►
From YouTube: Meshery CI Meeting (Feb 11th, 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).
B
B
A
Like
run
a
podcast
right,
some
friends
and
I
have
a
podcast
related
to
video
games
and
technology.
B
Oh
no,
oh
okay,
yeah
yeah,
hence
the
yeah,
hence
the
the
get
up
I
should
have
asked
before
I
joined.
I
wanted
to
make
sure
that
you've
got
the
to
the
extent
that
I
ever
missed
the
meeting
that
you've
got
a
host
key
in
case
you
in
case
the
recording
doesn't
automatically
start
or
in
case
you
can't
share
for
some
reason.
Yeah.
B
C
A
B
B
B
We've
got
measure
ui
and
it's
made
of
react,
etc,
and
we
have
this
softer
concept
of
ux,
and
so
I
think
what
I
ended
up
doing
was,
as
we
yeah
yeah.
I
did
so
so
the
area
area
is
sort
of
like
hey
this
gent
this
general
space,
and
in
this
case
it's
like
you
generally
there's
a
user
experience,
concern
and
then
specifically,
there's
a
instead
there's
a
javascript
concern
under
component
ui.
So
so
I
just
created
the
separated
out
those
two
different
labels
and
then
independent
bot.
B
They
have
reason
you
know
not
too
long
ago,
moved
from
managing
the
configuration
of
dependent
mod
as
a
separate
thing
and
now
you're
able
to
define
that
configuration
in
yaml
in
your.
You
know
github.
So
as
we
were
basically,
I
think
all
of
the
repos.
Now
they
all
had
dependabod,
enabled
and
they've
all
been
updated.
Now,
to
keep
that
dependable
configuration
in
yaml
and
the
repo,
and
some
of
that
configuration
is
to
auto
label
things
so.
A
So
sorry,
so
it's
needed
another
issue
to
address
that
to
modify
the
gml
configuration.
B
Yeah
at
this
point
all
right-
I
guess
I
can't
quite
remember
if,
but
at
this
point
the
config
for
de
bendabot
is
now
done
inside
of
the
dot
github
folder,
and
so
my
comment
here
was.
I
think
it
was
just
in
recognition
that,
like
hey
rodolfo
is,
you
know,
is
leading
this
area
and
if
we're
doing
changes,
we
should
make
darn
sure
that
rodolfo
understands
what's
going
on
and
oh
I
see.
D
B
Yeah
and
the
other,
the
other
thing
I
was
trying
to
do
here
was
like
hey,
wait,
a
second
there's
some
things
independent
about
that
we
should
update
like
who
is
the
default
reviewer
for
a
given
component
like
in
this
case,
I
had
assigned
dhruv
as
a
person
to
be
assigned
as
a
default
reviewer
for
ui
stuff,
and
then
I
was
thinking
wait
a
second.
I
don't
know
that
rodolfo
or
others
can
get
in
here
to
see
this
config
and
even
change
the
config
as
over
time
as
needed,
and
so
that
was.
A
B
Yeah
now
one
of
the
frustrating
things
I
think
this
is
a
github
limitation
is
that
I
don't
know
that
or
maybe
it's
a
depend
upon
limitation.
We
should
probably
experiment
I
at
the
time
you
could
only
explicitly
identify
an
individual
to
be
auto
assigned
as
a
reviewer
and
when,
ideally,
we
would
probably
use
a
team
and
there
and
there
are
a
number
of
teams
in
our
org.
B
Well,
okay,
I
I
did.
E
B
B
Okay
but
yeah
I'll
add
some
notes
into
the
meeting
minutes
to
say
it
might
be
worth
a
test
of
depend-a-pot
see
if
it
can
do
it.
Okay
sounds
good.
A
A
A
I
don't
know
if
you
are
here
jesus,
I
know
I
I
don't
think
so.
The
other
tickets,
the
other
ticket
or
the
other
issue,
is
being
worked
by
max
g3,
but
I
think
he
is
not
here
also.
B
B
Call
out
the
shawl
and
ask
if,
if
working
with
cyprus
is
of
interest
to
you
or.
B
B
We
shall
shall
raise
this
hand.
Nice
shall,
let's
do.
A
F
A
A
A
A
B
Yup
yeah,
that's
right!
That's
right
and
tunisia
is
on
and
so
vishal.
By
the
way
you
meant
you,
you
missed
your
chance
to
ask
about
cyprus,
ui
vishal.
If
I
were
you,
if
you're
gonna
take
on
that
issue,
I'd
take
up
rudolfo
on
his
offer
to
engage
and
kind
of
show
you
around
how
cyprus
works.
Nice,
okay,
yeah
yeah!
I
see
you,
you
do
have
that
question:
okay,
good,
we'll
rudolph,
we'll
have
to
circle
back,
that's
great
and
so
yeah
tenush.
D
Yeah,
yes
yeah,
so
what
the
issue
with
this
I
am
facing
is
that
when
we
how
we
normally
write
a
workflow
for
an
we
specify
the
pull
request
at
the
top
when
we
do,
particularly
if
we
do
this
here
what's
happening,
is
that
the
two
keys
that
natalify
need,
particularly
the
site
id
and
the
auth
id
they
are
not
getting
passed,
because
I
read
it
in
the
docs
that
pull
request
doesn't
allow
those
secret
keys
to
be
passed
on.
D
So
after
some
exploring
a
bit,
I
found
the
apart
from
the
pull
required.
I
found
something
called
pull
request
target
it
does
allow
it
to
pass,
but
the
issue
with
pull
request
target
I'm
facing
is
that
pull
request
target
builds
the
base
of
the
pier.
Not
the
merge
merge
commit,
which
is
what
we
need.
D
So
what's
essentially
now
happening
is
that
whenever
someone
sends
a
pull
request,
we
basically
don't
see
any
changes
in
the
url
that
is
generated
because
it
is
not
the
latest
commit
which
is
being
built.
G
D
Yeah,
so
so
what
essentially
this
tr
does?
Is
it
generates
a
netify
url?
As
you
can
see
at
the
top,
you
can
ignore
the
content
of
the
pull
request.
That
is
not
important.
Here
means
what
this
url
should
generate
is
the
latest
changes,
but
what
it
is
doing
is
that
it
is
building
the
base
of
the
pull
request,
so
none
of
the
changes
get
reflected
and
the
base
of
the
the
merge
command
will
get
built
if
use
pull
request
instead
of
pull
request
target.
B
B
Which,
which
keys,
are
we
looking
for
netlife
keys.
D
B
B
G
D
Yeah,
let
me
clarify
what
we
are
trying
to
do:
what
in
netly
5?
Normally
we
just
specify
the
branch
and
all
the
building
stuff
is
done
by
netlify.
D
So
what
we
were
thinking
was
that,
instead
of
wasting
our
building
buildings
there,
which
takes
about
seven
to
eight
minutes,
we
were
trying
to
build
the
site
in
github
actions
and
the
generated
public
directory,
which
that
is
essentially,
we
will
go.
We
are
going
to
push
it
into
netlify
and
so
nettle.
If
I
build
minutes
are
not.
D
D
D
D
A
Okay,
yeah
thanks
for
pointing
that
that
I
I
think
I
I
will
check
it
out
how
to
to
allow
pull
requests
to
access
those.
Those
variables
have
to
figure
out
something
related
to
that.
D
For
the
the
pull
request
target
was
the
other
thing
I
tried,
but
I
was
not
able
to
get
the
latest
changes
there.
Another
thing
I
thought
of
was
using
a
running
and
the
workflow
using
workflow
run,
and
there
is
another
tag
instead
of
pull
request
target,
but
I
haven't
explored
that
particularly
currently.
A
Okay
yeah,
so
let
me
do
a
research
about
this
in
order
to
to
know
more
about
this
problem,
and
if
we
can
stop
this
one,
I
I
will
send
you
a
slack
message.
Thank
you
daniels.
In
order
to
to
let
you
know
well,
I
am
discovering
sure
yeah
thanks
for
pointing
the
that.
A
B
So
vishal's
audio
doesn't
work
or
he's
not
in
a
place
to
be
able
to
do
it,
but
he
so
I
gave
him
the
link
to
your
blog
post,
which
has
your
slides
in
it
rodolfo.
B
If
you
think
about
it,
if
you
have
time
to
do
it,
a
link
even
to
your
youtube
to
the
youtube
recording
might
be
helpful
or
actually
wouldn't
might
be
helpful
like
for
sure
that
recording
that
overview
that
you
gave
oh
there,
it
is
okay,
now
yeah
yeah,
that's
what
you
need
michelle
right.
There.
A
Yeah
here
is
a
meeting
that
will
explain
how
cyprus
works
and
we
gave
an
overview
about
how
it
works
and
what
is
needed
to
start
working
on
writing
tests
in
cyprus.
So
I
recommend
to
you
that
and
watch
this
video
and
also
read
this
blog
post
and
see
the
other
resources
in
order
to
get
familiar
with
cyprus.
A
A
B
Is
good,
so
that
was
just
a
quick
note.
Alonso
had
mentioned
in
chat
that
he's
stepped
off.
The
call
to
go
have
a
call
with
sriti
chandra
about
multi-language
support
for
mesh
redox.
I
don't
know
that
it
entirely
relates
to
ci,
maybe
a
little
bit
but
but
he'd
be
given
status,
so
I
was
just
tracking
it.
There.
F
I
am,
I
am
working
on
with
with
also
blowing
up
this
issue,
so
I
think
I
can
help
with
that
status,
and
I
I
think
I
I
first
started
working
on
this
trying
to
adapt
the
language
pluring,
but
that
that
doesn't
went
well
with
with
how
how
we
can
insert
that
plugin
into
our
or
objective
space.
F
So
what
alonso
was
trying
to
do
is
to
create
an
isolated
project
for
for
jackie
in
order
to
understand
understand
how
to
use
that
in
into
an
ayaki
project,
and
he
is
also
having
some
issues.
So
I
am
not
sure
if,
if
that
plugin
is
going
to
be
like
the
best
way
to
to
do
that,
but
we
we
need
to
to
check
what
is
the
last
status
from
from
our
launch
to
know.
F
If
that
is
going
to
to
be
the
best
approach
to
follow,
if
not,
we
can
start,
we
can
try
to
do
something
from
scratch
we
just
that
is.
That
is
what
we
we
already
done.
We
added
that
from
scratch.
So
maybe
we
can
start
looking
on
how
to
implement
that
in
into
a
more
automated
way
to
create
translations.
G
Can
we
like
have
a
github
action?
Whenever
we
create
a
publication,
it's
get,
it
gets
merged.
It
will
automatically
take
a
look
at
all
the
changes
that
were
made
in
the
documentation
in
any
one
particular
language,
and
it
changes
that
into
for
the
corresponding
other
languages,
like
whatever
it
seems
in
the
documentation,
with
the
help
of
the
github.
B
B
I'll
rephrase
it
if
I
can
I'm
shivaya
correct
me
if
I
say
it
inc
wrong,
but
I
think
shabbat
is
wondering
if,
if
there's
a
language
translation,
github
action
like
if
there's
if
there's
a
way
of
anytime,
that
update
is
to.
F
G
F
I
well,
I
am
not
sure
if,
if
we
we
want
an
automate
translation,
because
I
am,
for
example,
in
spanish,
that
is
like
the
most
effort
we
put
on
on
this
issue
when
we
did
that
by
an
automation,
an
automation
tool
for
transducer
translation.
It's
like
not
not
doing,
not
making
a
lot
of
sense.
What
that
tool
is,
is
doing
with
the
translation.
F
B
B
It
is
just
a
it's
a
restructure,
it's
kind
of
like
what
you
do
when
you
want
to
keep
versions
of
multiple
versions
of
the
docs,
so
one
it's
the
same
thing
that
you
do
to
keep
multiple
multiple
languages
of
the
docs
and
the
two
will
go
hand
in
hand
by
the
way,
so
so
yeah
so
the
docs
themselves,
it's
a
folder
full
of
files.
B
Those
are
the
docs.
So
then!
Well,
if
you
want
to
have
versions
of
them,
there's
a
few
different
ways.
You
can
approach
this.
B
You
can
at
a
point
in
time,
say
we're
moving
from
version
two
to
version
three
make
a
copy
of
the
folder
leave
the
old,
the
old
one
behind
put
it
under
the
folder
name,
v3
there,
it's
versioned!
It's
done
it's
it's
that
damn
simple
I've
been
talking
about
it
for
ever
I've
been
talking
about
it
for
a
year.
It
is
that
damn
simple,
it's
just
about
the
same
thing
for
language
as
well.
B
The
only
the
only
like
simplicity,
sort
of
snafu
stuff
is
well
what,
if,
like
the
navigation
to
the
header
or
the
footer
or
the
navigator
changes
like
well,
that's
okay,
like
if
you
go
back
to
v3
of
the
docs
you'll,
see
the
the
v3
version
of
the
header
and
the
footer
and
the
navigator,
which
is
entirely
appropriate
because
if
you
change
the
navigator
for
the
new
stuff
for
a
new
link,
a
new
page
you've
created.
B
Point
is
the
docs
do
need
to
be
restructured
to
include
whether
you
do
it
from
scratch,
jesus,
which
is
almost
always
the
not
just
a
lot
of
times,
not
the
thing
to
do,
but
it
is
you're
going
to
have
to
the
docs,
have
to
be
restructured
anyway
to
organize
in
such
a
way
that
you're
able
to
say
either
create
custom
front
matter
variables,
which
is
what
I've
done
a
little
bit
ago.
B
G
Because
even
for
large
projects
like
tensorflow,
they
have
separate
folders
for
the
translations.
They're
not
doing
it
manually
like
they're,
not
doing
in
an
automated
way,
with
some
kind
of
a
ci
workflow.
They
are
having
individual
contributors
for
each
language.
B
Like
in
terms
of
like
the
automated
translate
like
yeah,
whether
it's
automated
or
not,
you'll
have
to
have
a
folder
structure
and
a
like
a
into
jesus
point
like
yeah,
having
tried
to
help
translate
the
english
to
spanish
or
just
trying
to
help
be
a
voice
there,
like
yeah
the
translators
weren't
quite
getting
it
like.
Jesus
was
right
like
it
was
pretty
it's
it's
we're.
B
I'm
thankful
that
alonso
and
jesus
and
a
few
other
contributors
are
here
to
to
help
with
that,
because
it
really
does
take
their
understanding
of
the
technology
to
get
it
to
translate.
To
the
right,
I
mean
they
have
to
actually
do
the
translation,
because
it
I
mean,
but
but
to
what
you'd
said,
shabbat
there's
a
lot
of
merit
to
it
as
well,
because
if
you
do
go
to
a
lot
of
different
sites
and
you
use
like
chrome's
language,
you
know,
auto
translation,
like
is
fairly
close.
G
Yeah
and
it's
always
increasing,
like
the
implementation
of
the
transition
itself,
is
improving
day
by
day,
so
right.
B
And
so
yeah
there
is
a
there's,
a
an
open
issue
on
this
there's
a
jekyll
there's
like
two
there's
like
a
few
jekyll
plug-ins
to
sort
of
structure
your
site,
such
your
theme
such
that
you're
set
up
for
multi-language
support,
and
so
there's
a
one
of
them.
That's
like
the
most,
the
most
popular
of
them.
B
F
Is
I
was
thinking
about?
Maybe
what
we
can
we
can
do
is
each
time
a
new.
A
new
post
is
is
in
a
posted
into
jackie
onto
measuring
documentation.
We
can
create
on
that
that
that
new
document
is
in
master
and
push
it
to
the
website.
F
B
That's
all
yeah
that
makes
total
sense
I'll,
we'll
write
that
down
like
yeah.
If
we're
going
to
actually
seriously
have
a
translated
version
of
the
docs,
how
could
you
guys
ever
be
expected
to
keep
up
if
you're
not
being
notified
like
when
a
new?
You
know
when
there's
something
new
to
translate
so
yeah.
F
B
Totally
I
was
hoping
for
a
tamil
version
soon
shiva
either
that
or
hindi
transition.
A
B
A
B
A
Okay,
great,
so
the
next
topic
is
a
get
nighthawk
a
ci
setup.
I
think
abhishek
has
already
done
or
has
already
executed
some
tests
about
this
avishay.
Can
you
tell
us
please
what
do
you
do
and
what
do
you
think
can
be
the
next
steps?
A
H
So
I
was
able
to
build
the
nighthawk
server,
both
both
the
client
and
server,
and
and
also
the
test
server
last
a
couple
of
days
ago,
and
it
was
running
fine.
So
I
was
thinking
the
next
steps
would
be
to
start
setting
up
the
ci
and
start
publishing
the
artifacts
for
different
architectures
or
os
versions
of
of
these
binaries,
so
that
that's
what
the
so.
What
I
was
having
in
my
mind
is
that
we
could
write
different
github
actions,
which
would
do
we
should
do
the
build
and
release
flow.
H
The
idea
is
that
we
would
have
one
github
action
that
would
that
would
basically
build
the
build.
The
binary
for
one
particular
version
of
nighthawk
and
the
these
binaries
would
be
multiple
of
them,
basically,
each
of
each
for
each
for
a
separate
operating
system,
architecture
and,
and
then
all
of
the
all
of
them
are
released
and
the
same
as
a
part
of
the
same
getup
release.
B
So
yeah,
for
my
part,
it
is
go,
create
some
issues,
get
really
specific.
That's
that's
what
what
package
manager
are
we
going
to
shoot
for
first,
which
which
are
the
the
ones
that
we
should
shoot
for,
based
on
what
we
think
we
can
easily
compile
to
to
outline
where
those
where
the
binaries
will
be
published
and
what
how
many?
If,
if
there's
more
than
one,
get
nighthawk
container
image
to
be
pushed?
B
Let's,
let's
see,
let's
break
down
the
tasks,
let's
get
specific,
let's
give
them
out
to
folks.
Rudolfo
is
here
highly
capable
and,
and
I'm
going
to
pressure
him
to
help
on
the
project.
B
So
yeah
I
mean,
can
you
go?
Please
go
create
issues.
There
are
tasks
that
are
listed
in
the
spec
itself.
The
project
spec
that's
linked
here
at
the
bottom
of
that
dock.
Did
you
have
you
seen
those
tasks
yeah
sure
I
I've
seen
them
I'll
go
create
issues?
B
Thank
you.
Thank
you.
I
think
it'll
unlock
it
for
people
like
it'll
it'll
help.
Others
move
some
of
it
forward
in
parallel.
Do
you
know
if
there's
a
if
there's
a
standard
runner
that
we
should
be
using
hey?
Look?
You
can
only
build
this
thing
in
ubuntu
anyway.
Here's
the
basil
action
or
whatever
those
those
standard
components
to
it
are.
B
I
think
that
that
will
enable
others
to
be
successful
and
to
work
on
the
rpm,
while
someone's
working
on
the
tar,
gz,
someone's
working
on
the
x64
like
and
for
to
just
go,
lay
out
what
those
are
like.
We
don't
have
to
build
all
of
the
things
we
just
need.
A
handful
of
popular
platforms
to
support.
I
B
The
yep,
the
so
in
here
the
project
specification
hyperlink
yep
that
that's
the
one
so
yep
yeah
the
in
scope
out
of
scope,
section
it's
like
on
page
two-ish
or
three-ish;
okay,
fine,
it's
fine!
It's
on
page
five!
B
Specifies
some
it
says,
stable
and
nightly,
builds
for
debbie
and
ubuntu
centers
and
mac
like
hey.
Look,
let's
get
as
far
as
we
can
for
the
from
a
selfish
meshery
perspective,
shoot.
We
just
need
one
that
works
inside,
that
we
can
run
inside
of
mesherie's
container
image
and
then
one
and
then
and
kind
of
focusing
and
getting
that
one
right
means
that
there's
a
pattern
to
follow
for
the
others
yeah.
B
But
I
think
for
a
lot
of
there's
a
number
of
very
well-intentioned
people
in
the
community
who
have
all
kinds
of
other
things
happening
in
their
lives
and
are
also
all
like,
like
the
rest
of
us
all
of
us
learning
as
we
go
so
yeah
this
yeah.
So
if
you're
able
to
break
down
where
you've
had
success,
no
doubt
people
will
pick
those
up
so
sure
I'll.
B
Question
abhishek:
do
you
know
the
os
that's
being
used
as
the
base
image
for
measuring?
No
doubt,
that's
alpine,
I
don't
know
if
you.
H
I
B
Yeah
and
part
of
what
abhishek
you
or
or
or
anyone
else
like
I
just
I'm
calling
on
abhishek
specifically
because
because
he
lets
me
bend
his
arm
is
if
people
have
there
are
things
to
just
identify:
hey
the
the
the
static
binaries.
They
will
be
published
to
the
releases
section
of
that
github
repo
or
they
will
be
published
here
or
there.
Whatever.
Those
things
are.
B
That
document
is
just
waiting
for
opinions,
feedback
comments,
suggestions,
we're
gonna
this
next
thursday,
we'll
have
another
cncfc
network
call
nighthawk
is
a
standing
topic
there.
It's
an
awesome
time
for
anyone
who's
here
to
elevate
their
works
for
some
people
surrounding
the
layer,
five
community
to
see
what
you've
done
get
to
know.
You
there's
some
very
seasoned
technologists
that
come
to
that
meeting.
B
B
B
It
is
it's
layer
five
I
o
get
nighthawk,
actually
there's
a
link
right
there.
It's
open,
get
nighthawk
issues
right
inside
the
meeting
minutes.
B
E
Story
come
again
on
what
to
put
a
thought
on
what
you
said
about
before
about
the
image
that
is
alpine
and
not
open
to
and
yeah.
I
was
looking
on
the
one
that
was
I'm
going
to
build
and
I
had
the
imperson
that
we
have
the
mastery
on
one
to
20,
but
I
think
it
is
for
the
sites
or
or
something
that's
I
think
I'm
wrong
or
or
for
the
person
builds,
I
don't
know
or
I'm
confusing
the
projects.
E
So
I
didn't
look,
but
I
tried
last
week
to
just
to
start
creating
to
start
watch
some
issues,
so
I
can
also
select
some
issues
or
something
to
work
on,
not
sure
what
was
your
question?
Sorry
for
yeah?
That's!
That's
kind
of
it.
It's
just
like
yeah
around
the
thing.
B
That
was
mostly
it
like
an
update.
I
created
an
issue
that
I
thought
that
might
be
part
of
what
you
were
trying
to
do
sort
of
just
like
an
initial
build
on
okay.
E
So
I
and
you
also
assuming
why
didn't
I
get
the
email
and
also
I'm
surprised,
I
didn't
get
the
email
for
what
for
the
ones
I
signed
up.
I
I
I
edit
my
I
myself
with
the
water.
I
have
to
see
what's
going
on
on
my
mail,
probably
okay,
so
I'm
I'm
assigned
to
it
that
that's
great,
I'm,
I
will
say
I'm
a
little
bit
behind
on
being
reliable,
how
soon,
except
yesterday
and
asap
you
need
this-
you
you
need
this
to
have
it
not
to
block
the
other
releases
and
so
on.
E
B
E
Okay,
so
we
I
I
we
whatever
should
have,
can
we
have
a
checkpoint
on
saturday
I
saw
you
sometimes
worked
working
on.
B
E
The
word
did
not
came
in
right
in
in
this
right
moment
in
in
my
head,
but
I
would
I
wouldn't
have
a
problem
to
put
labels
on
people.
Sometimes
I'm
doing
this
mistake,
but
it's
a
starting
point
also.
Sorry.
E
Yeah
I'll
put
it,
I
put
it
in
other
words,
for
me,
it
will
be
great
in
order
to
actually
finish
to
have
a
checkpoint
on
on
saturday.
What
did
I
build?
How
I'm
building
my
blog
so
and
and
like
this?
Since
I
will
know
I
have
a
tight
deadline,
I
will,
and
I
have
I
can.
I
can
be
focused
only
on
that.
That's
the
thing.
E
So
I
I
would
like
to
have
only
this
timely
on
saturday
and
on
monday
at
check
box,
like
you
know
the
check
box,
so
I
don't
know
sure
when
you
you
need
to
check
your
kids
if
they
did
their
homework
or
something
I
don't
know
how
to
say
it
but
yeah
and
if
they
didn't
okay,
where
are
you
standing?
How
much
did
you
do
or
what
is
the
thing
and.
B
Yeah,
it
makes
sense,
makes
sense
very
good,
very
good.
B
E
Okay,
so
abhishek
rodolfo
and
for
the
other
one.
E
Yes,
yes,
yes,
and
I
will,
I
don't
know
sync
with
them
when
it
will
be
a
good
time
for
at
least
one
of
them
on
saturday,
and
at
least
one
of
them
on
on
monday,
okay,
to
to
meet.
E
E
B
Now,
just
quick
quick
recap
for
me
on
netlify
did
we
have
explicit
items
that
we
were
going
to
try.
B
And
we
didn't
like
to
enter
shavai's
point,
which
is
like
a
really
great
point
like
when,
when
you
do
a
search
of
the
github
for
github
actions
with
netlify
doing
netlify
builds.
I
think
when
I
did
a
google
search,
I
found
a
million
articles
saying
how
to
move
from
github
actions
to
netlify,
but
that
wasn't
what
I
was
looking
for.
I
was
looking
for
like
a
github
action
that
just
kind
of
had
you
know,
would
land
the
preview
in
your.
B
One
last
thing
for
me
to
read
off
of:
if
I
could,
we
had
gotten
an
email
from
github,
not
a
special
one
or
anything,
just
an
email
from
github
saying
you
know,
as
that
you
have
run
over
your
storage
limit
for
layer,
five
io
for
the
org
and
it's
500
megs
a
month,
and
I
thought
my
goodness.
I
didn't
realize
that
we
were
necessarily.
B
Persisting
many
of
the
build
artifacts
you
know
because
there's
tons
of
builds
that
go
on
across
the
various
repos
and
things,
but
in
general
those
are
transient,
like
you
know
that
when
we
do
benefit
from
previewing
a
build
artifact
like
sometimes
it
would
be
nice
to
have
a
build
artifact
like
mesh
ctl
for
that
pr
for
measuring
ctl
like
sitting
right
there
in
the
pr
where
you
can
just
click
on
it
download
it
you
know,
do
some
user
tests.
B
That
would
be
kind
of
helpful.
We
don't
do
that
that,
so
I
was
a
little
bit
confused
as
to
like
geez.
Where
did
we
run
over
the
500
mega
limit
and
and
anyway,
to
help
overcome
that
yeah?
I
sent
out
an
email
about
with
some
specifics
about
where
we
were
running
over
the
limit,
or
also
do
you
remember
where
we
were
running
over.
B
There's
a
configuration
in
github
on
a
per
repo
basis
and
a
and
on
the
org
level
that
you
can,
I
think,
by
default,
it'll
it'll.
It
has
a
retainer,
a
90-day
retainer
of
artifacts
that
were
binaries
or
other
artifacts
that
were
created
during
build
time.
So
we
reduced
that
down
significantly.
As
a
matter
of
fact,
I
said
I
would
reduce
it
to
15
days
and
like
because
the
bills
were
blocked.
I
just
reduced
it
to
one
day
because
the
bills
are
blocked
and
because
we
don't
need
any
transient,
artifacts
we're
not
using
that
functionality.
B
Yep
I'll,
go
I'll,
come
clean
and
say
I
actually
reduced
it
to
one
day
and
so,
but
which
so
tuneus.
You
know
like,
depending
upon
what
approach
you
end
up
in
and
rudolfo
kind
of,
where
we
land
on
layer.
Five
preview
builds,
if
part
of
that
thinking
and
part
of
that,
workflow
was
persisting
a
temporary
folder
in
the
repo
to
understand
that
that
keep
in
the
back
of
your
mind
that
there
that
that
setting
is
there
and
could
affect
that
person.
I
B
You're
still
kind
of
tracking,
I
think
toward
like
the
fact
that
vishal
raised
his
hand
to
kind
of
dig
into
some
cypress.
You
know
functional
tests.
Oh
awesome,
rudolph.
You
you'd
led
an
initial
charge
to.
B
Do
functional
testing
of
kubernetes,
you
know
spinning
up
a
kubernetes
cluster
laying
out
a
service
mesh
deploying
an
app
is
that
part
of
like
if
vishal
is
successful,
with
creating
a
few
functional
tests
in
cyprus
that
those
tests
will
be
using
that
temporary
kind
cluster
right.
B
Right
right,
yeah
so
anyway,
so
so
I'm
excited
that
that
that
you
all
are
working
on
a
on
such
tests.
So
I'll
repeat
that
really
quickly,
mostly
for
cans.
Welcome
to
the
call
attends
we're
about
to
we're
like
two
seconds
away
from
ending
the
call
so
very
nice
to
have
you
on
the
call
quickly
I'll
say
if
you
don't
have
a
link
to
these
meeting
minutes,
we'll
make
sure
we'll
post
one
in
slack.
B
So
you
can
go
check
them
out,
but
we
were
just
wrapping
up
a
discussion
on
the
fact
that
each
time
that
someone
submits
a
pr
to
the
message,
repo
there's
a
number
of
tests
that
are
run
and
one
of
those
tests
is
to
spin
up
a
kind
cluster
invoke
measures,
operations
to
deploy
a
service
mesh,
deploy,
a
sample,
app
and,
and
you
know,
go
through
a
few
different
tests
and
some
of
the
contributors
here
are
working
on
new
tests.
So
pretty
neat.
A
Great,
so
I
think
we're
getting
over
the
meeting.
So
thanks
for
everyone
to
attend
this
meeting,
and
if
you
have
more
questions,
please
be
feel
free
to
reach
us
in
the
channel
or
by
a
select
message
to
me.