►
From YouTube: 20180606 kubeadm office hours
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
Oh,
all
right
today
is
June
6
2018.
This
is
the
kuba
diem
office
hours
today.
The
tentative
plan
is
to
go
through,
and
kind
of
do
must
do
like
a
doc
slash
test,
sprint
planning
what
I'll
do
is
we
currently
have
some
items
that
are
on
the
agenda
we'll
go
through
that,
then
what
I
was
hoping
to
do
is
to
go,
must
do
like
a
brainstorming
session
for
individual
things.
A
Individual
items
I
also
want
to
take
I
want
to
assign
one
or
two
technical
leads
to
kind
of
follow
up
and
sort
of
manage.
The
current
set
of
Doc's
issues.
I'll
volunteer,
as
one
of
them
I
need
another
person
to
manage
from
the
technical
side,
not
from
the
dock
side.
So
is
there
someone
else
from
the
technical
side
that
wants
to
be
on
point
to
kind
of
triage
all
of
the
issues
that
we
have
and
make
sure
that
there's
consistency.
A
A
A
So
right
now
we
we
are
as
part
of
this
we're
gonna
kind
of
like
do
a
walk-through
of
the
docks
and
make
all
the
issues,
but
somebody
needs
to
rationalize.
We
have
several
issues
that
are
logged
in
currently
in
the
in
the
Covidien
repo,
as
well
as
separate
issues
in
the
website.
We
somebody
to
rationalize
all
that
stuff
and
to
make
sure
that
we're
actually
getting
things
fixed
and
not
having
like
10
issues
for
the
same
thing,
that's
kind
of
what's
involved.
Okay,.
A
B
Jennifer
the
coordinates
issues
are
mostly
appearance,
a
most
converged
right
now
we
only
have
one
remaining
and
it
needs
a
little
bit
of
a
grammar
check
in
there.
Yep
I've
got
it
open.
Okay,
so
I
wanted
to
talk
about
this
core
DNS
service
thing
that
we
have
so
basically
the
signified
people
confirmed
to
me
that
we
are
not
renaming
the
called
the
coordinate
service
to
coordinates
and
it
remains
cube.
Dns
right.
F
I
got
that
and
I
saw
the
comment
and
the
issue
when
that
PR
is
actually
merged.
I
only
wanted
to
flag
it
because
it
seemed
like
a
usability
issue.
I
didn't
want
to
stay
silent,
oh
and
it
wasn't
it
wasn't
a
blocker
I
mean
the
argument
makes
sense
right,
but
it
does
I
think
represent
some
potential
user
confusion
in
the
short
run,
I
actually.
A
F
B
The
next
one
is
the
so
I
would
like
all
the
commands
are
now
submitted
as
PRS.
We
are
only
missing
the
one
which
Lucas
is
adding
right
now,
which
is
I.
Think
a
great
note,
I
think
yes,
yeah
you're
only
missing
this
one.
Everything
else
is
submitted,
so
we
are
finding
terms
of
the
commands,
so
the
use
of
stories
now
these
are.
F
This
is
not
a
great
time
in
the
release
cycle
to
be
trying
to
get
this
kind
of
thing
in
for
a
release.
It's
the
kind
of
thing
that's
really
optimal,
to
do
at
any
time
once
when
11s
is
out
the
door
or
went
at
least
once
we've
frozen
the
docks
repo
and
then
those
PRS
can
all
go
against
master
and
we
can
iterate
on
them.
However,
we
need
to
so
some.
A
A
F
B
Yeah,
so
this
one
is
about
the
we
had
a
bit
of
an
argument:
what
we
should
do
with
configuration
files
when
we
should
we
embed
them
inside
the
website.
Should
we
generate
them,
or
should
we
link
to
go
Docs
so
look
exactly
sure
what
we
should
do
here.
I
think
that,
like
the
decision
was
to
include
the
current
config
like
alpha
2
and
also
link
to
the
go
Dawgs.
A
I'm
of
the
opinion
that
you
can
talk
about
using
the
config
well,
I,
don't
know
the
config
is
really
large
rights,
just
a
lot
of
content.
Yeah
yeah!
We
don't
even
have
a
page
for
all
this
stuff
right,
there's
no
landing
home
that
talks
about
configuration,
and
we
really
should
so
like
part
of
the
set
up
page
and
we'll
do
it
in
the
walkthrough,
but
would
basically
have
a
shell
out
to
a
whole,
separate
page
which
goes
through
config
in
all
its
gory
detail
and
the
I.
Don't
really
have
too
strong
of
opinion.
A
My
personal
opinion
is:
when
you
mirror
content,
it
becomes
maintenance
burden
for
people
over
time
and
we're
gonna
have
to
have
more
maintenance
burden
and
I
know
that
Jennifer
is
like
Doc's
are
amazing,
but,
like
developers
have
a
tendency
to
ignore
things,
so
then
it
becomes
an
issue
right,
but
if
you
cannot
eclis
link
to
something
that
gets
updated
within
the
go
doc
references
as
part
of
the
code,
it's
a
forcing
function
for
us
to.
Actually
you
know
it's
part
of
PR
reviews
make
sure
today.
So
that's
that's
my
opinion
and
I'll
stop
there.
F
Do
but
I
think
what
we
need
to
do
for
the
moment
is
to
branch,
because
I
mean
I'm,
not
averse
to
having
a
conversation
about
the
go,
Docs
and
I
understand
the
argument.
Reference
Docs
are
always
hard.
However,
you,
however,
you
slice
them
and
yes,
the
closer
to
the
code.
The
content
lives,
the
more
likely
it's
going
to
get
maintained,
but
that
said,
using
the
go
Docs
relying
on
the
go
ducks
represents
a
departure
from
anything
else
that
we
do
when
the
website
I
can't
make
a
call
on
that.
F
If
you
want
to
bring
a
proposal
to
sig
Docs,
that
would
be
great.
This
could
conceivably
solve
problems
for
other
SIG's
and
other
parts
of
the
kubernetes
ecosystem
and
the
docs
I
don't
know
I,
don't
think
we
can
just
make
a
call
about
this
for
111,
but
I
would
welcome
you,
know,
sort
of
a
proposal
to
consider
you
know
sort
of
the
cases
where
the
go
ducks
approach
would
be
appropriate
what
the
UX
for
that
would
be
like,
and
then
we
can
take
that
up
as
a
proposal
to
see
Doc's
generating
reference.
F
Docs
of
all
sorts
is
a
thing.
That's
on
us
right,
that's
why
we
follow
with
a
one
sort
of
wrinkly
exception
to
the
queue
medium
Docs,
because
that's
a
fairly
new
thing,
but
you
know
I'm
not
gonna,
stand
on
my
sword
about
the
go
ducks.
First
of
all,
it's
not
really
up
to
me
it's
up
to
the
sig
Docs
maintainer
of
the
group,
but
I
think
you
can
see
where
this
is
a
departure
from
the
way
we've
done
things
in
the
past.
F
A
Walkthrough
so
I'm
I'm,
fine,
with
picking
I,
want
the
least
amount
of
friction
right
now
at
this
stage,
so
folks
can
just
go
so
if
somebody
wants
to
maintain
the
configuration
apparatus
so
Lumiere.
The
action
item
here
is
is
basically
we're
going
to
generate
or
dump
the
data
into
the
docks
they
try
to
find
a
long-term
path
going
forwards
that
make
sense.
A
G
B
B
A
A
Once
we
get
that
in
I
think
we
should
probably
do
a
separate
walk
through
and
just
config
as
well
as
items
for
troubleshooting.
So
let's,
let's
do.
Let's
just
let's
make
a
stance
and
just
say
like:
let's
get
the
generated
content
in
there,
it's
already
versions,
so
we
can
go
back
so
we
didn't
need
to
do
both,
but
once
it's
in
there
let's
do
a
separate
walkthrough
on
the
deets,
because
this
will
affect
a
lot
of
troubleshooting
issues
that
are
going
to
happen
in
the
111
time
frame,
because
we
change
config
so
drastically.
B
F
A
A
A
Start
to
get
that
list
together
and
then
luba
mirror
and
myself
will
try
to
anything.
That's
not
already
tracked,
will
try
to
get
it
in
place
and
try
to
get
those
PRS
against
them.
Does
that
seem
like
a
reasonable
approach
to
people
right
now,
so
I'm
good
I
see
thumbs
up
so
alright.
So,
let's
once
more
into
the
breach
right.
The
first
portion
here
that
there
are
separate
tabs
up
on
the
top
in
it
I'm
not
going
to
go
into
content
structure,
I'm,
just
gonna,
try
and
separate
out
the
pieces
right.
A
One
thing
that's
confused
me
is
we
have
a
creating
a
che
clusters
with
ku-band
and
we
have
a
using
Ku
band
to
create
a
cluster
this.
This
always
like
confused
me,
because
in
order,
I
would
almost
like
to
say
creating
a
single
node
cluster
and
saying
using
cube
ad
or
creating
create
a
single
node
cluster
with
cube.
A
DM
is
that,
okay
with
folks
souls
good,
alright,
so
Chuck,
if
you're
taking
notes,
that's
kind
of
like
the
first
item,
no.
B
A
E
E
G
A
H
H
H
A
So,
just
while
you're
doing
the
copy
pasting,
the
one
thing
I
wanted
to
update
for
the
instructions
is
after
cou,
medium
is
installed.
I
wanted
to
pre
pull
the
images
before
you
actually
do
in
it.
I
talked
with
Lucas
about
this
any
any
A's
or
nay.
As
to
that
one.
The
reason
why
I
want
to
do
a
people
images
is
because
sometimes
in
it
gets
wedged
right
there
there's
a
number
of
issues
that
kind
of
occur.
A
A
vast
majority
of
them
is
qat
a
minute
failed
and
there
there's
all
kinds
of
connectivity
problems
that
can
potentially
happen.
But
if
you
do
the
pre
pull
images
as
a
first
step
of
initializing,
your
masters,
you
know
for
certain
that
you've
kind
of
eliminated
a
whole
case
of
problems
that
happens
in
the
pre-flight
checks.
Now
it
does,
but
you.
A
I
D
B
G
E
D
A
F
A
F
A
F
A
F
A
F
J
A
B
D
F
You
have
that
section
cut
my
eye
too.
When
I
went
looking
for
information
elsewhere
in
the
docs,
we
have
more
general
information
that
doesn't
provide
specific
release,
numbers
and
specific
dates.
I
couldn't
find
anything
this
precise,
but
I
agree.
It
does
not
belong
in
Cuba
diem
and
I'm
happy
to
take
an
action
item
back
to
sig
ox
to
say:
okay,
because
we've
been
trying
to
get
our
get
our
hands
around
all
of
this
stuff
right
because
we
need
our
virgins
need
to
map.
So.
F
A
A
Before
you
in
it,
though,
there
needs
to
be
like
a
link
to
config
somewhere
here
like
it's
something
that
outlines,
because
it
says
it's
like
notes
afterwards,
I,
don't
know
what
other
people
think
it's
like
right
now
we
have
this
like
you,
have
the
command
listed
there,
but
then
you
have
the
content.
The
context
of
hey
it's
listed
here,
I.
A
A
F
These
all
look
like
these
need
to
be
interleaved
in
the
relevant
procedure
right.
These
are
all
things
you
need
to
think
about
before
you
call
this
command,
you
need
to
choose
a
pod,
Network
plugin
in
the
next
step.
So
is
this
more
like
a
checklist,
you
know
things
need
to
gather
before
you
start
it
is
it's.
F
F
We
don't
have
consistency
at
this
level
and
I
mean
we
have
prerequisites,
but
that's
more
like
you
need
a
cluster
alright
for
a
task
or
tutorial,
and
that
said,
the
prerequisite
section
is
the
closest
thing.
We've
got
to
a
standard
for
that
kind
of
content.
It
should
also
be
written
to
look
more
like
prerequisites
if
you
want
to
read
more
about
the
flags.
It's
like.
Okay,
that's
not
really
a
thing!
You
have
to
do.
F
B
A
A
A
B
B
F
A
A
A
Then
there's
the
well
we'll
get
into
the
other
guys
in
just
a
second
here
but
I'm
trying
to
be
as
efficient
as
fast
as
possible.
There's
a
bunch
of
details
here
that
don't
make
a
lot
of
sense.
So
why
don't
we
I
would
like
someone
else
besides,
just
little
mirror
falling
on
this
one
I'm
gonna
throw
a
Chuck,
Liz
or
Jaison
on
this.
So
one
of
you
pick.
A
A
B
You
so
me
and
Jason
discussed
that,
probably
like
the
link
of
issues
like
the
filtered
labor
issues.
We
have
should
probably
go
here,
because
the
users
can
immediately
check
if
we
have
already
covered
this
particular
problem
they
are
experiencing.
So
I
think
this
was
the
best
placement
Jennifer
mentioned
that
we
should
track
it
somewhere
else.
Well,.
F
Not
if
these
are
ongoing
issues,
but
if
there
were
known
issues
in
111
that
are
planned
to
be
fixed,
they
go
in
the
tracking
issue.
They
go
in
the
workarounds
in
the
release
notes,
but
if
they
are
I
like
known
things
that
are
kind
of
stable,
known
things
that
you
don't
know
when
you're
gonna
fix
they
go
gear,
we
can
basically.
H
A
A
One
second,
so
doc
the
troubleshooting
guy
here,
I've
already
assigned
Liz
as
part
of
this.
If
there
are
known
issues,
what
we
should
do
is
kind
of
rally
on
this
one,
but
this
whole
troubleshooting
guide
needs
to
be
gone
over
with
a
fine-tooth
comb,
because
right
now,
some
of
these
things
don't
even
apply
and
I'd.
Much
rather
have
it
be
up-to-date
with
things
with
aspects
that
are
useful
with
things
that
people
have
actually
seen.
I
A
Here's
a
big
problem
that
I
don't
have
an
answer
to
our
cluster
guide
is
very
easy
and
friendly
for
pre
provision,
nodes
that
don't
integrate
with
cloud
providers.
What's
the
current
state
of
the
external
cloud
provider
integration
Docs,
and
can
we
link
to
it
in
some
way
as
the
after
saying
for
or
pre
can
addition
for
Koob
ADM,
Jennifer
I,
don't
know
the
state
of
that
other
than
like
I
know
that
it's
out
of
hand
wavey.
It's.
F
All
hand
wavy
I've
been
trying
to
get
an
answer,
because
this
is
breaking
umpteen
gajillion
things
right
now.
It's
supposed
to
be
moved
out,
I'm
supposed
to
be
getting
a
status,
I'm
real
sorry
to
continue
to
hand
wave,
but
I
will
take
an
action
item
to
push
on
the
sig
to
get
resolution
this
week,
I'll
tell
you
Mike
I'll,
give
you
a
guess.
You
know
for
right
now,
I
think
for
111.
F
We're
gonna
have
to
at
least
link
out
to
those
topics,
even
if
we
don't
provide
anything
way
to
navigate
navigate
to
them,
they're
still
getting
built
as
part
of
the
dock
site.
That
content
all
still
exists.
So
if
we
have
links
in
there,
they're
not
going
to
break
for
now
for
now
and
that
content
goes
OK
in
the
111
site
on
the
win,
11
release
I'll
be
on
the
hook
to
fix
it,
but
that
doesn't
solve
the
problem
of
the
content
itself.
D
I
D
A
A
D
A
D
A
F
There's
no
link
to
the
stuff
navigate
to,
but
you
find
it
when
you
search
that's
still
in
the
dock
site
in
that
in
the
chat.
There
is
also
some
content
about
cloud
providers.
That's
kind
of
officially
sanctioned
still
I'm,
not
quite
sure
why
it's
there
so
yeah
the
docs
are
even
more
confused
than
I
realized,
but
but
AWS
in
particular
seems
to
have
an
official
place
we
can
link
to
if
we
need
it.
F
I
I
think
it's
complicated,
though,
because,
like
in
the
case
of
AWS,
you
have
to
set
kind
of
that
hostname
override
otherwise
cube
ATM
and
it
won't
even
work
list
of
whether
you
try
to
enable
a
cloud
provider
or
not.
Well,
if
you,
as
long
as
you,
try
to
enable
a
cloud
provider,
you
have
to
override
it
and
and
I,
don't
think
that's
the
case
for
the
other
clouds
out
there
and
that's
that's
why
I
called
that
out
for
troubleshooting
yep,
so.
A
I'm
gonna,
let
you
three
start
to
take
point
on
that
particular
aspect
and
we'll
kind
of
like
loop
back
around
to
it
and
try
to
think
about
how
we're
going
to
do
with
it.
But
there
should
be
definitely
like
some
type
of
precondition
is
a
prerequisite
so
like
right
now
as
part
of
the
initial
installation
guide,
we
kind
of
have
like
a
before
you
begins
as
part
of
this.
D
G
What
I
think
it
is
missing
a
year
is
before
troubleshooting
a
shooting
is
guide
a
smoke
test
guide
that
I'm
to
in
checking.
If
the
cluster
is
really
working,
I'm
thinking,
for
instance,
that
the
smoker
guy
that
is
East
6th
at
the
end
of
kubernetes,
they
are
the
way
there
are
free
tests
that
make
you
check
if
the
dns
really
work,
if
exact
cucit
on
exact
walks
and
so
on,
we
think.
A
A
B
A
Why
don't
we
do
like
a
add,
a
next
steps?
You
know
now
your
cost.
Now
your
cluster
successfully
stood
up.
What's
next
right,
like
link
to
the
what
you
can
do
afterwards
like
what's
next
add
the
other
bits
into
there
as
part
of
a
smoke
test
right,
your
cluster
that
seemed
legit,
yes
right,
cool,
so
I
will
add
that
to
where
is
the.
A
All
right
so
I
want
to
go
through
meta
topics
now
because
we're
running
at
a
time
to
go
through
it.
I
do
know
that
Chuck
and
ribbon
are
working
on
the
h.a
update
and
well.
Maybe
we
should
set
up
a
separate
little
conversation
to
have,
or
put
it
inside
a
slack
to
see
if
other
people
are
interested
to
basically
do
a
walk-through
and
a
Redux
of
the
current
h.a
documentation
that
calls
out
using
Canadian
phases
and
all
the
details
for
that.
A
It
would
also
kind
of
use
some
of
the
new
stuff
that
Lukas
has
created
to
make
it
a
cleaner
deployment
option
for
a
stack
master,
because
that's
a
simplified
h.a
environment
for
most
folks-
and
this
is
kind
of
something
we're
working
on.
But
we
want
to
make
sure
that
you
know
it
the
lines
with
whether
folks
are
thinking.
How
does
like
how
does
communicating
via
slack
and
maybe
setting
up
some
other
time
to
talk
about
this
stuff
or
or
not
a
contrary
thing
would
be.
A
The
the
problem
was
some
of
the
ancient
documentation,
too,
is
its
kind
of
split
and
I.
Don't
know
why
it's
split
where's.
My
main
thing
here
setup
has
a
piece
of
a
cheat
documentation,
but
then,
if
you're
going
to
be
tasks,
there's
the
separate
other
upgrade
documentation
and
some
of
that
needs
to
be
pruned
and
cleaned.
A
A
D
D
A
Then
there's
the
there's
lifecycle
aspects
of
this
yeah.
So
what
I'm
kind
of
thinking
is
that
there'd
almost
be
like
underneath
setup
there?
There
would
almost
be
separate
issues
right
like
one
is
setting
up
the
cluster
and
there
I
don't
know
if
it
makes
sense
to
have
it
area
set
up
or
not.
So
here's
our
more
like
upgrades
to
me
I,
would
put
upgrades
under
tasks.
Okay,
that
seems
reasonable.
F
F
A
I'm
gonna
do
the
same
thing,
so
what
I'm,
gonna
try
and
do
after
this
meeting
is
I'm
going
to
take
a
look
at
Jennifer's,
spreadsheets
I'm
gonna.
Take
the
current
a
look
at
the
current
backlog
that
exists
inside
of
our
repository
and
then
I'm
going
to
take
a
look
at
the
the
notes
that
we
currently
have
from
walking
through
some
of
this
stuff
and
try
to
slice
through
all
of
them
to
get
a
concrete
list
of
action
items
that
we
can
execute
against.
Does
that
seem
react,
a
reasonable
approach.
F
This
bridge,
it's
got
suggestions
for
how
to
think
about
it
to
not
not
what
to
do
but
you're
like
indicating
whether
we
remove
something
or
move
it
and
where
we
move
it
to.
So,
if
you
want
to
use
that
or
attach
it
to
an
issue
whatever
you
want
to
do
with
it,
I
gave
everybody
edit
permissions.
Okay,
you
put
that
in
the
channel
right.
Oh
sorry,
no
I
put
it
in
the
minutes.
Okay,.
A
I
will
look
through
the
medicine
I
think
we're
at
we're
nearing
time
so
I
want
to
see.
If
there
is
there
other
sort
of
documentation
things
that
we,
you
know,
we've
only
kind
of
touched
it,
but
really
we
need
to
go
back
now
and
actually
like
rationalize
these
separate
sources
of
information
and
make
sure
that
we
can
execute
in
a
reasonable
timeframe.