►
From YouTube: Helm Developer Call 20180802
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
Hello,
welcome
to
the
August
2nd
edition
of
the
helm,
public
developer,
call
LV
or
host
Matt
butcher
kicking
off
the
morning
in
our
usual
form,
by
starting
with
announcements
followed
by
the
developers,
core
developers,
giving
stand
ups
for
last
week
in
this
week
and
finally
finishing
with
any
discussion
topics.
So
as
far
as
I
know,
the
only
announcement
for
the
week
is
that
the
RC
2
for
the
current
hell
release
was
released
earlier
this
week,
I
was
still
on
vacation.
A
B
C
Thanks
so
yeah
I've
been
missed
a
few
of
these
meetings,
so
summertime
stuff.
So
the
big
thing
that
for
this
week
is
I,
opened
up
pull
requests
yesterday
for
kubernetes
1.11
support
for
mountain
cedar.
C
No
claws,
okay,
sorry
I
guess
meant
it
so
that
could
use
some
heavy
testing.
Things
in
particular
is
testing
different
kubernetes
versions
with
it
wait
logic
and
upgrades
those
other,
as
if
anybody
has
some
cycles
on
their
vehicle
that
they
could
slide
that
in
be
appreciate
it.
So
we
can
get
that
sluggard
in
for
the
next
release
other
than
that
gonna
be
working
on
I'm
three
stuff
again
working
on
the
SK
mother
I
got
that
big
chunk
out
of
the
way
see.
C
D
So
this
last
week,
I've
been
kind
of
helping
out
with
the
helm
to
just
making
sure
that
the
release
candidates
are
going
through
we're
testing
them
and
kind
of
keeping
track
on
what
bugs
and
what
not
need
to
be
fixed
before
we
go
into
the
next
RC
or
into
the
next
or
into
the
final
release.
We're
trying
to
move
along
the
2.10
release
cycle.
D
So
I
was
testing
that
out
and
then
compared
with
Adam
yesterday
on
the
helm,
to
or
kubernetes
1.11
support,
I'm
just
kind
of
going
through
some
of
the
quest
for
going
through
some
of
the
the
API
changes
in
kubernetes
just
to
make
sure
that
we're
kind
of
like
backwards
compatible
and
certain
things
here
and
there.
So
that
was
good.
What
I
hope
to
get
to
do
for
the
rest
of
the
week
is
just
keep
on
moving
through
RC
2,
2,
RC,
3
I.
D
D
F
A
The
next
couple
of
weeks
we
have
a
deadline
kind
of
looming
on
that,
because
mid-september
is
when
we
need
to
be
all
done
with
the
transition
through
our
kubernetes
2cn
system.
Maroa
and
I've
been
working
on
some
of
the
other
CNCs
migration
stuff
he's
getting
ready
to
take
a
leave
for
a
little
while
so
I'm
trying
to
make
sure
that
nothing
gets
dropped
through
the
cracks
on
that.
But
we're
really
just
down
to
a
couple.
Little
nitpicky
sorts
of
things
to
get
various
accounts
shifted
over.
So
we'll
keep
working
on
thorough.
A
A
G
A
In
tray
or
just
another
card
storage
account
somewhere,
but
nothing
has
been
decided.
That
kind
of
crucial
thing
we're
trying
to
be
able
to
capture
at
this
point
is
metrics.
We
currently
don't
get
very
reliable
metrics
off
our
Google
cloud
storage
account,
partly
because
we
don't
control
it.
So
we
would
like
to
be
able
to
put
it
somewhere
where
we
can
accurately
report
to
the
community.
G
A
G
A
H
For
this
tool
to
make
standing
up
chart
repository
is
a
lot
easier,
especially
as
we
go
to
a
more
distributed
model
and
move
away
from
the
stable
repo.
It's
going
to
become
fairly
more
important,
so
so
I'm
not
quite
ready
to
share
the
pros.
Well
yet,
but
the
kind
of
TL
DR
is.
We
want
to
make
char
repositories
as
easy
as
to
stand
up
as
basically
praying
and
get
repository
and
github.
So
that's
a
goal
and
more
on
that
soon.
Hopefully.
H
D
Yeah,
I
think,
there's
two
of
them
from
michelle,
but
she's
not
present
at
this
time,
but
I
can
talk
about
the
ones
that
I
wanted
to
do,
which
was
for
I
was
talking
about
releasing
helm,
2.0
or
two
10.0
today.
After
doing
that,
but
I
guess
we've
already
kind
of
mini
discussed
that
as
in
like
we
want
to
cut
an
RC
three
before
doing
it
to
ten
zero,
which
makes
sense
and
then
the
other
thing
also
I
just
wanted
to
throw
out
a
proposal
for
a
plug-in
manager
for
helm.
A
Do
we
need
this
sort
out
on
your
first
propose
your
first
discussion
topic:
do
we
need
to
sort
out
the
when
we
want
to
try
and
push
that
do
we
want
to
release
the
current
version
with
or
without
atoms
attached
for
kubernetes
one
mm
without
okay,
so
that
keeps
the
RC
release
train
rolling
ahead
for
that
one
and
then
I
think
we
have
three
or
four
other
small
features
that
would
go
in
with
Adams
patch.
So.
D
C
Since
we're
gonna
have
to
change
some
of
the
stuff
to
be
a
little
easier
for
news
releases
in
the
future
with
the
renée's
client
and
how
we
operate
on
the
object
set,
we
might
be
able
to
figure
out
a
way
to
include
the
apply
logic.
So
then,
if
you
works
like
scaled
points
or
something
like
that,
that
would
be
captured
in
it
as
well,
so
I've
been
working
on
a
prototype
I'm
done
to
the
point
yet
of.
D
E
Little
little
thing
I
either
bring
it
up
all
the
time
or
can't
remember
to
forget,
but
so
currently
in
support
queue.
We
will
label
something
question
and
support
cuz.
We
can't
or
don't
know
that
it's
a
bug
so
I'm
gonna
come
and
say
it's
a
bug.
We
don't
necessarily
know
I
think
the
common
things
you
kind
of
say
well.
This
is
just
support
question
and
if
you
verify
its
bug,
we'll
call
it
a
bug,
but
maybe
we
should
just
add
the
bug
label
and
then
make
another
label
for
verified.
E
So
we
can
clearly
see
what
people
think
are
bugs
and
then,
or
at
least
what
we
you
know.
We
know
it
might
be
a
bug.
Anyway.
We
know
it's
not
just
a
support
question
throw
a
bug
on
there
and
then
you
know
and
actually
get
it
verified
with
a
verified
label
on
there.
They
can
be
easier
to
go
and
find
the
ones
that
are
potential
bugs
because
yeah
right
now,
it's
a
bunch
of
support
questions
is
what
it
looks
like
and
it's
hard
to
tell
what
are
actually
support.
Just
purely
support
questions
versus
things.
D
If
I
propose
a
different
solution,
as
in
the
negative
of
a
it's
been
verified,
as
in
it's
been
unverified,
because
if
we
do,
if
we
say
that
we
start
adding
a
label
that
says
it's
verified
now,
what
that
basically
means
is
that
all
of
the
existing
bugs
that
are
all
the
existing
tickets
that
are
out
there
with
the
label
bug
by
the
definition
of
it's
not
being
or
not
having
to
label
verified,
then
they're.
Basically,
if
the
convention
is
we
go
six
months
later,
those
things
look
like
they
haven't
been
verified
where
they
have.
D
So
what
if
we
added
a
label
that
just
says
needs
more
verification
or
needs
testing,
or
something
like
that,
just
to
make
sure
so
that
when
we're
triaging
these
issues,
then
we
can
say
this
may
be
a
bug,
but
it
needs
more
information
before
we
actually
confirm
it's
a
bug
and
then
we
can
remove
that
label
and
say
it's
a
confirmed,
bug
or
something
like
that
confirmed
yeah,
unconfirmed
or
yeah,
something
along
those
lines.
Any.
E
A
A
E
F
E
F
F
A
A
We're
still
largely
focused
on
refactoring
code
base
and
doing
some
of
this
small
API
breaking
changes
that
we
have
been
wanting
to
make
all
longer
that
issues
have
asked
for
for
a
long
time
and
we'll
keep
doing
that
until
we
feel
like
that's
satisfactory
right.
We
want
a
good,
solid
foundation
and
also
the
proposals.
They're
still
actually
work
happening
on
Humphrey
proposals.
A
Basically,
DSL
for
chart
ability,
so
those
are
kind
of
the
main
things
that
are
happening
right
now.
Last
week
we
spent
some
time
going
through
a
proposal
is
probably
next
week.
Err
that
we
have
your
we'll
spend
some
time
going
through
some
of
those
proposals
just
to
make
sure
that
we
keep
it
good
rolling
cadence
on
those.
A
A
F
F
A
H
Question
when
I
chat
with
him
I
think
he
I
think
he
hasn't.
He
hasn't
been
that
much
in
the
loop
of
it
and
I
think
he
plans
to
look
at
the
proposal
and
stuff.
But
what
he's
I
think
what
he's
demoing-
and
he
did
a
blog
post
on
this-
is
basically
using
helm,
template
and
then
piping
it
to
customize
to
make
kind
of
more
I
guess
more
interesting
customizations.
That
kind
of
don't
really
exist
in
it
in
a
value,
so
yeah
more
file
or
shouldn't
exist
in
about
his
Miami
file.
H
So
he
wants
to
get
people
move
people
away
from
you
know:
moving
moving
kind
of
kubernetes
constructs
up
into
the
values
out,
ya,
know
file
and
instead
doing
that
somewhere
else
and
I
think
that
ties
in
pretty
well
to
helm
free
with.
If
Humphrey
is
just
kind
of
client
side
thing,
and
you
could
pipe
that
to
something
else,
that's
a
good
to
go
and
do
those
common
customizations
that
could
be
a
nice,
a
nice
model
or
even
using
the
the
Lua
scripting.