►
From YouTube: Argo Contributors Office Hours Mar 17th 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,
welcome
to
contributors
meeting
and
I
will
be
your
host
today
and
I'm
not
going
to
share
my
screen
here,
because
we
don't
have
much
on
the
agenda.
A
Just
two
topics
and
first
topic
is
feedback
about
feedback
from
the
person
who
was
treasuring
the
issues
today.
I
think
it
was
real
and
leo
anderson
I
see
leo
on.
The
call
can
please
share
if
you
have
anything
to
share.
B
Yes
sure
so,
yeah
no
outstanding
issues.
This
week
I
had
a
good
interaction
with
rishab
during
the
week.
He
was
pinging
me
and
for
for
issues
he
didn't
know
how
to
how
to
proceed.
So
we
were
able
to
close
a
few
issues.
I
guess
we
just
have
one.
B
That
is
that
I
didn't
know
about
how
to
correctly
reply
to
the
user.
It
seems
like
a
a
an
issue
in
our
docs,
but
basically
there
is
a
link
in
our
metric
documentation
that
goes
to
grafana
dot,
argo
proj,
dot,
io,
something
like
that.
C
B
A
Manifests
allocated
in
available
open
source,
it's
in
argo
approach
or
argo
brush
deployments
repo,
but
and
it's
hosted
in
google
on
a
kubernetes
cluster
and
basically
provided
by
intuit,
and
not
many
people
obviously
have
access
to
it,
and
I
think
it
has
to
be
me.
I
have
to
go
and
fix
it
and
most
likely
we
don't
run
h.a
grafana,
basically
there's
just
one
instance
of
grafanapod
and
I'm
guessing
it's
diet
and
we
don't
have
monitoring
for
for
that
graph.
B
It's
okay,
okay,
cool,
so
I'll
check
that
with
you
offline,
but
that's
all
about
it.
No
more
issues
to
report.
A
Cool,
thank
you.
Thank
you.
Oh
before
we
move
on,
we
need
to
find
who
is
going
to
do
triaging
next
and
I
I
want
to
volunteer
basically
last
time
I
was
primary.
I
didn't
really
do
tearing
much,
but
I'm
pretty
sure
next
week
will
be.
I
will
have
more
time
so
I
can
be
primary
for
for
the
next
week
and
we
need
secondary.
A
Let
me
know
please:
if
you
want
to
be
a
secondary
for
next
week,.
A
Oh,
thank
you,
I'm
looking
at
list
of
topics
for
today,
so
there
is
just
one
for
me.
I
just
wanted
to
say
congratulations
to
new
argo
cd
approvers
and
the
viewers
actually,
and
I
think
we
it
didn't
happen.
It
doesn't
happen
often
for
finally,
we,
the
argo
cd
finally
got
a
bunch
of
new
approvals
yesterday
and.
E
A
Yes,
I
think
it's
a
big
deal
and
yeah.
I
want
to
name
everyone,
so
it's
keith
michael
leo
and
pasha
pasha.
He
is
not
here
and,
and
they
became
a
reviewer
terry,
who
is
already
approver
of
workflows.
He
became
active
in
cd
and
he's
also
a
reviewer
there,
and
I
think
why
I
think
it's
a
big
deal
is
because
we
pretty
much
double
number
number
of
approvers
in
our
cd
and
hopefully
now
there
is
no
no
need
to
being.
You
know
very
few
approvers
who
are
active.
A
A
Yeah,
so
we
are
basically
congratulating
new
approvers
in
the
project
yeah
and
I
I
I
wanted
to
just
say
that
in
my
opinion,
you
should
not
be
shy
to
merge,
prs
basically
and
break
things,
because
that's
how
we
learn
so
feel
free
to
you
know,
push
merge
everything
you
think
should
be
merged
and
if
something
goes
wrong,
we'll
fix
it
together.
A
Right
yeah:
no,
we
cannot
applause,
you
because
we're
in
zoom
and
that's
it
actually
now
we
either
need
new
topic.
Oh
yeah,
okay.
We
have
a
bunch
of
congratulations
in
chat
and
there
is
nothing
else
planned
for
for
today,
okay
and
yet
before
we
wrap
up,
I
can
just
share
with
you
that,
like
last
week,
we
spent
working
on
fixing
some
vulnerabilities
in
argo
city.
I
think
you
noticed
there
was
some
fixes.
A
We've
got
one
more
and
I
cannot
share
details
before
we
publish
the
cve,
but
that
was
keeping
some
of
us
busy
last
week
and
I
think
just
want
to
share
that
update,
because
that's
why
we're
not
super
responsive
and
once
we
are
done
with
the
last
cv,
we
will
be
more
responsive
again,
all
right.
Anyone
wants
to
talk
about
anything
else.
We
have
like
30
minutes
left.
F
Oh,
there
was
a.
There
was
a
pr
somebody
was
asking
about
the
other
day
and
I
wanted
to
see
yeah.
Let
me
see
if
I
can
pull
it
up,
because
I
wanted
to
see
if
it's
been
sitting,
it's
kind
of
a
design
proposal
that
I
think
was
from.
A
We
spoke
about
it
yeah.
You
should
reminded
me
that,
as
I
understand,
red
hat
wants
to
work
on
it
in
the
next
release
and
yeah.
I
kind
of
promised
you
to
also
you
know
to
do
the
review
and
we
were
trying
to
refresh
our
roadmap,
and
I
in
there
is
a
pull
request
for
me
that
tries
to
you
know,
update
list
of
items
that
we
are
working
in
the
next
release
and
I
added
multi-source
support
in
the
list
as
well.
E
G
A
I
was
trying
to
say
that
I
saw
your
comment
in
the
pull
request
related
to
roadmap
about
multiple
sources
in
application
and,
as
I
understand
you
want
to
work
on
it,
so
I
I
basically
moved
I
updated
my
pr
and
moved
that
multiple
sources
story
into
the
next
release.
Yeah-
and
I
was
asking
you
know.
If,
if
I
misunderstood,
you
please
tell
me,
and
I
will
move
it
back
to
future.
A
Okay,
yeah
so
and
but
yeah,
it's
still
missing
review,
but
you
know
I
will
try
to
as
soon
as
possible.
A
A
Okay,
I
right
now.
This
is
this:
is
it
for
real,
any
any
other
topics,
or
we
can
wrap
up
early
today.
C
Just
wanted
to
check
on
analysis
controller
that
we
use
in
are
girl
arts.
Can
it
be
also
made
like
a
analysis
engine
as
a
separate
project,
I'm
just
thinking
through
before
I
put
up
a
proposal
for
that.
I
wanted
to
get.
You
know
your
thoughts
on
it.
C
Can
it
be
made
as
a
separate
project
for
because
it's
a
controller
in
itself
today
it
is
embedded
in
robots,
but
can
we
take
it
out
like
a
notification
engine?
Can
we
write
something
like
analysis
engine
that
can
be
integrated
into
multiple
projects,
not
just
sit-in
products.
C
I
Yeah
that
should
be
possible.
I
mean
I,
I
wasn't
sure
if
you
wanted
to
use
it
as
a
library
versus
the
the
crs,
so
what's
possible
is
that
you
can
saw
roll
out
but
not
use
the
rollout
and
not
use
experi
experiments,
but
only
use
analysis,
templates
and
analysis
runs.
I
I
think
what
is
might
be
possible
is
just
to
add
command
line
flags
to
like
disable
the
like
the
go
routines
that
start
the
rollout
controller
and
the
experiment
controller,
and
have
it
only
do
the
analysis
stuff
and
that
way
people
have
an
option
just
to
install
analysis
portion
is
that
is
that
that's
what
you're
looking
for
right.
C
Right
so
so
I
don't
know
if
I
got
completely
wha,
what
do
you
what
you
propose
here,
but
let
me
reiterate
what
I
understood,
but
what
you're
asking
what
you're
saying
is
that
in
the
rollout
controller
itself,
we
can
change
the
command
line
flags
to
just
run.
The
analysis
is
what
you're
saying
right.
I
Yeah
kind
of
like
how
cube
api
server
is,
like
a
whole
bunch
of
controllers,
compiled
into
a
single
binary
right
right
now,
that's
true
for
rollouts,
with
three
different
ones.
There's
rollout
experiment
analysis
runs,
there's
no
way
to
only
run
like
some
of
them,
but
it
should
be
easy
to
add
a
flag
to
just
run
like
the
analysis,
and
that
way
you
can
just
have
it,
be
an
analysis,
controller
and
choose
not
to
install
the
rollout
or
experiment
sears.
C
I
Oh
okay,
yeah,
it
depends
if
you
want
to
use
it
as
a
library
or
if
you
actually
want
to
use
the
crs,
so
it
using
as
a
library
is
going
to
be
some
more
work
kind
of
like
how
github's
engine
was
a
lot
of
work
because
because
it
brings
in
a
ton
of
kubernetes
dependencies
and
that
doesn't
make
it
very
friendly
as
a
library
okay,
but
I
think
if,
if
you
did
want
to
use
the
just
the
crs,
that
should
be,
I
think,
pretty
easy
to
to
enhance
it
just
to
run
selective
controllers.
I
Yeah,
that
was
actually
the
intent
of
breaking
it
down
that
this
way,
so
analysis
runs
and
templates
have
actually
no
concept
of
a
rollout.
That's
that's
how
it's
able
to
be
used
by
both
the
experiment,
cr
and
the
rollout
cr.
So
if
there's
another
like
keiko
specific,
because
some
resource
that
wants
to
do
analysis,
it
that
should
be
completely
possible.
G
Actually,
I
had
one
more
question
regarding
the
road
map,
so
there
was
one
more
proposal
by
me
that
was
regarding
sharding
application
controller.
So
right
now
it's
added
for
future
future
releases.
I
I
was
just
thinking
like
if
we
are
even
planning
for
2.5,
would
it
be
better
that
we
have
the
proposal
check
in
this
time
time
frame,
because
then
it
would
be
easier
to
work
in
the
next
release.
A
C
A
A
I
want
to
try
to
find
time
to
review
it
from
you
know
as
soon
as
possible
to
proposals
from
from
you,
shita
and
and
jan's
proposal
yeah,
and
thanks
for
reminding
about
it,
and
I
encourage
other
approvers.
Now
we
have
so
many
approvals.
We
can
review
these
proposals
yeah.
D
D
Have
some
ideas,
additional
ideas,
yeah
and
just
drive
it
to
something
that
would
benefit
the
project
right
so
yeah?
Basically
we're
calling
for
everyone
to
review
it
right,
not
only
reviewers
or
members,
but
also
to
the
community,
because
it
will,
in
the
end
affect
the
community
right.
So
they
they
shall
have
the
word
as
well.
So
that's
why
we
have
those
proposals
up
in
the
public
yeah,
but
I
I
agree
with
you
alex
that.
D
At
least
the
the
approvers
should
have
a
look,
and
I'm
also
I'm
I'm
having
a
debt
there
so
to
review.
The
other
proposes,
that's
not
being
sent
by
me.
I
A
D
It's
it's
basically
removing
removing
the
single
point
of
failure
that
we
currently
have
in
the
application
controller.
Alright,.
D
I
With
yeah
all
competing
for
the
shard
id
is
still
incorporated
into
the
name,
but
then
they're
just
renamed,
and
they
all.
D
A
D
You
you
could
have
like
a
a
leader
election
on
an
application
resource.
I
think,
like
you,
know,
a
chart
competing
so
all
instances
in
a
chart
competing
for
for
the
applications
to
manage
the
application
resources.
That
would
be
one,
but
I
think
it's
yeah.
It's
it's
in
the
proposal.
Okay,.
A
All
right,
I
think
this
is
done
right.
We
discussed
the
topic,
so
is
there
anything
else.
E
I
maybe
I
would
just
say
I'm
sorry
if
I
have
internet
connection
problems,
but
I
mark
I'm
actually
continue
to
mark
pull
requests
after
verification
with
a
rotten
label
that
we
agreed
before.
So
I
would
appreciate
if
you
go
over
these
pull
requests
and
close
them
or
reopen,
or
to
do
something
with
that,
because
they
stay
for
a
long
time
here
and
actually
contributors
still
not
answering
here.
So
I
think
it's
worth
to
close
them
at
some
point
yeah.
I.
A
Actually,
I
closed
few,
but
thanks
for
reminding
I
think
it's.
We
agreed
that
people
would
review
and
comment
if
you
object
to
closed
apr
and
here's
the
link
in
the
chat
with
all
the
pr's.
That's
supposed
to
be
closed
and
so
yeah,
just
whenever
you
have
time.
Maybe
let's
give
it
another
few
days
and
you
please
comment
if
you
want
to
keep
prs
open,
if
not
we'll
close
them.
E
Yeah
I'm
continuing
here
but
yeah
no
problem.
E
B
On
the
on
this
label,
the
the
rotten
label,
the
idea
is
to
adding
is
to
add
this
label
for
prs
that
require
attention
or
for
prs
that
we
should
close.
This
is
to.
E
Yeah
so,
in
addition
I
would
say
the
strategy
here
that
on
each
pull
request,
I
am
checking
if,
if
it's
really
valuable
and
it
can
be
merged,
I
am
first
of
all
I'm
contact,
contributor
and
asking
if
he
has
some
plans
and
important
to
say
that
a
lot
of
pull
request
was
merged
and
people
answered
on
them
and
yeah,
and
if
I'm
trying
to
contact
person,
then
still
not
answering
for
a
few
days.
I
just
mark
this
pull
request
as
rotten
and
waiting
for
you
to
took
closer
to
take
some
additional
decisions.
I
A
This
special
any
other
topics
from
anyone
all
right.
If
not
then
yeah
we
can
save
you
20
minutes
and
and
a
little
early
today,
cool
thanks
for
taking
the
meeting,
see
you
next
week.