►
From YouTube: Helm Developer Call 20171214
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
B
B
So
we
can
have
some
nice
talks
about
him
for
announcements
this
week,
not
any
that
I'm
aware
of
we're
not
planning
on
doing
a
2.8
until
we
get
there's
the
kubernetes
1.9
PR
out
there,
so
that
just
needs
the
view
and
I
guess
that's
when
the
committee
of
1.9
is
actually
released.
So
we
can
get
on
that
and
circle
efforts
on
that.
And
then
you
have
a
helm,
dev
summit
CFPs.
If
you
still
have
them,
have
we
extended
the
deadline
until
after
the
holidays
were
we
need.
B
C
I've
done
a
little
in
here
me:
I've
done
just
a
little
bit
of
issue
triage
and
stuff.
Coming
back,
I'm
coming
back,
I'm
gonna
get
to
do
more
soon.
After
the
new
year,
it's
gonna
be
great.
I've
been
pulled
away
to
do
some
other
stuff,
though
yeah
it's
a
little
bit
issue
maintenance
and
stuff.
There's
a
couple
things
I'm
looking
into
doing
a
that's
it
for
me.
I
won't
hear
about
cute
when
this
is
all
over
there.
Somebody,
gentlemen,
setting
it
to
go.
Oh
sorry,
Michelle
doing
connects.
A
A
Not
like
image
but
automatically
deployment
they're,
definitely
two
different
people,
some
in
say,
gaps
and
then
tent
outside
ups.
They
got
I,
say
House
meeting
so
yeah.
That
was
interesting,
really
good
discussion,
I
didn't
do
anything
code-wise
last
week
took
on
stuff
and
I
have
several
items
that
I
wouldn't
discuss
at
the
end
of
this.
At
our
wondered
around
home
summit
the
you
know
mailing
list
for
security
notifications.
A
A
D
Oh
okay,
I
wondered
that
could've
been
anyone.
Unfortunately,
I
don't
have
much
to
say
here.
As
far
as
the
columns
compliments
go,
I
talked
to
a
lot
of
people
about
Helmut
kook
on
and
came
back
again
to
work
here
with
people
looking
to
use
hell
more
broadly
but
again,
still
concerned
about
the
whole
security
situation
and
I
hope
that
we
can
talk
about
that
or
in
the
time
after
stand.
Ups
with
some
things
that
I
heard
at
Keuka
I
think
we're
interesting.
D
E
Sure
I
worked
on
what
was
the
status
Helms
status
can
now
output
llamo
in
JSON,
thanks
Matt
Fisher,
for
reviewing
that
it
kind
of
gives
a
pattern
that
can
be
used
on
other
commands.
I
wanted
to
look
at,
although
it's
gonna
be
a
little
bit
more
difficult,
passing
in
custom
user
agents
into
helm,
so
I
could
do
helm
fetch
and
set
a
user
agent
so
I'm
using
like
the
chart
CI
system,
that's
my
specific
use
case.
I
can
say:
hey
it's
charge,
CI
it
can.
Nor
me
when
we're
getting
metrics
from
next
site.
E
You
know
on
the
chart
system,
we
can
look
at
user
agents
to
see
where
all
the
downloads
and
things
are
coming
from
and
we
want
to
grab
some
stuff
there,
but
it
turns
out
that
that's
a
little
harder
with
some
refactoring
to
make
that
happen
over
on
the
charts
repo
I'll
talk
to
that
this
week
we
are
instituting
owners
files
for
charts
themselves.
So
if
somebody
owns
a
chart,
we
can
get
them
an
owner's
file
and
they
can
approve
stuff
themselves
to
go
into
the
chart
without
the
chart.
E
E
The
old
keep
open
label
that
used
to
be
in
place
isn't
working
under
the
new
tooling,
although
it
did
under
munch
github
the
tool
that
you
used
to
do
this,
which
was
only
available
on
kubernetes
kubernetes
and
no,
you
cannot
turn
it
off
for
an
org
and
it
just
closes.
It
warns
you
before
it
does
things
and
tells
you
how
to
handle
the
situation.
So
it
does
leave
comments
on
there
saying
this
is
now
stale.
If
you
don't
want
it
to
be
stale
handle
it
this
way
and
freeze
will
work
for
90
days.
E
E
Never
do
it
we'll
see
how
that
goes,
because
now
people
want
to
debate
it,
but
the
PR
is
in
so,
and
you
already
use
to
keep
open
label
on
at
least
one
issue
and
so
we're
trying
to
bring
it
so
does
kubernetes
kubernetes
and
there's
been
a
request
for
this
for
mother,
so
PR
is
already
in,
but
yes
that
was
just
turned
on
like
20
hours
ago
or
something
maybe
less
so
start
to
see.
Stale
things
being
notified
here
shortly.
F
C
F
Don't
know
what
to
do.
Do
I
type,
my
minutes
first
or
do
I
talk
first
and
then
type
my
minutes.
So
last
week
was
cook
on.
We
had
lots
and
lots
and
lots
of
stuff
going
on
Taylor
killed.
It
Taylor
got
a
couple
of
sort
of
impromptu,
helm,
related
things
going
so
like
the
a
bunch
of
helm,
helm,
helmets,
hell,
Helms
people
I,
don't
know
all
went
to
the
after-party
together,
which
was
pretty
cool
all
this
helmet
helmets.
F
A
F
Lot
of
people,
one
of
the
ones
that
stood
out
the
most
was
I
met
with
SK,
which
is
a
South
Korean,
the
biggest
South
Korean
telecom
company.
There.
They
use
helm
to
roll
out
the
zillions
of
things,
but
in
particular,
they've
run
out.
They
roll
out
all
of
their
OpenStack
clusters
on
a
mega
kubernetes
cluster
using
helm,
and
he
walked
me
through
the
high-level
architecture
and
explained
how
they
were
using
Jenkins
to
do
it
and
on
how
they
manage
their
massive
amounts
of
configuration
data.
It
was
pretty
cool.
F
It
was
definitely
a
good
example
of
somebody
who
took
some
small
tools
and
thought
really
big
and
did
some
really
sophisticated
stuff.
I
guess
Alibaba
also
uses
helm
to
do
massive
deployments,
and
they
talked
about
that
during
their
keynote.
So
and
then
you
know
just
talk
to
all
kinds
of
people.
Listen
to
all
the
kinds
of
discussions
about
how
people
you're
using
helm
what
they
liked,
what
they
didn't
had.
Some
conversations
with
people
like
Brian
grant
and
other
people
in
the
community
to
kind
of
get
a
feel
for.
F
What's
going
on,
I
met
an
old
friend,
Jago
McLeod,
who
is
from
I,
worked
with
him
at
nest
when
I
was
at
Google,
and
he
is
now
on
the
API
machinery
team
he's
the
PM
for
API
machinery
and
I
think
for
architecture.
So
he
was
telling
me
that
they
that
they
plan
on
bringing
several
a
couple
of
people,
at
least,
but
maybe
several
people
to
the
helm
summit
in
February.
So.
F
F
If,
for
some
reason,
you're
not
gonna,
be
out,
maybe
say
something
because
I
think
it
probably
will
make
the
most
sense
for
us
and
not
have
meetings
again
until
January,
but
I'll
leave
that
up
for
the
end,
all
right,
that's
it
for
me,
I
think
yeah,
there's
something
about
bit!
Nami
I
was
gonna,
mention
Oh
bitNami
launched
their
new
Kubat
stuff.
I
was
none
of
them
are
here
yet,
but
we
do
need
to
figure
out
here
in
this
group.
F
B
Okay,
it
all
passed
the
notes
back
onto
you
again,
so
you're
welcome,
so
yeah
last
week
was
a
lot
of
fun
and
on
that
note,
about
Alibaba
cloud.
I
actually
talked
to
the
Alabama
cloud
folks
about
how
they
were
using
helm,
charts
with
tensorflow,
and
they
mentioned
in
their
keynote
that
it
was
30%
faster
than
vanilla,
kubernetes.
B
So
I
talked
to
one
of
the
PM's
over
there
and
he
said
yeah,
that's
a
that
is
a
forked
version
of
kubernetes
that
they're
using
with
helm
to
get
that
30%
increase
because
they're
using
mean
of
GPUs
with
tensorflow
to
actually
do
the
modeling.
So
that's
essentially,
they
haven't
like
hacked
on
helm
or
anything
like
that.
It
was
just
a
different
flavor
of
kubernetes,
something
using
pretty
cool
stuff,
but
same
as
Matt
I
was
engaging
and
talking
with
people
all
over
that
are
helm.
Users,
contributors
and
whatnot.
B
I
was
talking
with
Suzy
on
some
things,
because
we
haven't
heard
from
them
in
a
little
while
with
like
Troy
topic
and
whatnot
and
seeing
where
they
are,
with
their
adoption
of
helm
and
just
kind
of
getting
feedback
and
hearing
about
that,
and
there
was
also
some
people
have
mentioned
out
there.
Yet
Tory.
Wasn't
there
the
thing,
but
I
saw
a
couple
of
people
else
that
are
that
work
in
the
bank
to
cover
office,
so
they
meet
up
with
foreign,
but
there
was
the
helm
or
not
the
helm.
B
Those
sig
apps
meeting,
there's
two
there's
a
cig,
apps
update
and
then
there's
a
cig,
ABS
deep
dive
and
both
of
those
sessions
met
arena
hosted
ended
all
that
which
was
awesome.
There
were
some
really
good
discussions
around
just
kicking
off
helm
the
helm,
dev
summit,
some
interesting
ideas
around
helm.
Three.
There
were
also
some
interesting
use
cases
around
like.
Why
should
we
use
CR
DS
versus
why
we
shouldn't
so
there
were
discussions
there
and
bad
about
these
different
things,
but
essentially
it
was
a
really
good
thing
for
him.
B
H
Okay,
so
sorry
for
the
late
arrival,
as
with
everyone
else,
I
I
was
a
kook
on
and
a
lot
of
time
and
the
same
thing
that
both
Fischer
and
butcher
were
in
so
I,
don't
know
how
much
to
report
from
from
there
and
so
Adam
the
fingers
Reece
yes,
I,
like
that
I
I
approve
of
this
title.
Okay,
so
the
big
thing
is
that
I'll
be
working
on
a
lot
of
the
home
Summit
stuff,
so
I'm,
sorry,
if
somebody
like
needs
help
with
a
PR
and
you're
like
completely
overloaded.
H
H
If
I
remember
right,
you
were
going
to
want
to
help
out
as
well
with
the
CFPs
question
mark:
okay,
yes,
okay,
so
Fischer
and
Adam
we're
gonna
have
a
meeting
tomorrow
with
Michele,
so
I'm
gonna
reach
out
to
you
guys
offline
after
this
to
schedule
a
good
time
tomorrow
to
do
that.
When
we're
all
in
our
Star
Wars
hangovers
and
then
after
side
of
that
I
just
had
a
little
bit
of
the
PR
stuff.
G
H
G
Just
forget
about
me:
yes,
a
big
thing
for
last
week
was
I.
Did
a
PR
41.9
on
the
beta
I?
Did
it
especially
for
Verena,
because
I
know
he
he
cares
a
lot
about
that.
So
I
was
sure
to
get
that
done.
Everything
appears
to
be
to
be
working.
Okay,
there's
a
few
accenting
questions
on
what
types
to
use
in
certain
situations,
because
everything
is
moving
around
so
I
got
some
tall
book
to
do
it
on
that
and
some
questions
to
ask
so.
G
B
A
Great
yeah
so
on
the
home
summit,
we'll
need
a
sink
on
that
so
Taylor
already
has
his
posse
of
people
for
reviewing
CFPs.
But
if
you
want
to
get
involved,
contact
Taylor
is
that
okay,
Taylor,
okay,
so
we're
gonna
do
something
Friday
to
kind
of
just
go
over
like
clearly
and
how
to
review
the
review
process,
and
we
need
to
figure
out
like
we
can
figure
that
out
on
Friday,
or
we
can
figure
that
keep
figure
it
out
here.
A
When
do
we
want
to
make
the
CFP
at
deadline
and
I
think,
based
on
my
experience
with
two
con
I
understand
now
how
to
review
CFPs
in
a
way
that
you
don't
actually
have
to
wait
until
the
deadline
to
review
all
of
them?
So
I
compose
that.
But
if
anybody
has
any
thoughts
on
when
to
close
the
CMP
great
Taylor,
would
you
wait
till
tomorrow?
I
mean.
H
A
I
think
so,
because
you
always
want
to
give
like
people
enough
time
to
look
like
some
people
will
not
come
and
if
their
talk
isn't
selected
because
of
policy
stuff,
so
we
want
to
get
I
would
say:
give
people
like
six
weeks
plan
and
the
combos
probably
want
to
in
for
six
weeks
before
the
deadline
of
the
event
or
the
event
takes
to
send
out
notifications
and
I
think
that
I
might
be
the
cutoff.
So
if
we
could
do
it
even
before,
then
without
would
be
better.
My.
A
A
Right-
let's
just
bump
it
from
this
conversation
to
tomorrow
and
discuss
it
music,
if
that's
okay,
so
the
next
thing
I
have
is
the
security
mailing
list
so
right
now
there
is
a
home
security
at
disk,
calm,
email
that
you
can
report
security,
vulnerabilities
and
I
tested
out
this
last
week
or
not
speak
a
few
weeks
ago
and
we're
still
getting
emails
forwarded
from
that
email
address
and
it
works.
But
obviously
we
should
change
it.
So
we
should
figure
out
like
what
we
match
means
like
you
know,
just
and
then
separately
but
related.
A
Because
I
personally,
don't
love
having
to
use
my
personal
gmail
account
or
something
all
the
time,
but
again
like
this
is
kind
of
the
name
of
the
game.
Now,
because
we
everybody
in
tomorrow's
community
is
using
my
Google
Groups.
So
if
this
is
not
an
issue
for
anybody,
then
we
can
continue
using
Google
Groups.
But
if
it
is
an
issue
for
people
like,
if
you
have,
if
you
don't
want
to
create
a
gmail
to
use
Google
Groups
specifically,
then
you
kind
of
need
to
hear
everyone's
opinion
on
that.
A
H
B
I'm
curious
about
this
too,
because
before
I
thought
that
the
a
precedent
was
that
everyone
else
in
the
kubernetes
community
is
using
Google
Groups.
But
if
the
CMC
eff
are
currently
using
and
hosting
their
own
mailman
service,
then
it
could
be
potential
or
it's
possible
that
we
could
talk
to
the
CN
CF
group
that
is
using
mailman
and
see
about
like
creating
mailman
mailing
lists
like
making
that
decentralized
thing
so,
like
Ubuntu,
has
a
mailman
mailing
list
for
basically
everything
around
and
also
Debian.
They
have
their
own
mail
that
mailing
list
and
it's
around
security.
B
It's
around
package
management,
it's
around
just
general
development
of
DB
and
Ubuntu
and
community
stuff,
and
all
that
general
troubleshooting
and
Wow.
So
would
it
be
easier
to
just
eventually
shift
everything
over
to
mailman?
If
that
is
what
works
for
everyone,
so
everyone
can
use
their
work.
Email
accounts
rather
than
everyone
having
to
sign
up
for
Google
accounts,
but
that
will
probably
require
more
infrastructure
on
CNC
outside
so
I'd
like
if
we
could
and
if
they
said.
That
would
be
like
an
interesting
thing
we
could
do
then.
B
Yeah
I
think
that
would
be
a
very
valuable
thing,
is
to
evaluate
how
much
effort
will
be
required
to
get
on
to
CNCs
mailman's
service
or
if
it
requires
mint
with
ourselves
in
the
sense
of
like
growing
some
money
at
them
can
be
able
to
host
a
mailman
service.
For
the
rest,
the
CNCs
projects,
just
getting
an
evaluation,
will
probably
be
a
great
thing
to
do
that,
and
then
we
can
figure
out.
How
do
we
migrate
over
the
security
mailing
list?
F
A
B
Migrated
over
to
the
teens:
yes,
it's
like
a
background
thing,
not
necessarily
as
something
we
want
to
do
now,
because
we
currently
have
the
common
core
dance
and
everything
else
already
on
Google
Groups.
So
if
you've
already
done
that
and
created
it,
then
we
can
just
already
migrate
over
and
that
doesn't
leave
anything
happen.
A
Yeah
I
can
do
that.
I
just
feel
like
mailing
lists
are
really
annoying
like
I.
Just
would
hate
to
have
to
like
ask
someone
to
make
a
gmail
account
and
then
our
Google
account
and
then
join
this
mailing
list
and
then
be
like
oh
yeah,
but
we're
gonna
migrate,
because
we
don't
want
you
to
have
to
later
make
a
count.
You
know
I'm
saying
like.
E
D
A
way
that
you
can
tell
it
to
send
email
messages
to
one
of
the
other
accounts
that
you've
one
of
your
other
email
accounts
that
you've
tied
to
or
verified
through
your
Google
account.
But
I
will
say
that,
even
though
that
option
is
there
it's
very
hard
to
corral
the
traffic,
for
example,
to
Matt's
point
if
you're
reviewing
a
Google
Doc
and
you
write
comments,
and
somebody
responds
here
you're
watching
the
comments.
Those
seem
to
always
go
to
your
gmail
account.
Those
don't
bounce
off
to
some
other
account
that
you
have
tied
to
it.
D
Those
Google
Docs,
you
do
need
the
Google
account
and
you're
probably
going
to
have
some
amount
of
email
flowing
to
it.
You
know
what
Matt
called
worth
or
they
did
earlier,
the
opposite
of
personal
email.
I,
don't
think,
that's
a
good
reason
not
to
buy
for
mail
man,
because
you
can
still,
if
you
like
to
use
Gmail,
you
can
well
sign.
A
So
I
would
argue
that,
as
as
the
community
grows,
people
like
you
know
right
now.
The
home
community
is
very
also
ingrained
in
kubernetes
communities.
I
think
as
the
as
the
kubernetes
events
get
more
boring,
which
is
which
is
a
great
thing.
I
would
argue
that
the
communities
will
be
less
tightly
coupled
and
so
I,
don't
want
to
say
we
have
to
set.
A
We
have
to
follow
the
precedent
set
by
the
communities
community
just
just
because
it
is
the
way
it
is
because
it's
it
might
be
the
case
that
has
are
things
where
we
won't
be
as
together,
but
right
now
we
all
know
me
so
right
now
we
are
so.
If
we
do
want
to
do
the
temp
thing,
that's
fine
I
just
wanted
to.
B
A
B
B
So
there
was
two
ideas
it
one
of
them
was
about
yeah.
We
wait
until
helm
three
and
figure
that
out
there.
So
dev
summit
is
a
good
idea
and
then
we
could
circle
that
back.
But
an
idea
was
that
on
new
Hellman,
its
with
new
releases,
we
would
deploy
till
a
release
that
is
apps
feet.
One
and
then
helm
in
it
upgrade
would
just
upgrade
the
existing
tiller
deploy
instance
retaining
the
retaining
the
current
API
version.
So
then
anyone
using
a
newer
version
of
kubernetes.
We
get
apps
p1
older
versions
to
get
extensions.
D
E
E
Could
do
that
because
that's
getting
into
some
of
the
capability
stuff,
but
one
of
the
problems
with
that
is,
if
I
remember
it's
using
string
comparisons
and
in
a
string
comparison
100
is
less
than
nine
because
it
evaluates
the
one
first,
because
it's
not
a
version
comparison
so
we're
about
to
run
into
a
problem.
There.