►
From YouTube: ErrTooMuchHelm = fmt Errorf"A Maintainer's Story About Helping with Helm" Taylor Thomas, Microsoft
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
ErrTooMuchHelm = fmt Errorf"A Maintainer's Story About Helping with Helm" Taylor Thomas, Microsoft
A
A
A
So
what's
interesting
about
this
as
much
as
Martin
and
I,
were
joking
I've
been
doing
this
for
about
two
and
a
half
years
three
years,
maybe
at
this
point
and
Martin's
been
doing
it
for
a
little
while,
and
so
we
just
want
to
give
it
a
little
bit
of
a
talk
about
how
you
can
be
a
better
contributor
or
to
start
contributing
or
whatever
it
might
be.
Okay,.
C
A
So,
like
it
says
up
here,
I'm
one
of
the
core
maintainer
zazz,
well
I,
have
worked
on
aks
at
Microsoft
and
I
work
at
hell
on
helm,
stuff
as
well,
and
then
I
mean
there's
all
my
social
media
stuff
as
well,
and
I
have
spent
way
too
much
time
with
containers
in
my
career
and
have
some
amazingly
terrible
battle
stories.
If
you
ever
want
to
hear
them
and
so
yeah.
A
That's
me
so
first
off
is
this
question,
so
you
want
to
be
a
contributor
now,
I
just
want
to
start
it
off
with
a
note:
I,
absolutely
love
the
helm
community,
one
of
the
things
that
I
really
loved
and
gravitated
towards
as
this
community
particularly
is
just
kind
of
a
welcoming
nature
of
everything.
It's
very
warm
and
generous,
and
the
people
who
and
the
things
people
do
to
help
each
other
out
and
kind
of
get
people
involved,
made
a
huge
difference.
A
C
Yeah,
but
like
everything
when
you're,
when
you're
stepping
on
in
something
for
the
first
time,
you
can,
you
know,
be
a
little
bit
scary,
you're,
not
sure
where
to
go
whatever.
So
you
know
it
can
be
a
bit
overwhelming
we're
milling,
ooh
yeah.
We
stumbled
over
that
Ward,
so
moving
on
so
I
think.
Take
it
easy
on
yourself
from
your
force
coming
in
okay
and
there's
a
few
ways
you
can
go
about
this,
and
probably
one
of
the
first
things
to
do
is
get
into
our
human
communication
forums.
C
So
we've
we've,
we
wrote
on
slack
so
we're
on
to
the
kubernetes
workspace
and
we've
tree
tenants
out
there.
So
we
have
helm
users,
which
is
for
your
common,
all
wrong
questions
on
helm
and
it's
good
place
to
kick
off
and
dev
when
you're,
it's
some
dev
orientated
contributions,
maybe
around
the
PR,
etc,
and
then
there's
also
the
charts
channel.
Okay,
we
also
have
a
Twitter
account,
so
please
follow
that
and
you'll
get
updates
on
that
and
also
join
our
mailing
list.
Next
up
I'd
say
get
yourself
up
and
running
okay.
C
So
what
I
mean
by
that
is
have
a
look
at
the
documentation.
That's
really
a
good
first
step,
then
maybe
start
playing
around
with
it.
You
know
if
you
can
have
a
cluster
or
somewhere
on
the
cloud
or
just
local
cluster,
with
mini
cube
or
kind
etc
and
get
it
go
on,
because
it's
easy
to
deploy
head
to
get
him
going
and
then
basically
deploy
your
charts
out.
We
have
tools
like
atom
create
and
we
go
through
to
him
in
a
minute
and
join
our
dev
weekly
calls.
C
Let's
do
it
in
my
currency,
which
is
a
5:30
p.m.
Irish
time,
our
British
time,
whatever
you
want
our
6:30
central
european
time
and
then
and
that's
a
great
place
to
come
in
and
if
you
ever
have
a
question,
please
we
open
it
up
to
the
floor
as
well.
Have
a
question.
So
are
you
just
want
to
come
and
listen?
I
just
came
and
listened
for
a
while.
So
I
was
like
this
shady
figure
in
there
in
the
background
and
then
eventually
one
day,
I
asked
a
question.
C
You
know
and
yeah
got
me
going
from
there
and
the
last.
One
then
is-
and
this
is
kind
of
something
that
you
know
comes
up
a
lot
of
time
in
open
source
communities
choose
here,
you'd
like
to
work
in,
and
that
sounds
like
an
obvious
taking
poor
people
come
in
and
they
really
want
to
help.
And
then
you
say
what
kind
of
area
you're
interested
in
and
and
yeah
just
you
know,
have
a
tinker.
Watch
it
like
or
even
ask
questions
it
doesn't
really
matter.
Do
you
know
what
I
mean
boat?
A
So
that
leads
us
into
that
final
point
from
the
final
point
that
Martin
said
into
the
big
chunk
of
what
we
want
to
talk
about
is
what
are
the
ways
in
the
areas
in
which
you
can
contribute.
So
there
are
four
major
ones
that
we
have
listed
here.
Community
management
is
more
I,
use
that
term
very
loosely
you're,
not
necessarily
managing
the
community,
although
that
is
a
possibility
as
well.
It's
more
about
the
communication
aspects
and
helping
out
with
meetings
or
are
planning
local
media,
we'll
talk
more
about
it.
A
Documentation,
I
think,
is
self
explanatory
issues
and
bugs,
and
then
that's
I
guess
self-explanatory
as
well
and
code
and
code
is
a
little
bit
more
complex
than
just
writing
the
code.
There's
some
other
things
to
know
about
as
well,
so
we're
going
to
go
ahead
and
dive
into
those
different
areas.
So
the
format
this
is
going
to
take
here
is
we're
gonna
kind
of
ask
each
other
some
questions
like
kind
of
panel
format,
but
it's
a
panel
of
one
yeah.
You
know
how
that's
gonna
go.
A
A
Con
helm
like
helm,
related
talks
because
there's
plenty
of
other
spaces
where
people
want
to
figure
out
how
others
are
using
helm
and
it
might
just
involve
helmet
for
part
of
what
you're
talking
about
you
can
answer
questions
on
the
mailing
list.
That
Martin
mentioned
we're
on
slack,
that's
very
useful.
That's
how
I
got
started
was
answering
questions
I'm
in
helm,
dev,
based
on
what
I
could
find
and
you
can
give
constructive
feedback
that
way
as
well,
whether
that's
through
proposals
or
through
different
PRS
and
things
that
are
going
on.
What's.
C
Nice
about
slack
in
like
IRC
before
is,
and
that
was
the
scariest
thing
for
somebody
we
don't
know
or
say
years
ago
was
you
know
you
asked
one
question
and
you're
talking
to
someone
and
next
thing:
20
people
have
come
in
and
or
at
the
same
time
and
yeah.
That
is
actually
very,
very
good.
It
opens
you
up
to
a
lot
of
people
so
yeah.
A
Like
that
yeah
so
question,
how
many
of
you
know
that
there's
a
helm
blog
raise
a
hand
by
chance?
Hey
it's
a
little
bit
more
than
I
was
expecting
so
guess
what
there's
a
helm
blog.
You
can
find
it
at
a
helm,
dot
SH,
slash
blog
now.
This
blog
is
actually
very,
very
useful,
and
what
we
look
for
this
isn't
some
like
carefully
curated
I
mean
I
work
for
Microsoft,
but
I
can
definitely
tell
you.
This
is
not
some
like
weird
Microsoft,
controlled
blog.
A
This
is
an
open
community
blog
that
all
you
have
to
do
to
submit
something
to
is
open
a
PR
and
then
ask
for
somebody
to
review
it.
So
if
you
don't
know
what
to
talk
about
or
if
you
want
to
run
an
idea
by
yes,
that's
fine
feel
free
to
ping
us
on
slack,
but
the
blog
post
is
actually
one
of
the
best
ways
to
get
involved
with
the
community
and
we're
looking
for
those
battle-tested
stories.
The
only
real
rule
we
have
is
no
product
pitches,
because
that's
not
a
good
use
of
people's
time.
C
A
C
If
you
were
looking
for
your
Doc's,
it's
quite
easy.
You
can
go
just
do
a
search,
Google,
search
for
helm,
tree
Doc's
for
the
new
helm
tree
doctor
coming
out
and
all
the
good
work
that
has
been
done
on
to
roll
in
and
all
the
people
that
have
pitched
in
it
or
if
you're
looking
for
the
corn,
stalks
or
helm.
Just
look
for
hellim
Doc's.
C
C
So
you
take
it
out
of
the
box
and
where
you
go
and
then,
after
a
few
things
going
wrong,
you
decide
to
take
out
the
instruction
manuals.
So
this
is
kind
of
the
same
to
where
people
actually
literally
die
vain
and
start
doing
things,
and
then
they
go.
What
happened
here
so
I?
You
know
you
say
alright.
This
is
what
happened,
and
you
know
you
give
a
link
to
the
doc
and
they
go.
C
A
C
So
we
definitely
we
love
contributions
to
Doc's,
okay
and
and
for
stuff.
If
you
were
spelling,
grammar,
etc,
please
get
in.
There
definitely
tailor
when
he
puts
in
these
dogs.
You
know
yeah.
So
when
a
grandmother,
sometimes
you
know
we
have
to
that's
a
joke,
sorry,
so
yeah
we
take
all
that.
Definitely
come
in
with
that.
C
If
you're,
adding
extending
in
the
talk
in
some
way
or
section
or
rad
new
capability,
I
suppose
what
we're
looking
for
here
is-
and
this
was
explained
to
me
a
long
time
ago-
probably
maybe
after
copiers
are
development,
and
you
know
how
many
developers
out
there
do
not
like
writing.
Docs
hands
open.
Okay.
C
A
It
really
is
important.
I
know
that
it
I
add
to
that
I
always
think
of
add
even
worse,
like
if
somebody's
had
a
stressful
week
and
they're
dealing
with
stuff
at
home
and
then
all
of
a
sudden,
like
they're,
trying
to
figure
out
something
at
work
and
you're
starting
to
use
Helm
and
then,
like
that's
the
kind
of
thing
I
put
in
my
in
my
head,
when
I'm
trying
to
write
Docs
is
I
can't
I,
just
like
see
it
and
I'm
not
gonna
get
frustrated
from
that.
That's
that's
one
of
those
things.
A
I
look
for
so
now.
One
of
the
other
questions
that
people
ask
is
then
like
what,
if
I
want
to
add
something
completely
new
to
the
docs,
whether
that's
a
new
example
or
a
new
way
of
doing
things,
and
that's
my
bad,
because
I
forgot
to
tell
my
computer
to
stay
up.
But
what
is
the
way
that
we
can?
What
is
the
way
to
go
about
like
adding
something
new?
Is
there
anyone
I
should
talk
to?
How
does
that
work?
Yeah.
C
I
think
this
applies
to
probably
any
contributions
you
want
to
do.
This
isn't
say
that
we
don't
want
you
to
put
contributions
in,
but
I
think
it's
always
worth
checking.
You
know
reach
out
and
slack,
maybe
paying
someone.
You
know
even
Cindy
Melton
main
enlist.
You
know
smoke
signals
we
don't
mind,
but
it's
sometimes
go
to
ask
because
you
know
you
see
it
sometimes
in
in
in
in
Helmand,
in
order
of
source
communities.
C
Somebody
puts
a
hell
of
a
lot
of
work
into
something
and
they
realize
then,
when
it's
going
to
be
reviewed,
that
it
was
already
in
process,
it
was
just
underway
in
or
else
it
was
something
we
looked
at
a
long
time
ago
and
it's
not
applicable
okay,
so
it's
kind
of
like
you
know
the
carpenter.
You
know
measure
twice
cut
once
so.
It
is
worked
connecting
and
find
out
what's
going
on,
because
we
want
to
get
the
PRS
in,
but
it
has
to
be
I,
suppose
soothe.
What's
in
the
project,
etc,
perfect.
C
A
C
So
why
do
I
want
as
difficulties
and
that's
why?
Sometimes
an
issue
goes
on
for
a
long
time
is
II.
Do
you
know
posts
and
comments
over
and
back?
They
take
a
hell
of
a
lot
of
time,
and
you
know
it
can
frustrate
you
if
you've
raised
the
issue,
but
if
we're
not
sitting
at
the
desk,
which
you're
eating
dinner
every
night
with
you,
we
don't
really
know
what
the
problem
is.
Unless
we
get
the
output.
C
A
And
I
just
remember
like
it's
for
us
when
we're
approaching
these
issues,
it's
not
that
we're
trying
to
like
favor
one
over
the
other,
but
if
there's
one
that
we
can
see
and
we
can
tackle
and
there's
a
chart
with
an
example
attached
to
it
and
all
those
types
of
details-
that's
the
first
one
we're
gonna
go
tackle
and
it's
not
because
we
don't
like
you
with
your
other
issue.
That
says
this
is
wrong.
It's
because
we're
we
can
actually
work
on
that
one
really
quickly
and
understand.
A
What's
going
on,
whereas
when
you're,
when
it's
less
clear
on
an
issue,
what'll
happen,
if
somebody
will
jump
on
say:
oh
I
think
I
have
this
problem
too,
and
then
the
original
commenter
comes
back
and
says
no,
it's
actually
this
and
then
clarifies
and
then
it
kind
of
explodes
into
this
thread.
That
goes
twenty
different
directions.
So
it's
all
about
just
trying.
A
A
C
So,
generally,
actually-
and
I
mentioned
it
today,
when
he
talks
this
morning
about
tea,
you
wouldn't
believe
how
use
of
the
scaffold
chart
is
okay,
so
if
you
debug
in
yourself
or
whatever
no
obviously
depends
on
the
situation,
but
for
a
lot
of
situations
where
you
come
in
and
it
may
be
syntax
around
the
chart
or
something
like
that.
Well,
we
often
do
I,
don't
know
if
all
of
us
do
but
I
think
what
a
lot
of
us
do
is
we
just
we
spit
out
a
scaffold
chart.
C
We
modify
their
own
to
whatever
we're
trying
to
do
to
see
if
we
can
reproduce
the
problem,
because,
like
really
you
know
it's
like
the
only
in
like
it's
the
unraveling
of
the
only
until
you
find
where
the
core,
the
problem
is
because
often
the
solution
isn't
as
difficult
as
trying
to
find
the
problem.
So
that's
how
we
often
work
around
the
scaffold
chart
and
that
may
be
a
starting
point
for
us
and
sometimes
where
we
do
and
I've
seen
from
people
like
mash
and
Taylor
etc.
C
Is
that
you
know
what
I
mean
we
speed
hoped
in
what
we
get
so
we
show
the
commands
were
running.
We
show
the
output
and
then
we
go
over
and
back
with
the
person
to
tune
of
thrown,
and
you
know
what
it
is
actually
quite
enjoyable
like
maybe
for
the
person
who's
got
the
issue,
it's
not
as
enjoyable
port.
You
know
when
we're
trying
to
work
it
out.
There
is
there's
something
nice
about
trying
to
find
the
problem.
Obviously,
if
it's
not
a
Pandora's
box
and
you're
orbiting
you
go,
we
will
shut
that
sorry
yeah.
A
C
Yeah,
so
what
we
generally
do
is
so,
if
you
want
to
go
into
the
community
into
the
community
and
the
contribution
Docs
you'll
find
where
all
the
labels
are
specified
and
generally
what
we
had
things
we
for
feature.
If
it's
a
new
feature
boat
for
a
bog,
we
have
refactor
and
then
we
have.
If
you
ever
see,
question
support,
don't
get
too
excited
stay.
Camp
question
and
support
usually
means
we're
not
saying
you're
you
this
you
have
is
not
a
problem.
C
All
we're
saying
is
we're
trying
to
ascertain
what
the
problem
is
before
we
stick
a
bug
on
it.
Okay,
so
that's
all
we're
trying
to
do,
and
sometimes
yeah
cuz
questions
support
we
use
as
well
for
when
somebody's
asking
a
question
like
you
know
what
I
mean
you
know,
how
did
you
get
to
I?
Don't
know
the
canals
in
Amsterdam
or
something
like
that.
You
know
it
was
often
a
kind
of
a
generic
late
question
border
yeah.
C
A
No
I'm
saying
this
one's
really
interesting
to
me,
because
people
think,
like
oh
contributing,
that
means
I'd
write
code
right
well,
now
we're
actually
the
code
at
the
very
end,
because
it's
not
the
only
thing
that
makes
the
project
run
so
now
we're
gonna
actually
talk
about
the
code.
If
that's
what
you're
curious
about
yeah.
C
So
I
think
that's
that's
a
great
point
from
from
Taylor.
Is
that
look
you
know
every
contribution
is
a
contribution
and
I
know
what
this
gets
said
in
a
lot
of
places
in
every
community
port.
We
mean
this,
so
it's
not
all
about
I
can
call
if
you
love
called
great
yay
if
there,
if
you
love,
docks,
yay,
test,
etc.
So
I
suppose.
The
first
question
is:
when
we
mentioned
pr's,
remember
it's
PRS
for
everything
doc
set
cetera.
So
how
do
you
review
and
test
your
peers?
Okay,.
A
So
this
is
one
that
I
know
is
very
interesting
for
people.
We
have
to
consider
a
bunch
of
different
things
when
it
comes
to
code
and
those
factor
heavily
into
what
our
decisions
are
with
a
PR
one
of
the.
If
you
attended
my
talk
yesterday
about
the
white
flag,
that's
a
good
example
of
it.
How
do
I
submitted
that
wait?
A
Pr
today,
oh
yeah,
that
thing
would
have
been
rejected
flat-out,
but
so
I
learned
from
that
and
I
think
we
all
learned
as
as
maintain
errs
from
that
experience,
that
when
we
look
at
it,
we're
looking
for
other
things
about
what
I
call
like
the
knock-on
effects
that
go.
It's
like
a
domino
effect.
He
goes
from
one
thing
to
the
other.
That
was
one
of
the
reasons
that,
for
example,
we
rejected
the
templating.
A
A
A
We
have
acceptance
testing
so
if
we're
just
trying
to
hook
it
all
together
into
helm,
3,
but
there's
still
manual
testing
and
edge
cases
that
we
have
to
handle,
because
that's
what
we
generally
do
when
we
grab
your
code.
Okay,
that
looks
like
it
works
now.
What
if
I
like,
try
to
shove
a
chart
that
looks
like
this
into
the
into
the
P
R
what
happens,
and
so
that's
what
we're
doing
with
it,
and
so
when
you,
when
we're
trying
to
get
feature
whatever
you
want
merged.
A
If
you
have
a
test
example,
whether
that's
in
the
issue
that
you're
linking
to
or
if
this
is
a
new
thing
to
something
where
you
upload
the
actual
chart,
something
that
we
can
use
to
test
and
replicate,
show
how
you
did
the
steps
to
replicate.
You
could
say:
I
tested
all
these
scenarios
and
did
all
these
things
Matt
Fisher's
really
good
at
that.
He
always
lists
every
single
test
scenario.
He
ran
it
through.
C
A
A
First,
there's
plenty
of
small
features
or
enhancements
that
people
ask
for,
and
those
are
the
good
things
to
kind
of
go
into
and
look
for.
First,
just
on
that
on
the
issue
that
you
see
with
the
bug
in
or
if
you
want
to
do
something
just
kind
of
claim
and
say:
hey
I'm,
gonna
work
on
this
or,
if
you're,
just
trying
to
open
a
PR
and
there's
no
specific
issue.
You
can
say
in
the
helm,
dev
channel,
I'm
thinking
about
doing
this.
A
A
So
we
expect
the
unit
tests
on
most
PRS,
except
for
there's
certain
parts
of
helm
where
it's
more
like
you
can't
really
test
it
in
a
unit
test
e-way
very
easily,
and
that's
what
we're
doing
for
the
acceptance
testing,
which
will
be
hooked
up
pretty
soon
here
to
helm
3,
but
make
sure
that
you
have
your
your
unit
tests
in
place
and
then
we'll
get
the
acceptant
test
in
place,
which
then
you
should
be
able
to
run
as
well.
We're
gonna
make
sure
that's
as
easy
as
possible.
Okay,.
A
C
C
That
so
mark
and
Josh,
they
did
a
great
project.
It
put
a
lot
of
work
into
the
acceptance
testing
project,
that's
in
the
helm,
org,
and
we
have
one
or
two
PRS
that
have
been
hooking
it
in
lately.
So
you
should
be
able
to
run
your
acceptance
test
now.
Okay,
who
put
me
on
the
spot,
have
a
look
at
the
make
file
actually
for
the
acceptance
test.
I
cannot
remember
the
tag
on
that
also
when
you're
your
own
local
tests.
A
A
C
I
actually
don't
know
Oh
No
yeah,
the
first
one
I
actually
did
ended
up
being
a
plug-in
on
its
own
I.
Don't
know
it
yeah
I
think
it
depends
on
the
situation.
So,
like
you
know
it
can
be
quite
quick
or
it
can
take
you
a
bit
of
time.
I
I
know
a
lot
of
people
starting
it's
a
good
thing
around.
Any
sane
takes
and
stuff
like
that.
C
That
shouldn't
take
long
and
I
suppose
one
point
of
to
say
is
because
it
for
the
last
definitely
couple
of
months
or
more
and
we're
trying
to
balance
support
of
Helen
v2
and
get
free
tree
out
the
door
that
we
are
overloaded
and
we
are
a
small
community
of
core
develop
or
core
maintainer.
So
you
know
just
have
patience
with
us
as
we
work
through
truly
support.
I
think
it's
yeah.
It's
depends
on
the
situation
yeah
and
it
I.
A
The
the
way
that
I
would
recommend
is
first
start
in
the
issue.
Sometimes
it's
really
hard
like
github
notifications
are
all
set
differently
for
everybody,
and
sometimes
it
happens
like
somebody
was
out
on
paternity
leave
or
somebody
was
out
on
vacation
or
whatever
might
have
happened
and
things
slip
through
happens
to
me
all
the
time
and
that's
where
generally
just
a
nice
gentle
reminder,
ping
on
slack
and
generally
in
public,
because
sometimes
like
oh
hey,
Martin
can't
handle
that
PR
right
now,
maybe
I
can
right.
A
Now
it's
like
Martin,
said
it's
a
little
bit
more
difficult
with
the
v3
stuff
because
we're
trying
to
push
that
out
the
door
and
make
sure
that
that
is
ready
and
solid
for
everybody.
So
it's
just
been
a
lot
of
focus
on
that.
So
there's
been
some
things
where
it's
just
like
smaller
things
have
kind
of
been
left
because
we're
trying
to
focus
on
those
things
that
need
to
get
done
for
the
PR
for
the
helm,
three
stuff
so
yeah.