►
From YouTube: Argo Contributors Office Hours Jan 20th 2022
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
Okay,
hello,
everyone
welcome
back
to
the
argo
contributor
office
hours.
I
will
be
your
host
today,
so
yeah,
let's
get
started.
First
order
of
business
should
be
to
do
the
github
discussions
moderators,
so
ishida
and
alex.
Do
you
guys
want
to
talk
about
last
week
at
all,
or
should
we
just
move
on
to
deciding
next
week's.
D
I
I
had
one
observation
all
right,
so
this
time
I
had
to
answer
a
few
questions
that
were
in
documentation.
It
was
just
I
guess,
user
couldn't
confide
it.
So
I
feel
like.
If
we
improve
documentation,
it
will
reduce
the
amount
of
support.
That's
but
nothing
else
like
I.
I
did
not
see
anything
like
no
alarming
reports
or
something.
E
F
A
I
was
just
going
to
ask
alex
which
portion
of
the
docs
did
you
find
yourself
linking
to.
D
I
think
there
was
thinking
the
question
about
thinking
and
then
so
why
I
start
thinking
about
documentation
is
because,
right
now
we
have,
I
think,
three
different
dogs
that
expect
describe
seeking
and
so,
and
it
took
me
some
time
to
find
you
know
all
these
dogs
and
send
it
to
a
user.
D
D
The
question
is
actually
this
particular
one:
it
was
in
support
slack,
the
user
was
just
asking
about
customizations
and
yeah,
and
we
have
resource
hooks
and
sync
options
and
seeing
strategies,
and
so
I
didn't
know
what
he
was
asking
about.
So
I
had
to
give
him
all
three
and
then.
E
D
D
A
F
I
work
with
sync
windows
a
while
back,
and
I
think
I
implemented
some
tests
with
different
scenarios
I'll
try
to
find
that
if
I
do
I'll
reply
in
this
in
this
ticket
referring
the
user
to
the
to
the
tests,
if
that
makes
sense,
I
think
it.
A
All
right,
so
we
should
probably
decide
who
will
be
moderators
for
next
week.
So
do
we
have
any
volunteers
to
start.
A
All
right,
so,
if
we
just
go
next
on
the
list,
it
looks
like
we
have
dan
would
be
next
and
jesse
would
be
next.
I
think
we
should
probably
only
pick
one
of
them
as
primary,
because
I
think
you
guys
could
both
be
primary,
but
maybe
give
someone
else
a
chance
to
be
secondary,
so
jesse
or
dan.
Do
you
want
to
volunteer
for
primary.
A
A
A
So
that's
good
all
right,
so
dan
you'll
be
primary
and
you'll
be
secondary,
all
right.
Okay,
we
can
move
on
to
the
next
topic.
Unless
anybody
else
has
things
they'd
like
to
discuss
regarding
moderation,
all
right
so
alex
it
looks
like
you
put
the
2.3
blocking
issues.
Do
you
want
to
talk
about
this?
A
little
bit.
D
That
that's
a
it
will
be
short
update,
so
we
kind
of
had
a
discussion
in
slack
with
jen,
jesse
and
all
approvers,
so
it
kind
of
2.3
is
much
smaller
right
now,
so
it
it
had
a
lot
of
issues
here
and
most
of
those
issues
doesn't
either
bugs
or
feature
requests
that
a
long
time
ago
got
into
milestone.
Then
we
let
it
sleep
and
we
were
just
moving.
D
So
I
did
that
and
basically
right
now
we
have
very
few
items
here,
a
few
of
them.
Okay,
and
maybe
I
can
walk
you
through
for
them
hopes
application
dependency
is
definitely
not
supposed
to
be
here
unless
jc
added
it
back.
But
it's
a
big
feature.
I
think
that
supposed
to
be
moved
forward
to
2.4
so
basically
ignore
the
first
one
aging
application
stack
application
set
to
argo
cd
tech.
I
feel
like
it's
almost
done
issued
in
jonathan
driving
it
and
the
next
ticket.
D
It's
a
it's
a
very
small
win
and
I
was
hoping
to
implement
it
basically
right
now,
so
this
is
right.
Now
a
radius
is
on
a
critical
path.
We
use
it
to
to
log
in
and
we
should
have
get
rid
of
it
for
a
long
time
and
I
basically
discovered
the
ticket
when
I
was
grooming,
a
list
of
2.3
issues:
okay,
merge
applications
set
into
rbc
installation
same
as
the
first
one
async
application
set,
and
then
namespace
resources
owned
by
cluster
scope;
resources
not
visible
in
ui.
D
D
D
D
Yeah,
sorry
not
notice
one,
but
maybe
there
are
two
of
them
and
let
me
select
you
the
right
one.
No,
it's
like.
D
The
the
change
is
kind
of
it's,
not
extremely
big
here,
if
you,
if
you
click
on
file
changes,
so
we
have
kind
of
a
simple
function
here
and
the
goal
of
the
function
is
to
take
a
parent
and
then
look
through
all
other
resources
and
look
for
children,
and
so
what
we
had
here
before,
we
were
only
looking
at
resources
in
the
same
name
space.
But
cluster
objects
has
no
name
space.
That's
why
logic
wasn't
working
and
and
then
the
simple
solution
is
to
start
looking
everywhere.
D
But
imagine
if
you
have
a
cluster
with
like
a
few
thousand
hundreds
of
objects,
so
it
will
be
slow
for
sure,
and
the
only
other
option
is
to
add
more
indexes.
It
is
that's
and
it
doesn't
look
too
complex,
but
I
have
no
idea
if
it's
going
to
work.
Basically,
the
only
way
to
see
if
it
improves
or
makes
performance
worse
is
to
implement
it
and
try
it
and
it
seems
like
it,
doesn't
really
fit
into
2.3.
A
D
D
We
would
have
lots
of
if
else's,
at
least
three,
if
else
so
one
if
here
and
then
two
in
two
other
places
where
we're
supposed
to
maintain
that
index.
G
D
We
will
spend
a
little
bit
more
time
kind
of
maintain
you
know
putting
data
into
index.
So
every
time
when
something
changes
we
will
supposed
to
update
index-
and
you
know,
update
the
parent,
but
that
the
methods
that
I
was
showing
before
you
know
iterate
children-
it
will
be
much
quicker
and
the
hope
is
that
it
will
actually
help
to
improve
performance
because
right
now
it's
a
rate
hierarchy.
D
D
C
Yeah
yeah,
I
think
it's,
it's
also
not
a
block.
It's
it's,
not
a
blocker.
For
for
the
release.
Right,
it's
like
is
a
new
feature.
I
don't
know
how
how
high
it
was
in
demand.
D
C
Well,
I
I
personally
think
that
this
this
is
a
great
to
have
feature
right,
because
I
mean
that's
that's
what
people
are
looking
for
in
the
ui
like
for
the
visibility
and.
C
But
yes,
I
I
also
agree
that
if
you
have
concerns
about
the
performance,
we
may
want
to
postpone
that
and
and
conduct
some
tests
before
we
release
it
into
the
wild.
G
Okay,
what
I
can
do
is
I'll
try
to
address
your
comments
alex
as
soon
as
possible,
and
you
can
understand
more
and
do
some
testing,
and
only
if
we
are
confident
enough,
we
can
include
it
in
the
series
or
we
can
postpone
it.
D
D
D
Thank
you
all
right
so
and
then
I
want
to
skip
one
issue
and
basically
end
point
if
it's
confused
by
port
sorting
back
it
was
here,
but
we
could
not
reproduce
it
ever
like
enough.
I
think
I
produced
it
yesterday,
so
I
kept
it
here.
I
understand
lots
of
such
bugs
because
they've
been
open
for
like
a
year,
and
this
is
one
of
those
bugs
it's
like
it's
a
bug
that
was
open
for
a
long
time,
but
I
kept
it
because
I
know
how
to
fix
it.
To
fix
this
place
is
small.
D
It's
in
in
github's
engine,
no,
not
this
one.
That
was
previous
attempt,
so
someone's
yeah
and
then
basically
I
have
questions
about
two
other
remaining
issues
into
the
three
milestone,
because
I
didn't
add
those-
and
I
don't
know
if
anyone
knows
if
anyone
actually
work
on
those.
So
one
is
unable
to
connect
private
repository
and
then
reconciliation
loop,
one
and.
D
I
I
know
that
jan
was
working
on
the
reconciliation
loop
yeah,
you
added
some
logging.
Do
you
know
like?
Are
you
planning
to
do
anything
else
for
the
ticket.
C
I
have
just
sent
this
pr
for
for
logging,
which
resource.
C
Is
triggering
the
reconciliation
okay
so
so
that
will
help.
C
Yeah
I,
but
I
think
so,
the
the
reconciliation
loop
I
mean
there
are
different
so
that
I
I'm
not
sure
if
we
can,
if
we
can
fix
it,
because
we
we
start
a
reconciliation,
if
a
managed
resource
is
changed
right.
So
we
we
have
a
watch
on
that
resource
and
then,
if
someone.
C
Resource
then
we
trigger
reconciliation,
maybe
maybe
a
path
going
forward
would
like
be
to
to
have
some
kind
of
throttling
per
resource.
I
don't
know
if,
if,
if
resource
changes
a
thousand
times
per
minute,
then
maybe
we
want
to
catch
that,
but
yeah
I
I
don't
know
so
I
was.
I
was
not
planning
to
work
on
that
for
1.
D
D
C
C
D
C
I
think
pasha
has
assigned
himself.
A
Yeah
it
looks
like
pasha's
assigned,
but
he
hasn't
commented
since
december.
Are
you
here
pasha.
D
D
I
I
I.
D
D
Not
not
a
blocker,
because
I
think
it's
been
there
forever
like
people,
you
know
that
the
ui
is
not
most
popular
way
to
add
repo,
because
you
have
to
copy
past
credentials.
It
looks
like
we
have
a.
H
Like
I
I
I
understand
what
is
this
item?
I
I
spent
few
days
for
a
producer
and
it's
actually
not
reproducible
at
all,
and
by
quote
I
I
didn't
saw
that
it
can
happen.
So
this
kind
of
error
that
communication
is
broken
between
client
side
and
server
side.
If
it's
true,
it
would
be
some
white
error
for
everyone.
H
C
Yeah
I
I
would
agree
with
that,
because
if,
if
it
would
be
like
that,
nobody
could
use
the
ui
to
connect
a
repository
using
ssh,
then
I
guess
there
I
share.
I
share
the
point
of
view
of
pressure
that
probably
many
more
users
would
have
come
up
with.
H
I
I
I
I
tried
to
reproduce
it
really
in
few
days
in
different
versions,
and
it
not
reproducible
issue
this
issue.
That
means
that
insecure
flag
from
ui
passes,
not
boolean,
some
empty
string
or
something
like
or
something
else
and
golden
server
cannot
convert
it
to
boolean
and
throw
such
error.
But
again
I
I
I
didn't
represent.
D
All
right,
so
I
feel
like
it's
specifically
safe
to
just
unassign
it
and
yeah
and
then
call
it
yeah.
D
And
I
I
have
one
more
quick
item.
I
think
it's
kind
of
I
already
unassigned
server
side
apply
from
2.3
and
I
just
wanted
to.
You
know
clarify
that.
I
did
it
right.
I
know
that
leo
was
trying
to
you
know,
drive
that
forward,
move
it
forward
and
jan
had
concerns
that
it's
not
that
easy,
as
it
looks
like
and
then
so.
Basically,
we
want
to
make
sure
everyone
agrees
that
it
should
not
go
into
2.4
and
we
move
it
to
two
should
not
go
to
two.
The
three
should
be
postponed.
C
Yeah,
just
just
for
some
more
context
that
and
that's
a
feature-
that's
wanted
by
many
right,
and
I
think
that's
also
something
we
from
from
red
hat,
look
for
as
some
significant
change
and
we
should
get
it
right
right,
so
maybe
maybe
not
rush
it
into
into
2.3
and
then
see
that
we
need
breaking
changes
going
forward.
C
D
Yeah,
I
I
I
can.
I
agree.
Basically,
I
I
was
tempting
to
edit
to
2.3,
because
we've
got
pull
requests
and
it
feels
like
a
you
know
like
easy
way,
but
you're
right.
We
might
regret
that
we
just
got
it.
C
Yeah
and
it's
it's,
it's
very
awesome
that
someone
sent
a
pr
for
that
right
so
and
I
guess
the
the
submitter
he
he
he
contributed
to
argo
cd
before
and
I
I
think
this
this
contributor
is,
is
very.
C
Smart
so
like
he,
he
worked
with
us
to
get
the
other
pr's
right
and
I
I
think
he
will
work
with
us
on
this
one
as
well
so
cool.
D
A
Thanks
alex
john,
it
looks
like
you
have
a
agenda
item
for
github
actions,
approval
changes.
C
Yeah,
just
something
that
I
wanted
to
bring
up
for
discussion
is
that
I
think
github
has
changed
the
default
for
executing
the
the
github
actions
workflows
right,
especially
for
for
so
I
think
there
was
a
spam
wave
a
couple
of
months
ago,
where
people
abused,
github
actions
in
popular
repos
to
mine,
some
crypto
coins,
or
something
like
that
and
then
github
actions
introduced
this.
This
feature
that
you
need
to
approve
the
run
for
the
workflows
right
and
made
it.
I
think
they
made
it
the
default.
C
So
every
new
contributor
needs
approval
from
someone
with
commit
permissions
to
the
repository,
and
there
are
three
settings
for
this
and
I
think
we're
we
are
using
the
default.
C
I
don't
know
if
it's
an
approach,
a
deliberate
choice
or
if
it
just
was
like
taking
the
defaults
from
github,
and
there
is
another
option
that
says,
require
approval
for
new
github
members
and
I'm
not
sure
if
we
should,
if
we
should
change
to
that,
because
what
I
have
observed
is
that
many
people
are
sending
prs
and
you
know
someone
has
to
come
around
and
push
the
approve
and
run
button
before
before
people
actually
can
see
whether
ci
passes
and
things
like
that,
and
I
I
guess
we
are
very
bad
at
improving
those
workflows,
considering
that
there
are
quite
a
few
that
still
require
approval,
and
even
if
we
approve
it
and
the
ci
fails
and
the
contributor
pushes
a
new
change,
then
we
need
to
improve
again
and
again
and
again
and
again
so
to
to
make
all
of
our
lives
easier.
C
D
D
We
had
this
type
of
pull
request
once
like
long
time
ago
and
never
got
the
you
know
the
pr
that
tries
to
mine
cryptocurrency
again,
so
maybe
it's
safe
to
switch
to
that.
You
know
less
one.
Conversion.
A
Cool,
so
it
sounds
like
we
all
agree.
We
can
switch
to
that
any
other
thoughts
on
that
topic.
D
I
was
just
thinking
you
know
about
consequences,
most
likely.
If
some
user
sends
this
type
of
pull
request,
he
will
be
blocked
because
I
mean
we
will
report
him
and
then
and
then
he
will
be
blocked.
So
it
looks
like
a
good
protection.
So
basically
the
only
option
for
for
hackers
is
to
keep
creating.
D
D
A
Cool
sounds
good
thanks.
John
all,
right
so
leo
looks
like
you
up
the
last
topic
regarding
snick.
F
Yeah,
this
is
more
informational
topic.
I
think
it's
gonna
be
very
quick
as
well
just
to
inform
everybody
that
I
worked
this
week
with
alex
and
hari.
We
enabled
sneak
as
a
pr
check
in
in
the
argo
cd
repo,
so
we
this
can
be
configured
in
different
ways,
the
way
it
is
configured
now
it
will
check
the
it
would
check
the
docker
container,
but
not
for
everything
in
there.
So,
for
example,
it
will
mainly
check
for
addition
additions
added
in
in
the
pr
itself.
F
It
doesn't
check
it
all
the
way
from
the
past,
so
the
the
the
validations
that
we
get
in
the
pr
is
just
the
validations
added
by
the
the
the
the
commuter
we
can
change
that
if
we
want,
but
I
guess
this
is
more,
is
this
is
less
intrusive
yeah
and
we
but
anyways
whatever
is,
is
making
more
sense.
We
can.
We
can
change.
A
Cool
yeah
thanks
leo
okay,
thank.
F
A
D
Had
a
super
quick
informational
topic,
I
so
we
immersed
yesterday
a
pull
request
that
upgrades
github
clients
to
certain
github,
kubernetes
client
to
most
recent
version,
and
we
also
had
to
upgrade
golang
to
1.17
and
the
reason
is
kubernetes
client
actually
uses
some
1.7
golden
feature
and
it
won't
compile
on
1.6.
So
just
be
aware,
you
will
have
to
upgrade
gold
link
on
your
laptop
as
well.
A
Great
cool
thanks,
alex.
A
Okay,
well,
if
not,
then
I
guess
we
can
get
10
minutes
for
our
day
back
and
we'll
see
you
next
week
thanks
everyone.
I.