't
know
what
is
a
fluke
test.
C
Aflacitas
is
a
test
that
sometimes
fails
and
sometimes
a
pass
and
a
failing
test
is
a
test
that
it
always
fa.
Is
it
always
failing?
So
I
I
will
take
a
look
on
this
test
these
days
and
see
what
is
happening,
and
if
it
is
a
failing
test,
I
will
submit
a
request
to
to
fix
that.
A
Yep
near
as
I
can
tell
it's
just
a
fluke
like
it
seemed
to
work
the
next
time
that
we
did
a
build
so.
A
That
seems
to
happen
with
some
regularity
in
github
github
workflows
that
the
runners
will
run
into
you
know
some
some
issue
will
occur
and
they'll
they'll
fail
fairly.
Randomly
I
mean
not
like
with
like
a
fair
with
like
to
a
fairly
high
degree,
actually
considering
that,
how
many
times
I've
seen
that
so.
C
A
Call
nope
but
looks
like
he
had
a
good
question
here
so
I'll
I'll.
Just
answer
that
in
a
comment.
C
C
This
topic
is-
or
you
might
have
more
context
about
this,
and
we
want
to
implement
this
test
to
be
included
in
the
ci
test
in
the
in
the
workflow
of
measuring.
So
one
question
that
I
have
abhishek
is
like:
how
do
you.
E
E
C
E
Oops
yeah
so
just
go
quickly
and
then.
E
So
there
are
two
scenarios
here
after
you,
log
in
basically
operator
gets
installed
automatically,
but
if
you
want
to
do
it
manually,
you
have
a
you,
have
an
option
in
the
settings
pane.
I
just
have
to
enable
this
to
install
an
operator,
so
the
automatic
installation
kind
of
is
a
little
bit
more
time
consuming
than
if
you
go
quickly
and
install
it.
So
so
I
can
see
that
my
cluster
is
also
configured
I'm
going
to
install
it,
and
once
I
install
it,
I
get
all
the
status
in
here.
E
One
of
them
is
this:
basically,
installing
it
manually,
other
one
is
just
like
being
in
the
page
and
it
just
automatically
installs
it
and
more
scenarios
similarly
are
listed
on
this
issue
like
the
one
you
just
displayed
so
and
basically
status
is
displayed
in
here
after
it
gets
installed
in
in
the
background
under
the
hood,
you
can
just
go
to
measuring
namespace,
so
I'm
in
machine
namespace.
Currently,
if
I
list
list
my
parts
or
let's
say,
yeah
parts
is
better.
E
E
E
E
C
Do
you
consider
that
I
should
test
that
the
pot
was
deleted?
Also,
sorry,
do
you
consider
to
add
a
test
that
verifies
the
list
of
the
bots
and
then
they
activate
the
measure
operator
and
then
validate
if
there
are
no
votes
anymore.
E
We've
listed
a
couple
of
scenarios
like,
for
example,
considering
the
operator
on
and
user
may
wish,
to
switch
to
a
different
cluster,
or
so
that's
a
scenario
and
like
basically,
a
list
of
all
scenarios
are
here
and
also
added
with
time
in
the
respective
comments
that
follow
so
so,
basically,
the
initial
step
would
be
to
just
test
the
on
off
like
have
a
unit
test
to
test
the
on
and
then
cover
like
basically
keep
adding
more
scenarios
to
cover.
C
Yeah,
so
it's
a
good
plan.
I
will
start
with
switching
between
on
and
off
and
then
we
can
continue
from
that
sure
yeah
great
queue.
I
will
work
on
that
this
weekend.
A
Sorry,
it's
sorry
to
interrupt
on
rudolph.
Actually,
maybe
one
that's
somewhat
relevant.
A
We've
spoken
about
it
on
this
call
in
the
past,
and-
and
it
is
the
notion
that
we
would
eventually
you
know,
we
have
a
this-
build
and
release
strategy
document
that
talks
a
lot
all
about
our
release
process
and
our
workflows
and
and
stuff
like
that,
and
we
we'd
also
talked
about
this
test
strategy
document
that
would
outline
high
level
like
our
approach
to
unit
testing,
functional
testing
and
then
testing
what
tools
we
might
use
so
why
we
would
use
them
in
what
fashion
and
that
kind
of
a
thing
and-
and
we
have
we've
only
made
it-
we've
made
it
not
very
far
at
all
in
that
one,
and
so
all
of
you
that
are
on
the
call,
are
welcome
to
jump
into
that.
A
There's
a
link
in
this
the
meeting
minutes
for
sure
to
the
to
what
is
mostly
an
empty
document
on
test
strategy,
but
the
subsequent
doc
after
the
test
strategy,
the
strategy
itself
would
just
talk
about,
like
oh
we'll
use,
cypress
and
we'll
use
github
action,
we'll
use
all
you
know,
we'll
do
all
these
things,
but
then
there
needs
to
be
a
a
test
plan
like
that.
A
A
It
might
be
the
case
that
it
might
be
the
case
that
we
would
begin
to
institute
well
like
a
once.
Every
three
month,
release
party,
like
bug,
smashing
sort
of
you,
know,
release
preparation
party.
A
If
we
can
think
we
have
to
use
that
term
just
to
attract
people
to
come
and
actually
that
type
of
a
session,
let
me
find
the
link
but
that
that
type
of
a
session
is
really
kind
of
nice
for
a
lot
of
well
newcomers
in
part,
because
anyone
can
help
anyone
can
just
be
a
user
and
find
bugs
and
complain.
A
A
Weird,
okay,
so
if,
since
I
totally
derailed
the
conversation,
if
you
don't
mind,
maybe
pulling
up
that
spreadsheet
just
momentarily
to
orient
everyone
to
it,
it
is
well
there's
a
couple
of
things
about
it.
So,
first
of
all
it,
this
is
a
call
for
participation.
A
This
is
a
new
spreadsheet
that
you
know
it's
not
overly
intelligent.
It's
it's
a
list
of
tests
to
perform
for
any
of
us
to
perform
to
verify
that
certain.
You
know
so
so,
like
some
of
this,
like
eventually
cyprus,
if,
if
rodolfo
is
creating
new
endn
tests
using
cyprus,
well,
it'll
it'll
automate,
like
line
number
three
row
number
three
there,
which
is
there's
a
category
of
tests
that
are
about
life
cycle
management.
A
Like
you
know,
managing
the
life
cycle
of
either
mastery
or
of
service
meshes,
and
the
operator
is
a
component
that
we
in
this
particular
case.
We
want
to
focus
on.
Does
the
mastery
operator
deploy
into
kubernetes
so
I'll?
Let
you
guys
you
know
everyone
read
it,
but
so
there
was
a
collection
of
you
know,
four
or
so
maintainers
this
last
weekend.
That
was
just
walking
through
those
walking
through
you
know.
A
We
when
we
made
the
v050
release
significant
release,
lots
of
new
architectural
components,
a
natural
fallout
of
that
is
that
there's
going
to
be
a
few
bugs
and
so
sort
of
a
quality
orientation.
For
a
little
while
and
some
point
releases,
you
know
to
patch
those
bugs
really
appropriate.
So
so
two
things
all
right.
Okay,
wait,
three
more
things
to
say
so
this
this
mystery
test
plan.
A
It
is
marked
for
v060.
It's
like
it's!
So
it's
one
of
those
documents
you
might
find
multiple
copies
of
or
like
in
the
sheet
how
it
has
sheet
1
at
the
bottom.
Maybe
we
would
end
up
having
a
v060
and
a
v070
and,
like
you
know
it
would
be
the
point
in
time.
Are
these
tests
passing
the
tests?
Will
change
and
grow
as
more
features
come
out
and
some
of
the
tests
will
just
the
goal
being
like
I
was
saying
before
a
lot
of
the
tests
would
be.
A
A
Well,
speaking
of
actually
I
should
I
should
say
this.
I
know
navendu
navendu,
one
of
the
maintainers
on
mastery
ctl.
He
is
calling
for
and
you
you
all
he'll
he's
calling
for
a
meeting
on
saturday
to
do
to
really
do
this
specifically
to
mastery
ctl.
I
expect
he
will
probably
use
the
messerie
ctl
command
tracker,
there's
a
separate
sheet
that
calls
out
like
every
single
sub
command
and
all
that
and
and
so,
if
you're
subscribed
to
the
community
calendar
you'll
see
it
on
the
calendar.
C
Yeah
sounds
good,
so
the
link
is
posted
here
and
I
I
will
be
waiting
for
that
invitation
on
saturday
to
be
joined.
If
it
is
possible,
okay,
so
and
passing
to
the
next
item
in
the
list,
there
is
a
releasing
home
from
abhishek.
I
think
there
is
a
pull
request
open
right,
abishi.
A
E
Oh
yeah,
sure,
okay,
so
so
so
I
guess
the
ca
is
currently
broken.
With
respect
to
helm,
just
the
helm,
release
and
chat,
release
and
asuco
is
I'm
just
pulling
up
the
pr.
So
basically,
asuka
is
working
on
a
custom
gate
of
action
or
to
to
sort
of
release
these
charts
and
just
a
second
and
just.
A
Refreshing,
okay,
I'll
toss
in
a
couple
of
comments
as
well,
which
is
to
say
that
we
have
well
actually
in
some
respects,
we've
offered
people
a
lot
like
too
many
choices
for
how
they
can
install
measuring.
Do
it
on
docker.
Do
it
in
kubernetes:
do
it
using
helm
using
the
master
ctl
using
get
get
some
brew,
get
some
scoop,
use
a
bash
script
and
whatever
other
way
that
we've
been
asked.
So,
like
almost
you
know,
it's
good
it's
healthy!
A
Eventually,
as
we
grow
and
grow
and
grow
people
have
preferences,
they
have
different
systems,
they
need
to
integrate
into,
they
use,
helm
and
only
home.
So
they
need
that,
and
so
fine,
I'm
good
we've
in
vet.
We've
done
that
investment,
one
of
the
every
time
that
there's
a
new
release
of
mesh
reserve
of
mescheri
the
whole
thing.
There's
a
new
release
of
mystery
ctl,
a
new
brew
package
is
released.
A
new
scoop
package
is
released
great
well.
What
isn't
automatically
released
is
a
new
helm
chart.
A
It's
only
ever
been
manually
released
and
you
know
stuff
changes
and
when
there's
no
reason
not
to
have
it
in
the
regular
workflow
for
all
of
the
build
artifacts
like
we
have
an
existing
chart,
it
should
be
built
with
anything.
That's
been
changed
and
published,
and
so
there's
a
suco
had
forked
a
general
github
action
for
building
helm,
charts
a
helm
chart
releaser.
A
A
We
publish
the
artifacts
for
helm,
I'm
sorry
the
artifacts
for
that
chart
the
built
artifact
for
that
chart.
We
published
it
into
a
different
repo.
Why?
Well?
That's
because
we
publish
it
into
a
different
repo,
because
when
you're
using
the
command
helm,
helm
install
such
and
such
well
helm
as
a
tool
will
reach
out
to
the
internet,
to
some
http
endpoint
to
some
repo.
A
It
doesn't
have
to
be
a
github
repo
with
just
some
helm,
repo
and
a
public-facing
http
endpoint
and
grab
the
chart
retrieve
it.
Well.
We
already
used
the
meschery
repo
for
docs
dot,
mesherry.io
and
granted.
We
could
host
the
helm,
chart
the
artifacts
in
the
same
docs.mastery
tario
at
the
time
it
was
a
different
setup.
A
We
chose
not
to
we're
hosting
the
we're
hosting
the
built
helm
chart
in
the
mesherid.io
github
repo,
and
so
when
a
user
uses
helm
install
or
when
a
user
visits,
artifactory.I,
sorry
artifacthub.io
and
you
find
the
helm
entry.
I'm
sorry
you
find
the
mastery
entry,
it
will
reference
meshri.io
it'll
like
that's
where
the
helm
chart
is
published.
You
know
the
built
artifact,
and
so
anyway,
the
github
action.
A
A
Like
what's
the
what's
the
it's
our
current
issue,
vishal,
you
might
actually
have
to
speak
to
this,
otherwise,
otherwise
we
might
want
to
reassign
it
like.
Can
you
verbally
speak
to
it.
A
And
so
it's
causing
users
enough
pain
like
there's
enough
people
that
use
helm
as
their
install
procedure
that
or
abhishek.
I
think
you
know.
If
suco,
we
should
poke
him
once
more
today
to
make
sure
he's
actively
iterating
on
it
and,
if
not
vishal,
I
I
get
it.
The
chat
is
a
challenge,
it's
enough
of
a
critical
issue
that
we'll
need
to
we'll
probably
need
to
reassign
so
abhishek
rodolfo.
A
If
we
give
one
more
day
to
asuko
and
to
vishal
and
beyond
that,
we
need
to
get
it
fixed
and
push
it
out.
It's
it's
mostly
about
configuring.
The
github
action
for
helm
chart
releasing
the
current
issue
by
the
way.
To
answer
my
own
question
from
vishal
is
that
it's
improperly
configured
the
github
action
that
runs
that
it's
called
helm
chart
releaser
the
git
of
action
is
failing
because
it
has
a
bad
reference
to
the
remote
repo
and
I
don't
think
github
is
retaining
the
logs
long
enough
to
actually
see
it
here.
A
Hence,
in
part
why
the
screenshots
are
on
the
pr's,
but
but
yeah.
That's.
C
C
A
And
jubril,
while
that's
going
on,
do
you
want
to,
if
you're
going
to
talk
today
about
the
git
of
action
that
you're
working
on?
Do
you
want
to
drop
that
as
an
agenda
item.
F
B
C
C
C
C
C
C
Okay,
so
anyone
have
a
comment
or
question
about
ci
in
general
or
how
to
start
learning
any
github
actions
or
something
like
that.
A
I
have
a
general
comment
actually
about
git
and
the
interactions
with
github.
That
is
mostly
just
to
get
everybody
excited.
I
guess
that
that's
that,
if
you've
been
paying
attention
in
measuring
measuring
will-
or
maybe
maybe
this
isn't
as
obvious.
Yet
some
of
you
have
been
working
on
a
new
two
new
capabilities
within
measuring
one
is
called
patterns.
The
other
one
is
called.
A
Filters
mesherry
will
allow
you
to
load
service
mesh
patterns,
which
will
be
a
yaml
file.
That
is
a
higher
level
construct
than
a
mesh
kubernetes
yaml
file.
It
will
contain
things
like
a
descriptor
for
a
canary
rollout
or
a
descriptor
for
how
to
the
fact
you
want
to
deploy
istio
you
want
to
deploy
an
application.
You
want
to
run
a
performance
test,
and
then
you
want
to
do
a
canary
rollout
based
on
the
performance
of
it.
A
Like
that
whole
thing
in
one
one
yaml
file
and
actually
have
it
fairly,
concise,
there's
a
new
format
that
we're
working
with
it's
called
the
open
object,
open
application
model.
Anyway,
that's
what
patterns
are
about,
you
can
load
and
store
patterns
in
git
in
github
and
meshri
will
connect
to
github
and
pull
in
the
pattern.
A
Same
thing
for
a
filter
filter
is
a
different
thing:
it's
not
yaml.
It's
a
small
binary,
we're
working
with
webassembly
and
and
the
first
couple
that
people
are
working
on
are
they
are
in
rust
and
there's
a
whole
awesome
filters,
repo
out
there
in
the
layer.
Five
org
anyway,
point
is
that
people
are
gonna
store
their
web
assembly
filters
for
envoy
in
something
like
github
or
some
other
git
based
repository,
just
like
we
are,
and
meshri
will
connect
to
github
and
pull
them
down.
A
It'll
also
ongoing
connect
to
github,
if
you
authorize
it
to
and
part
the
one
of
the
big
points.
The
reason
I'm
telling
you
all
this
is
because
there's
this
term
there's
this
buzzword,
git
ops,
it's
been
around
for
a
while.
Well,
it's
been
here
over
a
year
and
it's
not
not
really
a
new
concept.
It
is
so
much
as
it's
just
an
evolution
of
what
a
lot
of
people
have
been
doing
in
the
past,
and
now
it's
got
a
nice
term,
and
so
we
can
all
understand
what
we're
meaning.
A
When
we
just
say
a
quick
term.
Git,
ops
and
part
of
that
is
measuring
is
long
before
we
even
started
the
project
like
it's
been
a
goal
to
be
able
to
facilitate
well
for
people
to
do
visual
design
of
a
service
mesh
and
have
it
integrate
with
git
ops.
So
if
you're
using
measuring
some
of
you,
are
working
on
or
have
seen,
mesh
map
and
whether
it's
mesh
map
or
not,
that
you
can
design
it.
The
configuration
that
you
desire
for
measuring
to
manage
of
your
surface
mesh.
A
A
C
Okay,
jayman
should
say
that
he
want
to
work
on
ci
issue.
So
currently
we
have
two
ci
issues,
I
think
so,
and
the
first
one
is
to
not
run
the
notified
providers
workflow
every
time
sure
sure
eh
after
learning
heimanshu.
So
this
I
think
this
workflow
is
a
a
very
or
this
issue
is
a
very
easy
issue
to
to
solve,
because
you
only
need
to
modify
the
frequency
of
when
this
workflow
is
triggered.
So
I
think
you
can
start
working
on
this
one.
C
The
issue
that
is
related
to
that
is
this
one
notified
remote
providers
and
if
you
go
to
meshali,
github,
workflows
and
notify
providers
here
is
a
jam
l
and
at
the
top
of
this
gmail
configuration
is
the
the
events
when
the
this
workflow
is
triggered.
In
this
case,
it
will
be
triggered
on
every
push
to
master
and
on
every
tag
created
on
github.
C
C
C
A
Yeah
and
so
manchu
is
brand
new
to
the
community
and
so
as
such,
hasn't
earned
his
stripes,
so
to
speak
in
terms
of
being
invited
to
the
github
org,
the
later
five
io
github
org,
and
so
as
such
github's
policy
is
that
you
can't
assign
somebody
something
unless
they're
part
of
the
org
and
so
himanshu.
If
you
comment
on
this
issue,
then
that
unlocks
the
ability
to
assign
it
to
you
so.
C
C
Okay,
so,
okay,
this
one.
A
I
beat
you
to
it
all
right.
I
he's
assigned
mr
n
a
is
assigned.
C
So,
if
you
have
questions
I
mentioned,
please
let
us
know,
let
me
know
in
the
micheli
channel
and
we
can
discuss
there
in
order
to
to
have
the
opportunity
of
others
to
learn.
Also
reading
our
conversation
and
and
this
issue
does
anyone
have
any
other
question
about
this?
One.
A
No
just
that
the
ant,
like
a
lot
of
the
workflows
that
we
trigger,
though
the
workflows
will
define
the
thing
that
they're
interested
in
triggering,
but
a
lot
of
them.
We
don't
we
used
to
like
have
a
secondary
like
anti-trigger
like
sometimes
we
would
trigger
something
and-
and
we
couldn't
find
a
more
granular
way
of
like
saying
not
like
yes,
we're
interested
in
triggering
this
on
push,
but
not
when
such
and
such
is
also
the
case.
So
this
is
good
this.
A
If
you
will
trigger
filters
on
some
of
the
other
flows,
because
we're
sort
of
over
triggering
a
little
bit
today-
and
it's
in
general-
it's
not
that
big
of
an
issue,
because
it's
because
there
there
are
build
minutes
that
that
we
have
left
over
at
the
end
of
the
month,
but
for
one
of
the
repos
we're
actually
burning
through
the
build
minutes,
which
is
actually
exactly
why
we're
working
on
that
issue,
because
we're
built
we're
burning
through
the
build
minutes.
A
We
have
like
15
minutes
left.
I
wonder
if
jubril,
I
know
you
want
to
take
us
through
your
github
action
that
you've
been
working
on
which
stands
to
be.
It
stands
to
be
not
only
used
for
the
intended
purpose
by
which
you're
designing
it
for
smi
conformance.
B
A
F
F
F
A
There's
a
couple
of
comments
on
the
pull
request
that
so
right
now.
The
approach
that
you're
taking
is
to
use
is
that,
in
order
to
test
either
both
test
measuring
functionality
as
well
as
actually
perform
a
task,
and
what
you're
trying
to
do
in
this
case
is
you're
trying
to
perform
a
task
of
running
a
conformance
test
that
you're,
you,
you
know
part
and
parcel
to.
A
That-
is
the
notion
that
you'll
need
to
spin
up
kubernetes,
spin
up
measuring
spin
up
a
service
mesh
and
then
invoke
the
smi
conformance
test,
and
so
I
think
you
chose
kind
as
a
small
kubernetes
distribution
to
run
inside
of
your
or
inside
of
the
run
runner.
Here,
it's
a
ubuntu
runner
say
more
or
take
us
to
the
rest.
If
you
would
oh
yeah,
let
me
pull
this.
A
F
Okay,
so,
basically
to
perform
the
test,
I
have
to
spin
up.
Basically
the
environment
like
you
would
normally
do
if
you
are
running
this
locally,
so
I
start
with
an
ubuntu-based
runner
and
there's
this
environment
variable
you
can
set.
F
If
you'd
like
to
use
helm
or
messy
ctl
to
install
oh
measuring
on
the
cluster,
then
I
get
the
prerequisites
for
kind,
which
is
the
distribution
I'm
using
since
it's
small
and
you
can
run
it
in
docker,
then
install
cube
ctl,
and
then
this
is
the
conditional
statement
like
if
you're
using
helm
as
the
deployment
mode.
F
This
would
run
so
there's
no
need
to
waste
ci
time
installing
this.
If
the
user
is
not
going
to
get
to
use
it,
then
using
machine
cc
I'll
start,
I
can
install
merchant
on
cluster
and
then
you
know
in
the
future
or
when
this
is
ready,
users
will
be
able
to
pass.
In
your
token,
so
I
had
to
do
some
bash
magic
to
actually
get
this
working
because
the
other
time
there
were
problems
with
the
json
file
not
being
properly
sent
over
into
the
file.
A
Oh
yeah
yeah,
so
so
maybe
I
should
explain
it
real
quick.
This
is
part
of
the
point
of
you
running
through
this
I
did
abhishek.
Do
you
want
to
explain
this.
A
A
Yeah,
so
so
jubril's
working
through
try
to
make
a
github
action
to
be
able
to
run
mescheri
and
kubernetes
and
get
in
get
up
workflows
so
that
we
can
reuse
that
for
testing
and
give
it
to
different
service
mesh
teams
so
that
they
can
run
smp
performance
tests
and
smi
conformance
tests,
and-
and
so
he
I
had
actually
missed-
lost
sight
of
the
fact
that
he's
actually
making
it
he's
already
incorporating
for
reusability.
A
By
having,
if
statements
and
they're
saying
what
he's
creating
this
deployment
mode,
you
know
construct
and
part
of
him
doing
this.
Is
that
he's
using
kind
at
the
moment
and
whether
it
was
kind
or
mini
man?
Excuse
me
mini
cube.
A
The
cube
config
doesn't
necessarily
have
certificate
data,
and
so
he's
you
know
like
much
of
the
rest
of
us,
unfamiliar
with
that
command.
On
line
89,
I
figured
everyone
would
enjoy.
Hearing
someone
else's
voice
other
than
mine,
explain
what
it
means.
E
Do
you
want
me
to
explain
what
it
does?
Yes,
please.
So,
basically,
if
you,
if
you
can
open
the
cube
conflict,
do
you
have
a
sample
quick,
config.
E
E
The
certificate
data
is
the
fields
are
basically
a
reference
to
another
file
in
the
file
system,
which
contains
the
certificate
data
yeah.
So
if
you
can
zoom
out
a
bit,
yeah
yeah,
basically
like
how
you
see
here.
E
The
certificate
data
is
a
string
in
in
this
case,
but
in
case
of
minicube
or
other
clusters,
it
kind
of
is
a
reference
to
another
file
in
the
file
system.
So
when
you
upload
this
file,
meshery
doesn't
have
access
to
the
other
file
like
the
reference
file
in
the
host
file
system.
So
it
fails
to
initialize
the
cluster.
So
what
the
below
command
does
is.
E
Basically,
it
copies
the
content
from
the
reference
of
the
files
and
then
put
it
in
the
put
it
in
this
particular
conflict
so
that,
like
basically
it
like
you,
don't
have
any
reference
so
make
sure
you'll
be
able
to
read
all
the
conflict
properly
and
then
initialize.
It.
A
C
A
Thing
else
yeah,
yeah
jabril,
do
you
mind
sharing
again,
I
can't
wait
for
your
github
action
to
be
done
so
much
that
that
we
have
to
talk
about
it
more.
So
what's
your
current
hold
up
this
issue
here?
Okay,
so
you're
good
through
line
89
right,
so
yeah.
B
A
Just
do
something
else:
let's
do
mastery
ctl
perf,
like
there's,
there's
other
things
to
for
us
to
begin
to
achieve.
As
a
matter
of
fact
like.
Let's
look
at
like
end-to-end
tests
like
hey,
when
we
do
a
release,
does
can
you
install
kuma?
I
don't
know,
I
think
so
I
hope
so.
I
keep
my
fingers
crossed,
but
if
you
had
that
command
in
here,
oh
great,
as
a
matter
of
fact,
we
don't
actually
have
that
command
on
meschery
ctl
command
line.
A
Okay,
maybe
we
should
be
talking
about
designing
out
that
mesh
for
ctl
mesh
command
more.
A
Okay,
it's
it's
linked
in
the
meeting
minutes
and
you
have
a
link
to
the
meeting
minutes
right
yeah.
I
do
okay,
cool
you
don't
have
to
you,
don't
have
to
bring
it
up
now,
but
I
just
want
to
make
sure
after
the
call
you
can
go,
find
it
in
in
the
test
plan.
There
are
a
few
different
tests
around
measuring
operator.
Like
does
the
operator
deploy?
Does
mastery
server
talk
to
it
when
it
does
deploy?
A
Does
it
delete
it
when
it's
told
to
be
deleted,
and
I
think
we've
worked
through
most
of
those
bugs
manually
next
time
we
make
a
code
change
or
you
know
something
else
changes
like
so
my
point
is
like
there's
another,
like
example
of
from
line
89
onward,
where
we're
able
to
like
reuse
line
89,
I'm
sorry
1
through
89
and
then
beyond
that
we
should
be
able
to
start
to
do
some
other
tests,
like
run
other
commands,
do
and
so
there's
a
test
plan,
that's
being
formulated.
A
There
so
juro,
as
you
think
about
this
too,
because
I
know
you
want
to
is
in
measuring
console
that
measure
measuring
hyphen
console
the
measuring
adapter
4
console.
It
has
a
suite
of
tests.
They
are
written
using
this
framework
called
bats
bash,
automated
test
suite
something
like
that,
and
but
it
just
reminds
me
of
gi,
joes
and
bats.
That's
all
it
reminds
me
of
is,
and
it's
good,
mr
one
of
the
maintainers
micro,
michael
gefeller,
wrote
it
or
not.
He
didn't
write
the
framework.
A
I
kind
of
think
you
know
the
more
that
we
dog
food,
our
own
stuff,
the
better-
and
I
don't
remember
who
I
was
talking
to
and
explaining
what
dog
fooding
your
own
thing
is
or
drinking
your
own
champagne,
but
whatever
the
euphemism,
you
want
to
use
the
use
of
measuring
ctl
instead
of
bats.
Well,
that's
nice
and
that's
better.
A
A
So
if
you
would,
as
you
there's
some
other
thoughts
here
as
well
as
like
hey
look,
there's
some
great
projects
that
were
being
built,
that
meshery
itself
is
being
built
around
the
kuma
project
that
the
the
where
they're
doing
some
of
their
own
and
then
tests
linker
d.
They
have
a
great
user
experience
around
linker
d
cli
and
they
have
pre-flight
checks
that
we're
mimicking
and
they
have
other
like
every
single
one
of
these
service
mesh
projects.
A
Some
of
them
do
some
really
excellent,
open
source
engineering
and
ci
pipelining
and
testing
and
things
we
should
go
be
inspired.
I
mean
we
should
think
about
that
if
you
would
think
about
how
we're
using
bats
think
about
how
we
might
do
something
here
reflect
on
a
few
of
the
other
approaches
and,
for
my
part,
I'd
be
really
curious
about
your
perspective
as
to.
A
E
E
It's
kind
of
designed
for
just
github
actions
like
easier
scripting
for
github
actions.
F
F
I
can't
find
it
right
now,
but
basically
it
would
allow
me
to
run
commands
using
other
languages
as
opposed
to
using
bash.
So
it's
it
reminds
me
of
just
that's
why
I
brought
this.