►
From YouTube: 2021-01-07 CAPZ 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
All
right,
hello,
everyone
today
is
tues
thursday,
sorry
january
7th,
and
this
is
the
cluster
api
provider
azure
office
hours
happy
new
year,
everyone.
I
hope
you
all
had
a
good
holiday
and
yeah
it's
good
to
see
everyone
here.
So,
let's
get
started.
A
B
It's
hey:
it's
here:
I'm
working
on
the
erickson
side
on
the
cluster
api
power
metal
cube.
So
I'm
I'm
the
new
one
here.
So
I
want
to
say,
hi
and
then
we'll
have
to
contribute
to
the
cluster
api
provider
azure
as
well
so
kind
of
newcomer
so
and
happy
to
be
here.
A
Awesome
welcome
and
thanks
for
your
pr
you're
already
making
some
impact.
So
that's
awesome,
so
yeah,
okay,
so
let's
keep
going
unless
there's
a
new
one
else,
nope
so
yeah.
If
you're
here,
please
add
your
name
to
the
attendee
list
and
add
any
discussion
topics
you
have
to
the
open
discussion
section.
You
can
request
access
to
this
document
by
joining
the
cluster
lifecycle
mailing
list,
if
you're
not
already
in
it
and
yeah.
A
Okay.
So,
first
of
all,
I
just
wanted
to
call
out
in
case
you
missed
it.
There's
a
blog
post
that
came
out
in
mid-december,
basically
like
making
some
noise
about
capzi,
so
yeah,
if
you
haven't,
read
it
check
it
out.
C
A
Yeah
check
it
out,
okay,
so
first
discussion
topic.
I
wanted
to
bring
up
maybe
talking
about
the
release
plan
for
the
next
few
weeks
or
like
maybe
make
a
release
plans
for
the
next
few
weeks.
A
A
So
what
I
was
thinking
is:
maybe
we
could
wait
for
that
and
aim
for
a
last
feature:
release
of
0.3
end
of
the
month
and
then
switch
over
the
main
branch
to
start
working
on
v1,
alpha,
4
and
then
back
port,
any
smaller
features
or
fixes
that
we
need
in
0.3
after
that
into
the
0.3
release
branch.
How
does
that
sound
to
everyone
and
any
thoughts
or
suggestions?
A
Okay
cool,
so
hopefully
that
gives
us
enough
time
to
finish
up
the
bigger
things
that
we
wanted
to
get
into
0.3,
and
then
we
can
start
focusing
on
v1
level
4
after
that
I'll
make
sure
to
update
the
roadmap
as
well,
because
I
think
we're
a
bit
ahead
of
where
we
said
we
would
be
so
just
want
to
make
sure
it's
like
reflecting,
where
we're
at
all
right.
Any
questions
about
releases
in
general
or
any
feature
I
mentioned.
D
A
A
D
E
E
I
need
to
write
this
up,
but
it
might
be,
it
might
be
useful
to
put
together
a
tracing
where
it's
open,
telemetry
based
and
then
use
the
open,
telemetry
exporter
to
azure
app
insights.
That
tracing
can
be
done
through
a
managed
service
as
or
just
through,
like
whatever
open
telemetry
exporter
that
you
would
like
to
use.
E
This
is
something
that
I,
after
after
getting
the
tracing
in
there
for
using
geiger,
I
realized
quickly
afterwards
that
you
know
it
might
not
be
the
the
right
way
to
do
it.
It's
really
useful
for
local
development,
but
this
is
probably
not
how
you'd
want
to
run
your
cluster
out
there.
We
haven't
taken
any
opinions
on
it.
It's
just
we
write
to
a
tracing
provider
and
open
telemetry,
and
then
it
it
pushes
it
out.
E
So
if
we
had
an
open,
telemetry
operator
running
and
then
that
was
then
doing
an
export
for
us
to
like
wherever
you
want
it,
that
would
be
something
probably
not
maybe
not
in
this
milestone.
I'm
just
thinking.
If
people
have
extra
time
and
they
wanted
to
hack
on
something
that
might
be
something.
A
E
C
Cool,
so
when
I
was
working
on
the
windows
tooling,
I
noticed
that
in
the
ci
logs
we
were
pulling
down
cube,
cube
ctl
for
116,
so
I
looked
into
it
a
little
bit
and
we
have
kind
of
two
different
ways
to
install
some
of
the
tooling
that
we
have
so
there's
a
way
to
install
it
via
the
make
file
and
there's
also
a
way
to
install
it
via
hack
and
sure
tool
the
ci
uses,
hack
and
share
tool,
whereas
if
you're
doing
anything
local
development,
we're
kind
of
doing
this
make
file
approach,
I
opened
up
pr
to
kind
of
resolve
the
one,
but
then,
when
I
was
doing
that,
I
noticed
that
there's
some
other
things
and
david
pointed
out
that
this
is
one
of
the
things
that
it'd
be
nice
to
kind
of
move
towards,
using
all
local
tools.
C
I
just
wanted
to
see
if
anybody
had
any
opinions
or
if
it's.
If
we
get
some
agreement,
then
I
might
take
a
stab
at
kind
of
aligning
all
these
tools.
C
Yeah,
well,
I
guess
I
don't.
I
think
that
was
kind
of
the
direction
david
had
mentioned
was
just
being
able
to
have
tools
in
the
local
path
so
that
we're
always
consistently
using
the
same
tools.
I
don't
know
if
I
have
a
super
strong
opinion
there,
but
I
I
think
that
that's
a
pretty
good
approach,
so
I
but
I
think,
having
two
separate
ways
to
install
tools.
This
can
cause
strange
behaviors
here.
A
D
A
C
Okay,
if
you
look
at
the
the
logs,
it
actually
installs
it
through
the
ensure
ssh
tool,
and
that
puts
it
onto
the
path
and
then
we
call
the
make
file
which
puts
it
into
the
local
tool
and
then
for
customize.
We
use
that
local
tool,
whereas
with
cube
ctl,
we
were
using
the
one
that
was
installed
to
the
path
and
so
over
time,
probably
just
those
small
discrepancies
might
cause
some
sort
of
flakes
or
other
things.
So
I
think
of
aligning
in
one
direction
is
probably
good.
A
A
Good,
all
right
david,
you
want
to
take
the
next
one.
E
Yeah,
sorry,
just
filling
in
conversation,
yeah,
so
cool,
so
the
pr
is
out
there
for
create
and
update
asynchronously
for
machine
pool
everything's
passing
except
for
the
api
difference.
E
E
E
This
kind
of
leads
into
doing
the
surge
and
rolling
updates
with
drain
node,
which
is
in
progress
right
now.
If
anybody
has
any
more
feedback
on
it,
it
would
be
great
to
get
it
pulled
in
sooner
than
later.
It
is
a
really
big
pr,
so
rebasing
it
is,
can
be
challenging
so
yeah.
A
How
do
you
feel
about
doing
the
lazy
consensus
on
it
because
I've
approved
it?
I
think
nader
has
said,
looks
good
to
me
and
I
don't
know
if
we're
waiting
for
anyone
else,
it'd
be
great
to
get
more
eyes
on
it.
But
I
don't
want
it
to
hang
and
like
hang
out
there
too
long
and
get
rebased
too
many
times
so.
E
A
I
don't
know,
maybe
tomorrow
end
of
day,
I
don't
know
if
that's
too
aggressive,
but
if
you
want
ace
to
look
at
it,
you
could
just
ping
him
and
ask
him
if
we,
if
he
wants
us
to
hold
for
him
or
if
he's
not
gonna,
get
to
it.
D
A
All
right,
let's
do
that!
Do
you
want
to
comment
on
the
pr
and
say
tomorrow,
yeah
thanks.
A
A
A
If
not,
this
is
the
end
of
the
agenda.
We
already
took
a
look
at
the
milestones.
I
think
we're
good
on
that
front.
Am
I
forgetting
anything
that
we
should
look
at?
Do
we
want
to
check
out
testgrid.
A
A
Yeah
so
yeah
I
saw
the
upgrade.
Tests
are
back
to
mostly
passing,
but
we're
still
seeing
flicks.
So
we
should
investigate
that
and
then
everything
else
looks
pretty
good.
I
can
go
on
it
really
quickly.
D
I
still
have
the
pr
that
was
testing
the
removal
of
experiment
like
the
rejoin
thing.
I
don't
think
we
should
merge
it
in
because
keep
it
around
for
a
while,
but
maybe
for
like
zero
four,
I
guess
zero
five
sounds
good,
just
mentioning,
and
that
is
still
there.
I
I
keep
running
the
test
every
few
days
and
it
seems
to
be
passing
more
often.
A
Okay,
yeah,
it
might
yeah
it's
better.
Maybe
if
we
wait,
because
if
we
change
all
the
v1
off
of
three
templates
for
the
last
release
and
then
and
end
up
making
it
less
stable,
it's
not
great.
Where
is
what
am
I
looking
for
yeah,
so
this
is
yeah
better,
but
still
some
occasional.
D
A
D
A
A
Well,
well,
I
hope
you
have
a
good
rest
of
the
day
and
good
rest
of
the
week
and
see
you
all
on
slack.