►
From YouTube: Argo Contributor Experience Office Hour 17th June 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
Okay,
so
this
pr
is
pr4694
and
just
for
recording
purposes.
This
is
to
add
automated
ui
smoke
tests,
pr
that
was
contributed
by
keith
chung
and
has
gone
a
little
bit
stale,
but
would
really
help
with
adding
more
tests
to
the
ui
so
just
wanted
to
get
more
attention
to
it.
A
Okay-
and
I
I'm
looking
at-
I
mean
I'm
kind
of-
I
have
to
comment
about
that
pr,
because
I
was
waiting
for
review
and
I
know
that
basically
I
just
read
it
so.
I
gave
comments
to
keith
and
I
remember
what
we
were
discussing.
So
I
asked
him
to
try
to
use
a
page
object
pattern
where
you
kind
of
supposed
to
define
objects.
A
That
represents
your
elements
and
provide
methods
needed
to
simulate
the
sections
and
then
and
then
I
guess
it
took
some
time
and
implemented
it,
and
I
just
didn't
notice,
it
was
done
and
it
yeah.
So
I
will,
you
know,
continually,
make
it
again
and
maybe,
if
you
can,
if
you
in
that
meeting,
yeah
and
yeah
you
here,
you.
C
A
What
so,
basically,
if
I,
if
I
got
to,
try
it
so
now,
there
is
a
same
test
as
it
was
before,
but
now
it
uses,
you
know,
suggested
kind
of
page
method
as
a
page
object,
framework
yeah,
and
do
you
know
what,
like
I
guess,
to
really
kind
of
get
it?
You
know
100
done,
we
will
need
to
add
it
into
ci.
Does
this
pr
includes
existing
ci?
It
changes
or
it
has
to
be
similar
kind
of
manually.
D
A
There
was,
I
know
we
were
considering
kind
of
two
goals.
One
was
to
have
regular,
you
know,
have
a
selenium
test
that
verify
each
and
every
pull
request,
and
we
also
had
to
have
a
smoke
test
that
we
run
against
production
environment
and
it
would,
you
know,
verify,
do
kind
of
read
on
reactions
against
production.
A
Can
you
remind
me
what
the
pull
request
does
it
address?
Is
it
towards
ci
use
case,
or
you
know,
smoke
test
for
production,
environment.
A
D
And
I
think
this
just
serves
as
a
base
for
for
other
tests
that
we
can
add
yeah.
Basically,
it
can
serve
as
a
framework
for
other
new
tests
right
for
smoke
tests
and.
A
A
Okay,
here's
an
example:
basically
it's
a
typescript
class
that
can
be
used
by
any
kind
of
test,
ci
test
or
the
smoke
test,
and
it
kind
of
gives
you
like
an
argo
cd,
specific
dsl
that
and
basically
yeah
here's
a
real
application
create
panel.
So
that
is
not
real
panel
that
renders
ui
it's
a
wrapper.
That
gives
you
methods
to
manipulate
his.
A
You
know
browser
page
that
has
create
panel
open
and
you
can
use
it
to
set
name.
So,
instead
of
writing,
basically
it
all.
It
encapsulates
kind
of
information
about
how
exactly
create
panels
ended
and
so
test
is
not
so
fragile.
If,
tomorrow,
we
factor
create
pedal,
we
will
not
have
to
refactor
all
the
tests
and
instead
we
can
just
change
one
method,
setup
name
and
all
tests
that
use
that
method
won't
be
won't
fail.
You,
okay,
so
yeah
and
I
apologize
I
didn't
know
you
addressed
these
changes
yeah.
A
I
think
I
will
yeah
no
worries.
It's
not
such
you
know.
It
should
not
take
long
to
review
it
again
and
it's
really
safe
to
merge,
because
I
remember
when
we
spoke
all
these
changes
kind
of
very
isolated.
So
it's
basically
it's
new
subfolder
ui
test
and
it's
like
it
has
separate
set
of
dependencies
node.js
dependencies
and
it
won't
affect
in
any
way.
B
So
that
was
the
only
thing
that
I
had
for
ui
discussion.
So
if
we
oh
sorry,
jan
had
actually
made
a
point
really
quick
to
introduce
some
of
the
new
people
that
we
have
on
the
call.
C
Hi,
this
is
peribi,
I'm
basically
in
hyderabad,
india.
I
work
for
fa
networks,
so
I
just
I
mean
I
work
here
for
ingress
controller
of
fib
called
balancer
and
another
project
icon
controller.
So
I
just
found
this
project
to
be
a
bit
interesting,
so
just
want
to
learn
and
contribute
for
this
project,
so
just
found
this
url
in
the
repo.
So
I
just
trying
to
call
last
meeting
also
I
joined.
This
is
my
second
meeting.
E
Yeah
yeah,
but
welcome
again,
so
thank
you
thanks
thanks
for
the
quick
introduction,
I
think
I
see
two
two
new
more
people
on
the
list.
One
is
himanshu.
F
Hey
hi
folks,
I
work
here
in
ui
path
as
a
lead
engineer
and
we
are
working
on
a
product
to
bring
all
the
ui
path
product
suits
on
a
single
kubernetes
system
and
deploy
it
on
to
the
customer
premises.
So
for
that
we
are
using
heavily
on
argo
cd
as
a
deployment
tool
onto
the
customer
premises,
and
for
that
we
I
wanted
personally
to
contribute
some
of
the
areas
that
would
be
benefit
for
us
as
well
as
well
as
to
the
community.
G
Hello,
myself
rajiv,
I
mean
I
am
also
working
with
uipath
and
me
and
himaju
wanted
to
discuss
a
few
of
those
things.
So
that's
why
I
joined
the
call.
A
Awesome
and
great
yeah,
so
if
you
want
to
discuss
something,
then
you
you
know,
we
have
a
document
and
we
could.
You
could
just
add
topic
ideas
into
the
doc.
I
just
put
the
link
into
it
yeah,
and
this
is
how
you
get
you
know,
topics
into
into
that
meeting
and
yeah.
A
Since
I
I
didn't
prepare
for
that
meeting,
but
I
I
know
that
there
was
one
update
about
some
ui
changes
and
I
I
was
thinking
remington
gave
that
update
and
he's
nothing
missing
today.
So
I
can
briefly,
you
know
just
show
what
remington
was
doing
and
then
I
feel
like
it
was
important
because
he's
kind
of
trying
to
he's
pushing
some
changes
to
refresh
some
argus,
ui
style,
dependencies
and
yeah,
but
and
the
pr
to
make
these
changes
was
open
for
quite
long
time.
But
I
can
just
give
an
update
about
it.
A
So
as
you
as
you
might
know,
we
have
our
cd
ui
library,
which
has
a
set
of
components
and
it
had
no,
if,
like
basically,
we
it
didn't,
have
any
commits
for
quite
long
time
and
remington.
He
was
working
on
few
ui
projects
like
rollout
dashboard
and
he
was
trying
to
basically
get
the
same
functionality
that
we
have
in
the
cheaper
but
use
more.
A
You
know
modern
techniques
like
react
web
hooks
and
it's
not
easy
to
migrate
to
this
new
way
of
building
components
because
argo
city,
already
you
know
heavily
using
that
the
triposity,
and
so
we
just
had
a
discussion
with
him
and
we
agreed
to
kind
of
start
v2
version
in
that
repository,
and
you
can
see
here
there
is,
you
know
now
v2
folder
and
it
has
bunch
of
components
and
separate
storybook,
and
you
know
description
and
so
that
what
we're
trying
to
do
we're
trying
to
now
find
a
way
to
use
in
argus
gui,
v1
and
v2
simultaneously
and
without
you
know,
resulting
increasing
the
bundle
size
too
much
and
it's
possible.
A
So
basically,
we
won't
suddenly
get
twice
more
javascript.
It
would
be
pretty
much
the
same
size
and
I
guess
remington
will
try
to
you
know,
keep
replacing
v1
components
with
v2
and
but
you
know
don't
we
agree
not
to
do
it.
You
know
all
at
once
and
instead
just
do
it.
You
know
one
by
one
and
basically
a
good
example
is
now
we
have
an
open
back
on
application
list
page
if
I'm
not
mistaken.
A
Basically,
instead
of
just
going
and
fixing
back
into
in
mobile
components,
remington
will
try
to
replace
a
component
that
has
a
bug
with
v2
version
and
kind
of
fix
the
bug
and
at
the
same
time
do
that
refresh
he
wants
to,
and
this
is
basically
here
for,
like
kind
of
just
for
your
information
for
everyone
who
is
interested
in
ui
work
yeah.
This
is
happening
right
now
and
I
guess,
to
kind
of
to
step
back.
A
The
main
goal
is
to
kind
of
I
would
say
my
goal
is
to
just
use
more
react,
clip
hooks
because
it's
it
really
helps
with
state
management
and
in
react
and
makes
it
easier
to
use
to
edit
code-
and
I
kind
of
I
agree
with
it.
So
we
can
get
rid
of
some
self-written
components
like
data
loader
and
ocd
and
in
in
our
ui
project.
It's
kind
of
complex-
and
you
know
webhook-
is
a
better
alternative
to
it.
A
A
So
let
me
switch
back
to
the
working
document.
So
am
I
right
that
we
some
topic
was
added?
Okay,
he
wants
to
edit
the
topic.
We
can
do
that
right
now.
We
can.
You
know,
since
we
have
a
lot
of
time
today,
not
so
many
topics
I
propose
we
just
you
know
talk
about
what
you
know.
What
you
want
to
talk
about.
D
F
Okay,
so
I
happen
to
come
across
this
particular
issue
on
argos
board,
and
this
is
personally
that
I'm
pretty
much
interested
in
so
I
was
we
were
thinking
if
we
can
have.
I
was
thinking
to
contribute
in
this
area
where
we
can
utilize
values.schema.json
file.
If
it's
present
in
the
chart,
we
can
leverage
that
to
render
a
complex
form
on
the
parameters
tab
of
an
application.
So
those
are
something
that
I
was
thinking
if
we
can
have
it
and
maybe
yeah
yeah.
A
F
Okay,
so
value
schema,
json
is
actually
defining
the
metadata
for
or
the
schema
for,
the
corresponding
values.
The
amazon
file
right,
which
is
there
in
the
help,
and
currently
we
are
actually
only
rendering
all
the
values
as
a
string
and
which
doesn't
have
any
type
defined
other
than
the
string
and
the
passwords
are
shown
as
a
plain
text
and
there
is
no
form
of
a
validation
provided.
F
F
So
if
we
can
have
a
type
associated
with
each
and
every
parameter
field,
we
can
maybe,
for
example,
I
can
have
a
type
associated
with
one
of
the
field
as
a
boolean
and
then
maybe
we
can
render
it
as
a
checkbox.
And
if
it
is
an
array,
maybe
we
can
render
it
as
a
drop
down,
maybe
or
enum
as
a
drop
down.
So
those
are
the
capabilities
that
we
can
utilize
and
youtube
and.
F
It's
it
is,
it
is
right
now
I've
only
seen
it
with
the
helm,
application
parameter
because
value
schema
json
is
a
helm
specific.
F
That's
why
I
was
thinking
only
for
helm
right
now,
but
we
can
still
leverage
it
for
the
other
tools
as
well
like
yeah
yeah
so,
but
for
now
I
was
thinking
it
too
for
using
it
for
help,
but
then
we
can
leverage
it
for
the
other
areas.
A
F
As
even
mentioned
in
this
issue
that
there
are
the
many
libraries
that
we
can
use
it
right
now
for
render
it
so
there
is
a
pros
and
cons
with
using
a
library.
F
The
pros
is,
we
can
have
an
exhaustive
implementation
means
we
can
have
a
vast
implementation
and
we
can
support
lot
of
the
functionality
which
is
json.
Schema
means
what
a
lot
of
the
fields
that
we
can
add
in
our
json
schema
those
whatever
the
library
supports.
So
we
can
have
many
many
functionalities
that
we
want,
so
that's
the
biggest
advantage
of
using
library.
However,
the
disadvantage
that
I
could
see
over
here
with
the
library
is
each
library
has
its
own
property
to
render
few
of
the
fields.
F
So,
for
example,
if
if
I
take
the
library
that
is
been
available
with
the
react
application
that
library,
instead
of
apart
from
using
value,
schema
json,
it
uses
one
more
additional
json
file,
that's
called
ui
schema,
json
and
user
has
to
provide
all
the
type
of
fields
that
they
wanted
to
render.
So,
for
example,
even
though,
if
they
wanted
to
override
the
behavior
of
a
boolean
from
checkbox
to
a
radio
button,
maybe
so
they
have
to
provide
all
those
additional
functionality
into
ui
json.
F
A
Can
I
ask
you
like
I
never
I
didn't
know
that
helm
has
you
know,
helm
chart
might
have
adjacent
humor
and
the
question
is,
do
you
know,
is
it?
Is
it
like
a
pattern?
People
actually
address
it,
yeah.
F
There
is
an
option
in
an
argo
where
you
can
set
validate
the
schema,
so
it
will
use
this
file
will
be
used
to
validate
the
parameters
provided
by
user
and
that's
the
feature
is
already
there
in
the
argo.
H
Not
necessarily
the
helm
chart
parameters,
meaning
you
can
still
make
that
mistake
that
you
mentioned
like
you,
can
put
a
name
instead
of
a
boolean
and
it
breaks
the
helm
like
it.
Then
then
you,
the
helm,
template
command
breaks,
yeah
right,
but
and
if
we
were,
if
we
implement
this,
the
validation
would
happen
before
the
parameters
are
accepted
in
the
ui
because
well
we
would
probably
give
them
a
a
radio
box
instead
of
a
text
box
for
the
boolean.
I
Yeah,
but,
and
just
technically
after
this
conversation,
so
in
open
shift
console
today,
we
have
a
way
to
install
hand
chats
outside
the
context
of
argo
cd,
and
we
use
the
same
mechanism
where
we
generate
forms
on
the
fly
based
on
the
content
of
the
schema
file,
and
we
encourage
our
partners
to
put
the
schema
file
so
that
we
can
generate
a
nice
form
else.
It's
a
basic
form.
A
Okay,
that's
that's
to
me
a
big
big
learning.
I
know
that
we
have
a
pull
request
in
progress
right
now
that
is
trying
to
improve
config
management,
plug-in
support
and
ui
was
the
next
step
and
we
were
hoping
to
find
some
some
standard
way
to
render,
and
you
know
define
parameters,
and
this
seems
to
be
this-
this
favorite,
it's
like
if
helm
uses
it
already,
then
you
know,
and
we
can
kind
of
try
that
mechanism
on
heaven
first
and
then
we
can
expand
it
to
plugins
as
well,
and
this
way
plugins
will
get.
A
H
Yeah
yeah,
no,
this
feature
sounds
like
a
good
feature
to
do.
I'm
wondering
to
to
surf
the
json
schema
is
the
the
schema
would
be
served
to
the
ui
and
the
ui
would
then
understand
it
and
then
present
this
different
components,
depending
on
the
type
of
the
field.
Is
that
how
this
would
work.
F
F
A
F
Okay
next
thing
I
wanted
to
understand
if
I
wanted
to
continue
on
community
contributing
on
this
area,
so
yeah
yeah,
so
so
what's
the
process,
should
I
raise
a
proposal
on
this
and
then
how
will
you
go
forward?
So
I'm
completely
new
to
this.
A
Yes
idea
that
I
feel,
like
basically
small
enhancements,
not
necessarily
a
puzzle,
but
to
me
this
is
not
small.
I
would
you
know
just
to
take
your
time.
You
can,
I
would
write,
you
know
a
short
document
that
explains
how
you
want
to
do
it
and
then
don't
spend
like
would
not
spend
too
much
time
on
it,
but
at
least
it
will
give
a
chance
for
people
to
suggest
like
I
would
personally,
I
would
write
a
suggestion
to
try
to
move
that
component
and
make
it
not.
A
Yeah,
so
if
you
have
a
doc
who
will
have
a
place
to
add
some
details,
don't
forget
about
these
details
and
then
next
step
just
implement
it
and
yeah.
So
we
kind
of
the
idea
here
that
we're
trying
to
we're
in
the
process
of
defining
a
process
right
now.
You
know
an
idea
that
we're
trying
to
plan,
for
you
know
what
we
work
on
in
release
and
I
think
that
feature
will
go
into
next
release
because
we
already
kind
of
halfway
through
current
release,
and
there
is
just
we
would
have
to.
I
Right
and
during
this
release
time,
you
might
want
to
have
the
enhancement
proposal
in
place.
Well
reviewed
the
design
and
everything,
and
then
it
could
could
be
something
that
we
effectively
take
out
time
to
review
in
this
release
and
then
in
the
next
release.
The
change
can
actually
come
in.
F
Awesome
then,
then
I'll
I'll
work
on
the
idea.
Firstly,.
H
So
the
the
enhancement
proposal
would
describe,
for
example,
how
it
would
actually,
how
do
you
think
it
should
work
like,
for
example,
I
think
this
will
probably
need
repo
server
changes
that
unpack
the
json
schema
from
the
target
of
the
home
chart,
and
then
you
know
it
might
need
to
cache
it,
because
you
know
it
doesn't
want
to
do
that.
Every
time
you
you
go
to
the
parameters
page
of
the
help
chart,
and
then
we
might
need
a
new
endpoint
that
serves
the
json
schema
for
a
chart
or
or.
B
A
I
hope
it
will
be
not
so
difficult,
but
actually
I
I
feel
like
we
will
be
able
to
reuse
existing
api
to
get.
We
have
a
way
to
get
application
details
that
returns
kind
of
ui
specific
data,
but
it
will
need
changes.
Basically,
we
will
need
to
add
one
more
field
yeah
into
that
api
response
and
make
repo
server
code
changes.
F
Sure,
if
you
guys
give
me
another
couple
of
minutes,
I
wanted
to
have
again
a
short
conversation
on
this,
because
I
did
some
of
the
poc
on
this
and
I
we
can
leverage
some
of
the
additional
external
libraries
which
we
can.
If
you
want,
we
can
include
it
in
argo,
which
can
which
actually
renders
the
complex
form
based
on
the
schema
json.
F
So
there
is
the
option,
one,
the
and
the
option
two
is
we
develop.
We
take
the
json
and
we
develop
everything
in-house
so,
which
means
argo
ui,
plug
the
components
that
we
have
right
now.
We
can
use
those
to
render
the
complex
form.
So
those
are
the
two
things
I
am
I'm
not
too
sure
which
way
to
go,
because
both
of
them
has
their
own
pros
and
cons.
F
So
with
library,
it's
an
additional
library
and
the
problem
that
I
could
see
with
the
library
is
the
look
and
feel
the
how
it
has
been
rendered
and
how
the
look
and
feel
of
the
argo
is
currently
right
now,
whereas
we
can
control
the
look
and
feel
if
we
based
on
the
type
and
the
actual
schema.
If
we
develop
it
in-house,
we
can
control
everything
accordingly,
all
the
location,
so
that's,
but
library
provides
a
biggest
advantage
over
here.
F
Is
we
don't
have
to
invest
too
much
on
the
functionality
that
supports
so,
for
example,
if
a
field
is
some
thing,
we
wanted
to
make
it
a
mandatory
field
where
user
has
to
provide
the
value?
For
that,
then
all
those
functionalities
are
already
present.
What
I
have
observed
in
the
external
libraries-
and
we
don't
have
to
invest
too
much
so
basically
in
a
short
time,
we
can
have
a
huge
functionality
means
we
can
have
a
lot
of
functionality
already
implemented,
even
with
the
validation.
F
Some
of
the
basic
validation
is
also
been
present
in
those
libraries,
and
if
we
go
and
build
it
in
house,
the
problem
will
be
will
not
able
to
cover
as
much
as
functionality
in
one
core.
We
have
to
take
it
incrementally.
A
A
A
A
F
A
A
Six
four
all
right
say
it
again:
please
so
is
it
six
four
zero
one
one,
okay.
C
I
tried
to
pre-produce
the
issue
and
I
was
able
to
report
this
with
both
the
version,
the
one
he's
using
2.0.0
and
the
latest
one
2.0.2,
and
so
what
I
did
could
just
go
a
bit
up
down
all
that.
So
we
have
that
ignore
field
right
now
upon
setting
that
one
upon
setting
is
no
differences.
C
The
old
version
still
had
the
issue,
but
not
the
latest
one.
So
either
this
issue
is
legitimately
fixed
or,
while
kicking
some
other
issue.
This
issue
might
have
gone,
but
what
I
want
to
say,
the
latest
version
is
working
properly.
A
We
I
kind
of
I
I
I'm
aware
of
the
problem
like
just
for
everyone
else.
It's
a
there
was
a
change
in
kubernetes
and
default
value
of
that
field
became
for
true
right.
It
used
to
be
false.
Now
it's
true,
and
basically
this
is
how
kubernetes
controller
works.
They
drop
default
fields,
they
they
drop
fields
from
you
know
from
the
resource.
If
value
is
already
equal
to
default
value
yeah,
and
that's
why
rcd
sometimes
might
notice
the
difference,
it
might
say:
oh
you
don't
have
a
your
refrigerator.
C
Yeah,
the
point
is
the
kubernetes.
It
will
not
accept
the
parameters
so
no
arbitrary
index.
The
parameter
kubernetes
continuously
knows
that.
So
we
see
that.
But
if
we
ask
adversity
to
ignore
that
field,
it
does
so.
You
can
scroll
a
bit
down.
C
A
C
Okay,
so
is
it
about
like
maintaining
this
kind
of
information
somewhere
or
is
it
about
letting
it
dynamically.
A
I'm
sorry,
okay,
can
you
say
it
again?
Please.
C
A
It's
in
the
I
mean,
basically,
we
would
have
to
make
code
changes
to
to
do
it
and
what
I'm
trying
to
say
you
know
argo
cd
when
it
give
a
resource,
it
know
the
resource
type
and
basically
the
code
would
look
like.
Oh
if,
if
that
particular
resource
is
crd,
then
use,
as
you
know,
hardcoded
settings
kind
of
take
user
settings
and
apply
on
top
of
them
predefined
settings
for
resource
diffing
and
these
settings
would
ignore,
preserve
unknown
fields
and
we
already
have
status.
Basically,
we
already
have
code
that
ignores
status.
A
H
We
have
yeah,
we
have
some
exceptions
for
known
quirky
behavior,
for
differing,
so
creation,
metadata.creation
timestamp
being
null
is
a
very
common
thing
that
appears
in
the
manifest
and-
and
we
know
that
it
has
no
effect
on
differing.
So
we
special
case
and
ignored
that
this
is
kind
of
another
exception.
H
We
know
that
with
crdb
wines
preserve
unknown
fields
can
only
be
false,
so
so
there's
so
if
we
ever
see
it
false
and
there's
a
diff
for
that
being
empty,
then
we
can
also
accept
that
and
and
say
and
ignore
it
as
a
difference.
So
I
think
that's
the
feature
since.
C
A
And
I
just
found
that
code,
so
it
literally,
you
know,
does
that.
So
basically
it
says
that
if
you
so
we
can,
we
programmatically
can,
you
know,
combine
creating
this
resource,
diving
settings
and
and
then
we
have
special
logic
for
charities
and
we're
already
ignoring
status
field,
so
that
code
in
particular
settings
that
will
then
ignore
status,
and
we
can
add
more
code
here
to
ignore.
C
A
A
Right
yeah,
we
can
add
four
from
here
and
we
just
spoke
about.
I
mean
we
already
got
one
because
the
first
one
is,
you
know.
E
E
I
I
felt
I
felt
we
need
either
we
need
to
to
rename
the
needs
discussion,
column
or
we
will
need
a
new
column
that
says
needs
proposal,
maybe
but
yeah
we
we
can
discuss
this
later.
So
sorry,.
A
Okay
right
so,
and
we
have
next
feature
and
that's
I
guess
it
won't
be
as
easy
so
idea
that
kind
of
people
are
saying
who
could
you
just
use,
helm
install
you
know
for
help
chats.
E
I
think
I
think
this
is
this
is
about
the
the
install
crd
flag.
A
A
Okay,
so,
okay!
Basically
we
are
not.
E
E
I
H
A
E
A
Okay
yeah,
at
least,
if
it's
in
that
colon,
we
should
not
work
on
it
right
away.
All
right.
Let's,
let's
move
on
to
the
next
one,
okay,
this
one-
I
I
think
I
oh
it
come
how
come
it
got
closed?
Oh
all,
right
I
closed
it
as
a
duplicate
and
incorrectly
added
duplicate
back
into
the
list.
So
the
right
one
is
this
one.
A
All
right
so,
and
it's
also
in
the
proposal
so
f,
5,
15.,
okay
and
yeah.
Let
me
fix
it
right
now,
so
I
I
meant
to
have
that
ticket
here
and
simply
remove
that
ticket
from
from
the
project,
because
it's
closed
already
and
both
are
about
the
same
so
and
request-
is
to
hide
the
buttons
that
user
cannot
click
anyway,
because
of
you
know,
you
don't
have
permissions.
A
Yeah
and
this
I
assume
this
one
is
simple.
I
think
we
want
to
do
it
yeah
any
objections.
A
I
A
To
me,
it's
we
need
to
make
a
decision
where,
if
we
want
to
do
it
or
not
and
then
talk
about
details
all
right
next
one,
this
one
won't
be
that
easy.
E
I
guess
yeah,
but
I
think
this
is
this
is
duplicated
right
because
yeah,
because
the
other
proposal
we
were
chatting
about
it
has
basically
a
similar
goal
right.
So.
C
E
Application
length
proposal
probably
gets
extended
to
include
this
use
case
as
well
yes,
or
what?
What
this,
what
this
enhancement
issue
is
trying
to
solve,
at
least.
E
A
To
kind
of
so
we
have
few
issues
with
how
we
label
resources
that
belong
to
application.
Right
now
we
use
kubernetes
label
and
kubernetes.
The
value
of
that
label
is
application
name
and
it
has
limitations.
One
is
mentioned
in
this
proposal.
Basically,
the
label
value
cannot
be
longer
than
63
chairs,
and
so
that
you
know,
application
name
is
kind
of
length
of
it
restricted
as
well
and
a
few
other
issues
is,
you
might
have
two
argo
cds
managing
same
app
in
one
cluster
and
then
kind
of
name.
A
E
I
guess
young
was
strength.
This
is
this.
Is
this
yeah?
This
is
a
very
common
problem
with
with
operator
and
it's
it's
buried
deep
down
in
the
operator
sdk.
So.
E
H
We
can
we
client,
let's
close
the
issue,
because
it
says
right
now:
the
issues
still
open.
Okay,
yeah,
that's
a
good
idea.
Let's.
E
C
E
H
A
Yeah,
oh
look:
one
of
the
accepted
feature
requests
already.
H
G
A
H
Because
if
we
have,
if
we,
if
someone
says
skip
crds,
then
this
would
satisfy
this
use
case.
I
think
I
don't
know
if
it's
in
the
oh
yeah
you're
checking
if
it's
in
here.
A
H
A
A
H
A
Might
be
like
the
one
reason
to
do
it
is
maybe
if
you
use
help
repo
once
but
the
tripo
must
be,
you
know
open
for
everyone,
then
we
can
alter
the
cheaper
and
then
next
user
will
get
out
of
completion
kind
of
that's
the
only.
I
Yeah
I
mean
this
is
definitely
problematic,
because
helm
repo
ad
is
something
today
is
something
that
lives
on
your
file
system.
And
then
I
wonder
if
we
want
to
get
into
that
business
of
storing
things
in
the
file.
E
E
This
guy
wants,
he
wants
to
just
drop
in
a
url
for
a
helm,
chart
and
doesn't
want
to
he
he
you
know
it's
like.
He
specifies
the
chart
with
like
a
full
ul
and
doesn't
want
to
add
the
the
repository
to
our
city
right.
A
Yeah,
that's
my
understanding,
so
it
must
be
chart
in
a
public
health
repo,
because
if
it's
private,
that
means,
you
must
add,
you
know,
parameters.
Okay,
let
me
ask
him
in
okay:
it
basically
means
discussion
needed.
We
cannot
even
agree
or
disagree
because
we
don't
really
understand
what
he
wants.
H
A
E
This
this
I
mean
this
would
be
like
we
had
a
git
repositories,
yeah.
C
E
But
but
the
next
person
will
not
get
a
drop
down
from
that.
So
if,
if
I
start
with
a
blank
repository,
configuration
and
just
add
github.com,
so
the
argo
cd
example
apps
it
will,
it
will
just
fetch
the
fetch,
the
the
data
from
the
repository
without
adding
I
I
think
he
doesn't
want
to
actually
add
the
repository
the
the
heim
repository
to
our
cd,
but
I
think
he
wants
to
just
specify
the
repository
in
in
his
application
like
like
with
git
right.
I
see
if
that's.
C
A
All
right,
let's
move
it
into
this
discussion:
okay
and
then
next,
I'm
just
going
to
pick
few
from
the
top
until
we
run
out
of
time.
A
A
Which
this
customize
belongs
to
and
so
that,
if
remote
base
is
accessible
from
credentials
of
current
repository,
then
everything
is
going
to
work.
But
basically,
if
you
need
two
set
of
credentials,
first
credential
credentials
to
get
you
know
first
repository
which
has
this
customized
and
then
another
set
of
credentials
to
access
remote
base.
Then
this
use
case
is
not
supported
and
I
guess
this
person
suggesting
that
we
could
change
this.
A
A
We
pass
helm
credentials
all
waitlisted
contradictions
to
the
help
chart
also
to
the
repo
server
and
then
repo
server,
I
would
say,
like
basically
execute
helm
login
for
all
dependencies.
H
You
for
the
help
I
mean,
there's,
there's
a
little
bit
more
consequence
for
the
customized
one,
because
it
can
access
get
repos
to
that
contain
other
stuff,
and
then
you
say
like
oh,
I
need
the
source
code
from
this
other
repo
and
then
I
use
customized
abuse
to
get
to
somehow
getting
the
source
code
of
the
and
maybe
use
like
a
configmap
generator
to
actually
view
the
files.
H
That
means
it
needs
discussion,
and
I
think
it
would
be
fine
if
we
just
if
those
repos
were
in
the
the
allow
list
of
the
project.
H
G
E
A
All
right,
it
is
10,
I
mean
basically
it's
end
of
meeting
right
now,
so
I
think
we
and
we
did
almost
10.
So
maybe
we
can
take
like
one
more
minute.
Just
to
summarize
like
do
you
think
this
process
is
working,
this
kind
of
triggering
of
features
every
every
meeting
we
spend
20
minutes,
but
at
least
we
you
know
we
made
some
decisions
right.
H
E
We
it's
time
to
do
it.
What
needs
discussion
from
my
point
of
view?
Is
that
so
sure
we
can
we
can
someone
have
contribute
like
making
making
it
built
on
the
other
platforms,
but
we
can't
actively,
you
know,
ensure
that
this
will
still
work
going
forward,
because.
E
To
test
it-
and
this
this
is
this
is
my
concern
with
that
right,
so
I
know
we
have.
We
have
integrated
arm
built
support
from
someone
else
in
the
past,
but
it's
it's
constantly
running
out
of
you
know,
out
of
sync,
so.
H
I
interpreted
this
as
the
arm
binaries
that
people
download
on
the
the
new
m1
max,
not
the
container
images
I
wasn't
sure
which
one
this
was.
I
see
images
in
the
title,
but
the
cli
makes
sense,
because
everyone
eventually
will
need
arm
images.
I'm
sorry
arm,
binaries
images
doesn't
makes
less
sense
to
me.
A
A
I
One
quick
feedback
on
this
process
would
be
every
time
we
decide
on
declining
something
or
every
time.
We
think
that
something
needs
a
future
discussion.
It
would
be
great
if
you
could
invite
the
proposal
author
to
join
our
meeting
and
potentially
present
the
case,
especially
if
you're
declining
it
before
we
close
it.
We
should
say
hey.
This
has
been
provisionally
declined,
but
we
would
love
to
hear
your
thoughts
on
this
then
give
them
a
chance.
E
I
think
this
also
this
this
process
should
we
should
document
it
so
so
like
community
facing
right,
so
that
people
know
what
to
expect
and
in
this
documentation
we
probably
should
also
say
so.
We
may
invite
you
to
to
this
meeting
or
you
can
invite
yourself
to
present
your
case
or
whatever
right,
yep
agreed,
yeah,
yeah
awesome,
very
incredible!
That's
yeah!
Truly!
I
agree
that
it's
just
fair
to
to
let
those
people
know
well
in.
E
Yeah
I
mean
the
the
the
golang
people.
They
have
two
two
additional
columns,
that's
likely
decline
and
likely,
except
so
yeah,
probably
what's
moved
in
likely
decline
is
then
subject
to
like
inviting
someone-
and
you
know
defending
his
case
or
whatever
so
yeah,
but
we
we
can.
We
can
see
moving
forward,
what
what
what
we
need
and
what
we
don't
need.
E
E
A
You
know
the
kind
of
closed
status.
Basically,
I
feel
like
we
should
not
have
closed
issues
here
if
we
close
that
we
just
remove
it
from
the
project.
So
if
we
move
it
and
it
stays
active
and
declined,
that
means
we
still
need
to
communicate.
You
know
we
still
need
to
talk
this
person
who
requested
it.