►
From YouTube: Charts Chat 20180109
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
Unfortunately,
I
feel
like
I
put
way
too
many
of
them
on
there
this
time
again,
so
I
guess
I
could
rattle
through
them
if
you've
got
them
up
the
first
one
up
front
rattling
through
these
github
IDs,
as
the
maintainer
names
as
I
was
reviewing
charts
the
reason
this
came
up.
Was
there
a
number
of
companies
who
wanted
to
put
their
company
name
in
there
or
their
company
orc
in
there
and
so
I'm?
Looking
at
this
going,
how
should
we
do
this?
A
Should
we
expect
real
people,
especially
since
we're
now
starting
to
go
towards
owner
files,
or
was
there
a
reason
behind
it?
I
understand
the
contacting
thing,
because
there's
a
couple
of
owners
who
don't
have
their
names
or
their
listed
email
addresses
on
their
github
org
ID,
so
I
have
no
idea
who
to
carbon-copy
into
a
PR
review
or
something
so
I
totally
understand
that.
But
I
was
just
wondering
any
thoughts
on
this
I
think.
B
I
think
it
would
be
an
additional
field
if
you
want
a
company
and
have
a
company
contact
or
something
like
that,
because
that
we
need
the
people
right
like
we
need
to
be
able
to
yeah.
We
at
least
need
person
to
contact
I.
Think
if
we
have
a
company
in
there
like,
let's
say,
there's
a
critical
something.
Rather,
how
did
who
are
we
responding
to
on
the
other
side
of
that?
If
they
want
to
do
an
alias
I?
Guess:
that's:
okay,
but
not.
D
I
think
we
had
this
issue
with
the
Nam
epoch,
where
you
know
we
kind
of
used
bit
Nami
pot
and
containers
at
phenomenal.
Calm
email
address
for
full
of
all
of
the
bit
Nami
charts,
but
then
made
it
difficult
to
actually
find
out.
You
know
who
is
actually
who
happened.
Nami
was
actually
maintaining
and
I.
Think
now
that
we
have
the
owners
file.
That's
obviously
more
clear
and
I.
I
also
think
that
the
chart
yeah
more,
should
represent.
D
What's
in
the
owners,
file
and
I
agree
with
with
you
know,
Vic's
remark
about
just
having
actual
people
on
there
and
you
know
that
changes
we
can.
We
can
update
the
charts
and
change
those
people
as
well,
but
in
terms
of
just
adding
a
company.
If
you
know,
if
you
do
want
a
company
there
I
think
given
the
github
ID
can
also
be
an
organization.
Id
I
think
it's
certainly
fine.
If
someone's
to
put
a
company
github
ID
inside
the
chart,
memo
I
don't
well.
B
A
File,
in
one
case
it
was
they
said.
Well,
it
could
be
anybody
at
this
company
who
goes
and
does
it
so
we
want
to
have
the
company
name
there
and
then
so
it's
somebody
who's
associated
with
that
company
can
come
in
and
do
it
for
us
and
Mayra
pushback
was
well.
You
know
we
actually
want
people
to
the
Carbon
Copy
and
on
PRS
and
a
polling
to
review
these
things.
It's
we
really
need
names,
but
this
was
expressed
by
others,
and
so
that's
the
reason.
I
thought
I'd
bring
it
up.
The.
E
B
B
B
F
A
Now
well
and
that's
how
it
is
in
a
lot
of
places
right
if
you
go
to
homebrew
packages
right
for
mac,
you're
gonna
find
lots
of
people
who
don't
do
the
app
do
the
packaging
for
the
particular
app
right.
This
is
not
an
unusual
thing
to
have.
You
know,
though,
the
folks
who
create
curl
don't
go
to
every
Linux
distro
and
deal
with
the
packaging,
and
so
it's
it's
not
unusual
to
have
other
people
do
that
and
we
do
see
a
fair
amount
of
that
I
guess
in
the
church,
repo,
okay.
C
B
F
Know
I
was
also
thinking
that
we
we
don't
really
have
any
line
stating
anywhere
what
our
policy
is
for
revoking
from
an
ownership
right.
We
do
have
a
don't.
We
share
I
mean
I'm,
sorry
that
I
didn't
look
this
up
for
Isis,
okay,
but
it's
r1
to
call
together.
Don't
we
share
just
the
code
of
conduct
and
so
if
something
like
that
gets
broken
at
least
that
there's
that,
but
then
what,
if
someone's
just
not
I
someone
screws,
only
we've
got
someone
going,
rogue
I
think
what
would
our
policy
be?
F
F
D
One
of
the
reasons
we
haven't
had
to
deal
with
this
so
far
is
because
you
know
only
we
could
merge,
Piazzi,
viously
or
need
to
chop,
maintain
us
could
merge,
be
ours,
and
now
that
we're
opening
the
ability
for
more
people
to
emerge
be
ours.
You
know
as
long
as
they're
in
the
kubernetes
org,
there
is
obviously
more
potential
for
this
problem
right
yeah,
it's
a
good
question.
A
I
do
as
far
as
I
know.
It's
only
come
up
a
couple
of
times
and
the
steering
committee
is
tasked
with
going
and
figuring
it
out.
What
it
is
is,
there's
a
code
of
conduct,
everybody
can
go,
read
it
and
if
there's
a
violation
of
that
there
is
one
or
more
email
addresses
you
can
contact
people
privately
to
start
chasing
down
how
it
happens
and
most
of
the
stuff
is
handled
very
quietly,
offline.
I,
don't
think
it's
ever
been.
Somebody
who
I
don't
know
of
any
case
where
somebody
has
had
to
have
their
stuff
revoked.
A
B
A
Okay,
the
next
one
was
actually
handling
the
deprecation
of
charts.
There
are
a
few
charts
that
are
deprecated,
and
this
came
up
because
there
was
a
PR
to
go
change.
The
vanity
URL
to
k8s,
dot,
G
cRIO
from
GC
art,
io,
/,
Google
or
whatever
containers
or
the
kubernetes
containers
were
put,
and
that
was
changing
to
vanity
URL
and
it
came
to
an
issue
you
can
link
to
3167
and
one
of
the
things
came
up
was.
A
This
was
for
a
deprecated
chart
in
one
case,
and
then
we
noticed
we
just
put
deprecated
on
it
and
then
admin
and
I
got
into
a
discussion
was:
do
we
move
deprecated
into
a
different
directory?
What
do
we
do
these
things?
Okay,
so
maybe
we
just
delete
deprecated
or
this
deprecated
chart.
Maybe
somebody
comes
into
the
new
chart
with
the
same
name
and
they
start
over
on
versioning
and
next
thing.
A
You
know
they
start
overwriting
some
of
the
versions
but
on
others
that
are
in
our
bucket
and
we've
run
into
problems
right
and
so
the
proposal
that
I
had
and
I
you
know
was
spitballing
was
we
leave
it?
We
put
deprecated
on
it.
Maybe
we
put
a
comment
in
there
that
says
this
chart
is
deprecated
to
undec
eight
it.
You
know
you
need
to
bump
a
major
version
or
start
supporting
and
remove
the
deprecated,
and
you
need
new
maintainer
here
right.
A
We
put
something
in
there
like
a
comment
and
then
we
have
see
I
say
if
it's
got
deprecated
always
fail,
see
I
write
because
you
actually
have
to
say
we
run
deprecating
it
for
something
like
this
happened,
but
then
it
leaves
it
in
the
repo,
but
it
tells
everybody
that
it's
deprecated,
but
then,
if
somebody
comes
along
with
another
chart
of
the
same
name,
we're
not
gonna
run
into
numbering
issues.
That
was
that
was
my
spit
ball
proposal.
Does
anybody
have
a
better
idea
or
something
and
how
we
can
handle
this.
B
Good
first
time,
I
mean
what
we
want
is
for
people
to
know
right,
that's
the
main,
the
main
thing
and
then
to
be
able
to
talk
with
on
and
off,
which
I
think
that's
enough
red
light,
green
light
situation
that
explicitly
we
know
who
turned
it
back
on
and
who
turned
it
off
and
how
long
it's
been
off
and
all
that
kind
of
stuff.
So
I
think
that's
that's.
B
D
D
Kind
of
rel
dapper
guidelines
for
this
for
how
we
deal
with
these
I
I
added
a
comment
about
deleting
the
charts
from
the
repo,
but
my
Matt,
you
know
kind
of
rightly
raised
the
issue
that
if
we
delete
it,
how
do
we
know
what
version
to
restore
it
to
if
we
want
to
restore
one
with
a
city?
Name
and
that's
that's
actually
a
big
reason-
I
think
to
just
keeping
it
in
there.
D
So
we
you
know
we
have
that
reference
and
it's
easy
to
find,
and
also,
if
someone's,
just
going
through
the
repo-
and
you
know
looking
through
the
lists
of
charts,
they
can
they
get
a
message
and
read
me
or
something:
no
good,
so
yeah
I'm.
Definitely
for
that
and
see
I
know
that
we
have
liability
through
the
circle.
Ci
stuff,
I,
think
that
makes
it
all
easier.
Yeah,
I.
A
B
B
D
Also,
usually,
do
is
update
the
notes,
dot
txt,
as
well
as
the
readme
in
the
trot
repo,
so
people
actually
see
it,
cuz
helm
doesn't
actually
do
we
treat
the
deprecated
flag
with
anything
special
it
just
it's
just
it's
there
in
aspect,
but
helm
doesn't
actually
use
it.
It
was
actually
added
for
things
like
monocular
so
that
they
can
filter
them
out.
That
makes
sense.
I.
D
Actually,
speaking
of
monocular,
when
I
was
working
on
revamping
some
of
the
chart,
syncing
logic
I
noticed
that
there's
a
lot
of
charts
that
haven't
been
marked
as
deprecated.
It
should
be,
in
particular
things
that
have
moved
from
inky
pink
you
beta
to
stable
should
be
marked
as
deprecated
in
incubation
or
not
so,
I
think
we
need
to
go
back
and
review
what
other.
F
D
D
A
Okay,
all
right
we
get
on
this
one
then,
on
the
deprecation,
yeah
awesome.
The
next
one
is
it's
kind
of
a
funny
one.
Somebody
brought
up
in
an
issue,
that's
issue
3000,
so
check
this
out.
If
you
had
to
say
greater
than
or
less
than
right
is
version
1.0
greater
than
or
less
than
version
9,
so
write,
10
less
than
9
string
comparison,
which
is
greater
9,
because
9
is
a
greater
number
than
1.
The
first
number
you
compare,
and
so
if.
A
At
this
issue,
we've
got
in
our
files.
You
know
you
know.
If
you
know
the
capabilities,
major
1
and
capabilities
minor
is
greater
than
7
right.
Well
as
soon
as
we
hit
this
one,
oh
release,
that's
about
to
come
out
and
kubernetes
1.10
1.10
is
less
than
dot
9,
because
we're
doing
string
comparisons
here,
I
didn't
even
think
about
it,
but
it
totally
is
true.
Helm
does
have
a
way
of
handling
this
because
it
handles
ember
home
can
do
semantic
versions,
and
so
we're
gonna
need
to
rewrite
all
of
these
things
with
semantic
versions.
How.
A
Spring
spring
can
do
some
verse
off
content,
so
you
can
say,
is
it?
You
know
these
versions
and
do
the
conversion
but
yeah
it's.
We
have
to
go
in
and
fix
this
or
our
stuffs.
Gonna
start
failing,
but
one
died.
10
is
released
and
so
I
wanted
to
bring
it
up
here.
So
you
all
would
know.
That's
it's
greater
than
and
less
than
an
equal
to
major
and
minor
is
gonna,
have
to
change.
I'll
start
I
can
file
PRS
to
start
changing
this.
Unless
somebody
else
wants
to
do.
A
A
A
And
Sprague
has
the
ability
to
do
the
version
comparisons
for
us,
rather
than
using
the
greater
than
less
than
equal
to
that's
built
into
the
code.
Template
that's
string
based,
so
we've
got
the
parts
there.
We
just
need
to
rewrite
it
and
I
wanted
to
bring
it
up,
because
I
always
thought
this
was
great
and
I
had
no
idea.
D
A
A
If
you
look
at
the
values
that
you
mo
file,
he
removed
them
and
what
he
did
instead
was
he
created
a
helper
function
and
helpers
dot
TPL
that
grabs
the
it
grabs
the
default,
the
chart
da
version
or
what's
in
the
image
version.
So
if
the
image
version
is
empty,
it
uses
the
shark
app
version
as
the
tag
version.
A
And
I
thought
this
was
rather
neat
and
so
I
wanted
to
know
what
people
thought
of
that
at
first
I
commented.
You
know
what
you
remove
these,
we
need
the
version
in
there.
We
don't
just
want
to
grab
latest
and
I'm
like
then.
He
pointed
that
out
and
and
I
had
glossed
over
it
and
I
I'm
curious.
What
people
think
of
this
pattern.
B
Implicit
versus
explicit
I,
like
the
explicit
you
know,
here's
the
thing
that
we're
doing
and
here's
the
change
and
and
that
kind
of
thing
if
this
is
how
we
did
that
across
the
board.
I
think
it'd
be
a
lot
easier
to
tenet
again,
it's
the
right
way
to
do
it,
but
it's
yeah
since
there's
differences
among
the
charts,
I
prefer
I
prefer
being
explicit,
but
it's
neat
maybe.
E
D
Because
you
know
most
people,
you
know
when
they
can
configure
the
tag.
We'll
look
at
this.
We
look
at
about
you
so
yeah
more,
but
a
lot
of
people
don't
actually
care
if
they're,
just
installing
the
chart
directly
so
I
think
I
think
it's
fine
to
actually
use
admin.
I
think
this
is
actually
a
good
use
case
for
it.
So
what.
A
F
I
see
I
I
misinterpreted
that
there
we
go
yeah.
D
F
D
E
F
Think
that
the
one
plus
here
then
just
to
sort
of
take
back
what
I
was
saying
before
is
that
it's
just
more
dry
yeah
you,
you
would
have
less
mismatch
as
I
bet.
If
we
look
through
our
charts,
we
might
find
a
couple
of
mismatches
now,
I,
don't
know
we.
We
may
be
totally
up
just
today
right
now.
You
know
where
someone
doesn't
update
the
chart
version
or
someone.
B
D
Know
agreed
if
we
do
want
to
add
this
I
think
more
documentation
needs
to
be
there
to
actually
explain
what's
happening,
yeah,
where
I
think
this
would
be
useful
is
I,
think
this
sure
I
can't
really
think
of
it
right
now,
but
there
is
a
chart
that
I've
seen
that
uses
the
same
tag
version
for
multiple
components
into
values.
What
yeah
no-
and
this
would
be
really
a
much
better
way
of
doing
that.
You
know,
as
people
have
said
like
if
for
this
one
row
you're
just
putting
into
play.
A
A
E
C
F
F
Or
was
it
I'm,
not
sure
who
said
earlier
that
they
prefer
the
explicit
I
would
say
as
a
as
I
as
a
rule
explicitly
wins
just
because
if
the
goal
is
to
save
energy,
you
waste
more
time
with
the
confusion
of
over
with
the
extra
layers
of
complexity,
you
know
if
they
win,
you
a
lot
great.
If
they
don't
win
you
a
lot,
just
don't
do
it.
You
know.
F
Same
with
NGO
functions,
for
example,
you
can
overload
them,
I
mean
you
can't
overload
them,
but
you
can,
you
know,
say:
pass
a
struct
in
there's
a
sort
of
way
of
cheating
overloading,
but
that's
frowned
upon
because
you
lose
a
lot
of
other.
You
know
you
lose,
you
know,
type
checks
and
you
lose
other
things,
but
you
you
gain
flexibility.
So
the
only
projects
that
do
things
like
that
or
when
there's
a
massive
game,
you
know
I,
don't
know
it
might
be
nice
to
follow.
A
Okay,
would
you
like
me
to
go
back
then
and
say
so?
What
I'm
hearing
is?
Is
it's
a
neat
pattern
in
complex
situations?
We
could
see
somebody
using
it,
but
for
something
fairly
straightforward,
like
this
we'd
like
to
keep
it
more
explicit,
because
that'll
be
easier
for
people
to
follow,
even
if
it's
a
little
bit
of
extra
work
to
update
the
tag
in
the
different
places,
so
we're
moving
the
complexity
more
onto
the
authors
and
the
consumers.
E
A
All
right
and
then
I
had
another
one,
then,
to
move
on
to
the
next
one.
I'll
follow
up
on
that
one.
There
were
in
the
incubator
web
page
test
charts
that
there's
a
pull
request
for
there
were
calls
out
to
ec2
because
they
actually
have
VM
images
and
ec2
as
well
that
you
can
give
your
ec2
creds
to
and
interact
with
and
I
wasn't
aware
of
us
doing
charts.
A
A
A
You
can
run
yourself,
but
they
do
have
the
ability
to
pass
an
ec2
creds
to
via
the
PR
and
I
didn't
I,
actually,
don't
know
how
to
handle
those
kinds
of
cases
where
it's
just
one
public
cloud
provider
in
that
or
if
we
say
no,
we
don't
want
to
do
this.
We
want
to
make
them
generic
sure
you
can
run
them
anywhere.
Don't
have
passed
that
in
or
it's
okay,
even
if
we're
running
on
your
own,
because
somebody
has
stopped
and
say:
oh
I,
really
know
I'm
doing.
A
A
End
up
having
the
controller
and
then
it'll
probably
be
most
effective,
I
think
if
you're
running
all
of
it
in
Amazon.
So
if
you're
running
your
kubernetes
cluster
on
Amazon,
which
a
whole
lot
of
people
do
now,
then
you'll
get
the
benefit
of
just
working
with
another
thing
and
then
you're
managing
the
part
that
manages
everything
in
it.
D
F
A
Well,
that's
a
separate
one.
They
do
have
a
PR
for
the
agent
to
run
in
the
cluster
as
well,
so
you
can
have
the
agent
run
as
a
VM
and
ec2
or
something
in
the
cluster,
and
when
it
is
an
ec2,
you
pass
in
your
ec2
creds
in
your
vial.
You
GMO
file
and
I'm.
Just
wondering
is
that
is
that
kosher
like
I,
have
no
idea
I,
don't
think
I've
encountered
or
thought
about
this
before
wouldn't.
F
A
F
You
thought
I
hear
what
you're
saying
does
it
seem
to
cloud
providers
specific
yeah?
We
have
other
there's
another.
There
was
another
and
I'm.
Actually
there
are
several
right,
but
there
were
there
was
one
I'm
kind
of
losing
my
memory
thread
about
it,
but
there
was
a
dashboard
for
a
Microsoft
Project
that
can
only
be
run.
It
really
only
makes
sense
to
run
on
measure.
D
F
D
Yeah
sure
about
UK,
that's
I
think
that's,
as
could
be
seen
as
like
a
you
know,
more
of
a
specific
case
where
it's
actually
really
useful
in
kubernetes,
it's
I'm,
not
so
sure
about
is
well.
Is
it
possible
to
run
webpage
yes
on
other
clouds
like?
Could
someone
come
along
in
the
future
and
add
support
for
you
know,
gkv
or
probably.
F
A
D
D
Because
the
other
thing
you
could
draw
a
parallel
to
is
something
like
you
know:
the
data
doctor
talk,
for
example,
which
is
obviously
very
specific.
They
data
dog
that
allows
you
to
connect
it
to
you,
go
keep
Nettie's
cluster,
which
is
very
useful,
but
you
know
it
does
it's
not
gonna
work
with
New
Relic.
You
need
to
use
the
New
Relic
tripod
for
that,
so
that
I
don't
know
if
it
falls
into
that
category.
The
other
option.
A
C
D
D
A
How
about
I
ask
this
because
I
you
know
I've,
you
have
dropped
a
link
into
the
pull
request.
Can
y'all
go,
take
a
look
at
it
and
just
see
what
you
think
on
this
I
can
deal
with
the
application
specific
parts
on
it.
If
nobody
else
wants
to
I've
used,
webpagetest
I've
never
deployed
it,
but
I'm
familiar
enough
with
using
it.
I
think
you
can
run
it
in
different
places.
I
can
chase
that
down,
but
just
the
style
in
the
way
that
they're
using
ec2
just
take
a
look
at
it.
Give
feedback
I,
not
sure.
A
I
was
ok
with
it,
but
I
thought
I
would
ask
because
I
try
to
be
fairly
cloud
provider
neutral
and
everything
I
do,
especially
since
so
many
people
like
to
run
things
on
premise,
yeah
and
I'll.
Tell
you
the
the
author
of
these
charts
is
really
I
mean
he
is
very
flexible
and
willing
to
work
like
upstream.
Isn't
tagging
things
yet
at
this
point,
and
so
he's
gonna
carry
tags
separately
when
he
no
releases
are
happening
or
when
he
knows
it's
good
enough,
where
he
could
tag
it.
F
D
Honestly,
I
have
no
comms
about
even
having
this
in
stable.
We
wanted
to.
You
know
if,
where
I
see
it
like,
if
it
is
something
that
can
be,
is
currently
cloud
specific,
it
should
be
mentioned
in
the
readme
and
if
someone's
gonna
implement
support
for
another
cloud
in
the
future
or
some
other
way
of
deploying
it,
then
they
can
go
and
you
know
contribute
to
it.
I
don't
think
I
should
necessarily
stop
it
from
being
added.
Initially
yeah.
B
F
The
main
thing
that
I
and
I'll
just
make
a
comment
on
this
right
now,
but
the
main
thing
that
I
would
just
recommend
is
that
they
not
we
not
ask
we
not
even
recommend
people,
add
their
their
key
and
secret
into
values.
For
a
chart,
I
mean
that's
something
that
we
have
other
kubernetes
objects
for.
That
I
mean
like.
If
we
can,
we
can
set
those
as
environment
variables
or
something
thing
based
on
a
secret.
You
know
we
do
that
elsewhere
and
certain
traits,
except
you,
take
great
a
secret
if
there's
something
sensitive
and.
A
F
D
A
But
you
know
we
can
help
them
get
the
first
one
in
and
give
them
recommendations
for
follow-ups
as
well.
We
don't
want
to
block
and
tell
they've
implemented
every
possible
feature
if
somebody's
willing
to
continue
to
work
on
it
at
least
I,
don't
think
so.
I'd
like
to
really
kind
of
get
them
motivated
and
being
interested
in
contributing,
rather
than
us,
being
a
pain
in
the
neck
and
then
walking
away
cursing
your
name's.
A
A
C
Name,
that's
how
it
currently
works,
and
this
was
I
think
invented
by
Mattel,
and
what
is
documented
is
that
the
service
account
name
should
Norvin
should
actually
be
used
instead
of
using
the
phone
and
template
for
the
service
account.
So
what
the
question
is?
Should
we
change
it
I?
Should
we
keep
it
this
way,
I
mean
generally,
we
have
the
best
practice
of
using
the
phone
and
template
for
resource,
and
the
questioners
should
be
really
do
this
differently
for
service.
It
counts
and
yeah.
C
There's
more
in
this
discussion
and
Kenzie
also
contributed
because
App
Engine
and
generally
I
would
say
we
should
not
make
it
too
complicated
and
the
way
it
is
now
I
think
it's
fairly
easy
to
understand.
It
basically
allows
us
to
manage
our
big
resources
within
the
chart
or
externally.
So
this
is
basically
more
or
less
all
we
need
I
would
say.
C
Can
this
argument
was
companies
may
require
special
names
for
for
specific
policies
or
names
for
service
accounts
which
would
not
fit
into
this
I
would
tend
to
say.
If
a
company
requires
special
names,
they
would
probably
also
require
special
name
swaddle
resource
types
or
may
not
even
allow
to
use
public
earth,
so
they
would
have
to
roll
around
anyways
to
not
want
to
change
it,
as
we
do
isn't
in
deductive
documentation,
but
I'm
open
to
any
discussion.
Sir
I
know
Kenzie
has
a
different
opinion.
If
you
want
to
champion
yeah.
E
Well,
I
was
gonna,
say
because
I
know
a
lot
of
the
time.
There
is
actually
special
naming
around
specifically
around
credentials,
and
so
a
service
account
is
basically
a
credential,
and
so
where
that
fits
into
a
lot
of
security
policy
that
already
exists
is
you
would
treat
it
like.
A
user
name
and
user
names
often
have
very
strict
conventions
around
how
their
names
so
or
you
know,
service
account
names
around
how
they're
named
so
I
want
people
to
be
able
to
always
say
this.
E
Account
should
be
named
this
and
have
that
never
be
ignored
if
specified,
but
at
the
same
time
the
documentation
right
now
implies
that
you're
required
to
specify
it,
and
the
point
was
made
that
you
shouldn't
have
to
specify
it.
If
you
don't
want
to,
and
if
you
don't,
then
it
should
just
be
auto-generate,
which
I
agree
with.
D
One
one
thing
I
would
say
about
this
is
that
the
name
is
confusing
if
it's
called
service
class
name,
then,
to
me
that
sounds
like
it's
used
everywhere.
We
may
have
to
add
some
a
comment
saying
you
know
this
is
only
used
if
our,
but
not
enabled
is,
is
oh
sorry,
our
backup
created
is
great,
is
false.
E
C
C
D
Guess
the
question
is:
do
you
expect
that
to
create
that
service
account
with
the
name
that
you
specified
and
then
why
would
you
want
to
specify
a
name
for
a
service
account
if
you
don't
want
to
create
it
for
you,
you
know
it's
a
reason
for
you
to
want
that.
I
pasted
a
link
to
the
and
all
existing
claims
that
I
was
talking
about
this
pattern
that
we
use
in
the
Drupal
chart,
and
it
says
there
requires
persistence,
enabled
to
be
true
right.
Actually,
I.
Guess
that
doesn't
matter.
F
C
C
F
F
B
D
D
C
You're
saying
companies
have
special
policies
regarding
security
and
so
on.
In
this
case,
they
would
certainly
not
allowed
to
manage
avec
within
this
in
your
child
anyways,
so
that
we
do
it
externally
index
just
specify
the
service
account
line,
and
we
can
keep
the
best
practice
as
it
is
no
problem
at
all.
Yeah.
E
I
mean
well,
but
then
you're
saying
to
people.
If
you
want
to
be
able
to
specify
this,
you
can't
do
this
as
a
one-stage
process
anymore.
You
have
to
externally
manage
all
this
other
stuff
you're
saying
you
have
to
there's
not
a
there
right
now.
If
we
say
you
can
specify
the
service
account
name
and
have
this
create
the
Arabic
resources
for
you,
we're
saying
you
can
do
it
that
way
or
you
can
still
externally
manage
it.
F
F
E
F
E
D
A
A
Our
anytime,
our
back,
create,
is
false.
We
use
the
specified
service
name
and,
if
one
isn't
specified,
we
Auto
generate
one
right,
and
so
somebody
installs
ten
different
charts
that
use
that
and
have
our
back
creative
Falls
will
either
use
the
name
they
specify
or
Auto,
create
in
all
cases
right.
Yes,
so
they
could
end
up
if
they
don't
specify
ten
different
service
account
names
right.
Yes,.
A
A
A
E
F
Through
this,
though,
it
does
seem
like
this
falls
into
that
earlier
category
we
talked
about
before
work,
I've
done
about
a
a
space,
that's
in
need
of
best
practices,
because
I
just
kind
of
did
a
you
know,
grab
through
the
codebase,
and
there
are
a
number
of
charts
to
create
that
great
service
counts.
It
would.
F
F
E
E
B
F
F
D
Okay,
I
think
we'll
see
how
this
list
issue
plays
out.
I.
Think,
like
you,
know,
everyone
who
has
opinions
on
this
definitely
take
a
look
at
3:20
and
we'll
see
how
that
plays
out
and
make
a
decision
on
what
we
want
to
do
with
the
best
practices
there.
It's
also
a
question
of
you
know:
Matt
seems
to
mention
changing
pillar
to
do
something
and
I
don't
know
what
is
what
exactly
he
has
in
mind
for
that.
So
I
don't
know
if
this
applies
to
charts
or
whatever
it's
something
that
needs
to
happen
in
helm.
A
B
F
A
So
Renard,
if
you
could
make
pointers
in
that
issue
with
your
stuff
in
there
as
well
of
the
discussion
just
to
kind
of
fill
out
so
butcher
and
whoever
else
gets
involved
in
it,
we
can
have
the
full
on
on
all
sides.
At
least
cross-linked
in
that
would
be
useful
and
anybody
else
who's
got
those
things
just
throw
them
on
this
issue.
So
we
can
have
a
good
conversation
here
and
a
good
point.
When
somebody
comes
back
to
and
says
you
did
it
wrong,
we
can
say,
but
we
talked
it
out.
A
E
A
A
F
A
Good
yeah
I
was
gonna,
say,
do
you
think?
Can
everybody
go
like
go
out
and
review
some
charts,
because
that
would
be
awesome
right
in
fact,
there's
some
easy
ones
in
there
that
have
already
been
okay
to
test
they're
easy.
If
people
want
to
go
Mirjam
and
then
the
other
one
is
there's
an
aetherium
chart,
does
anybody
know
aetherium.
E
A
A
E
A
It's
long
been
a
problem
in
public
clouds
where
people
go
get
those
free
accounts
and
then
burn
them
to
mine
crypto
and
they
go
get
all
these.
You
know
free
little
card
things
like
Asher
was
handing
him
out
at
one
point,
and
so
you
get
these
things
and
then
you
go
use
it.
You
burn
the
free
resources,
real
quick
to
mine
crypto.
In
fact,
a
lot
of
people.
A
There
have
been
a
number
of
cases
where
people
have
had
their
their
cloud
accounts
hacked
so
that
other
people
will
burn
crypto
on
them
and
make
money
off
of
it
by
hacking
into
somebody's
cloud
accounts
that
isn't
a
more
common
occurrence
than
you'd
think
yeah.
So
yeah
we
do.
We
do
and
that's
why
I
saw
this
on
I'm
being
like,
like
I'm,
not
OK,
to
test
this
thing
until
I'm,
absolutely
sure.
D
C
C
A
D
A
D
Be
the
easiest
option
is
to
not
have
to
make
sure
they
don't
have
a
default
account
and,
due
to
kind
of
you
know,
minecrafting
where
you
have
to
certainly
learn
to
true
to
be
able
to
actually
deploy
giving
and
that's
you
know
that's
work
quite
well
for
Minecraft.
That
would
definitely
help
you
yeah.
A
A
D
D
E
C
A
A
Yeah
they,
if
anybody
didn't
realize
it,
they
have
had
some
choice,
words
to
say
about
the
times
of
our
meetings
being
all
around
Pacific
time
and
then
the
other
side
of
the
world
from
them.
So
yeah
all
right.
Well
we're
a
minute
over
thanks
everyone.
This
is
great.
We've
been
consistent
with
these
meetings
and
had
great
discussion
thanks
everyone.
This
is
fantastic,
I
love
it.
This
one
was
good
yeah
cool.