►
From YouTube: SIG Cluster Lifecycle - Cluster Addons 20210706
A
Right
welcome
everyone
to
the
cluster
add-ons
meeting
today
is
the
sixth
one.
I
just
had
a
brief
look
at
the
agenda
and
it's
empty,
but
I
saw
that
the
last
time
the
sum
of
code
project
was
discussed
were
any
of
you
there
at
the
last
meeting.
B
I
think
I
was
there
if
I
wasn't
there,
I'm
pretty
sure
nick
was
there.
I
I
don't
have
any
real
updates
for
google
summer
of
code
other
than
that.
As
far
as
I
can
tell,
no
one
was
selected
for
this
project
or
no
one
selected
this
project,
so
I
don't
think
we
actually
have
some
record
folks.
A
It
looks
like
yeah,
it
would
be
coding
by
now
already
well.
It
really
surprises
me
that
that
we
didn't
have
applications.
B
Yeah
there
was
a
lot
of
discussion
on
the
the
github
issue
and
some
interest
from
some
of
the
sig
node
folks.
So
maybe
it
would
be
good
to
go
to
the
signo
meetings
and
see
if
we
can
keep
pushing
on
that
work.
Even
if
we
don't
have
google's
nerf
code
folks
to
work
on
it.
A
C
Hi
friends,
I'm
I'm
just
running.
Oh
I'm
just
juggling
a
few
things,
so
I'm
not
really
clear-headed
on
any
particular
priorities
in
the
moment.
For
this
meeting.
C
I
you
know
it's
just
a
side
update,
but
I
was
able
to
talk
a
little
bit
with
michael
bridgen
maintainer
from
the
flux
project,
who
you
know
just
mentioned
again
that
he
would
be
continually
keen
on
seeing
oci
image,
support
in
source
controller
to
be
able
to
unpack
kubernetes,
manifests
and
other
add-on
type
sources
within
flux
for
github's
flows.
C
That's
of
interest-
but
I
think
that's
been
said
before
here.
A
Is
there
any
work
that
would
need
to
be
done
outside
of
flux
like
anything,
we
would
need
to
do
or
stuff
that
needs
to
be
discussed
elsewhere.
C
I
know
that
this
is
a
technical
need
for
ruck
pack
as
well
at
some
point,
so
there's
really
going
to
be
some
similar
code
being
used
there,
whether
or
not
that's
the
same
library
that
can
be
used
across.
Everything
is
up
for
exploration
for
sure.
I
know
that
when
we
were
speaking
last
on
the
cluster
add-ons
meeting,
the
ruck
pac
effort
was
just
experiencing
some
small
technical
delays,
but
just
other
prioritization
from
the
folks
involved
and
that
meeting
wasn't
recorded,
but
we
noted
it
down
in
the
meeting
notes.
C
I
believe
I
assume
that
that
might
still
be
the
case
that,
like
technical
development
on
rough
pack,
software
stuff
is
currently
paused
a
little
bit.
D
Can
folks
hear
me,
I
don't
know
my
mic
was
working
nick
yeah,
so
there
is
some
more
stuff
merged
to
that
repo
right
now,
some
of
the
basic
skeleton
of
the
types
themselves,
the
controllers.
What's
not,
there
is
the
pulling
of
the
stuff.
That's,
hopefully
what
we
plan
to
get
done
over
the
next
three
weeks,
so
our
sprint
cycle,
but
that's.
C
That's
where
we're
at
so
the
pulling
of
those
images
is
basically
the
same
category
of
problem
that
I
was
mentioning.
Then
it
is.
D
I
think
we've
mentioned
it
several
different
times
before,
but
the
the
constraints
for
pulling
the
image,
mainly
that
we
actually
use
the
cubelet
to
pull
the
image
through
cryo
and
not
directly,
is
going
to
be
one
of
like
the
main
differentiating
factors
between
what
this
and
something
like
image
package
does.
Unless
we
have
what
the
g-suck
project
is
doing
for
free,
where
we
can
just
tell
cube
to
pull
an
image
and
then
give
us
the
ability
to
inspect
it
as
a
volume.
C
Yeah,
like
a
csi
volume
driver
before
that
and
that's
nice,
because
it's
a
plugable
component.
D
Exactly
which
was
also
one
of
the
things
that
we
were
talking
about
as
an
interface
for
ruckback
is:
can
we
do
that
separately
and
just
abstract
it
away
for
everybody
else?.
C
Yeah
and
there's
certainly,
you
know
to
be
able
to
use
that
kind
of
csi
driver
within
flux.
A
source
controller
would
be
interesting,
or
vice
versa,
to
be
able
to
use
source
controller
as
an
http
source
for
the
ink
cluster
pull
of
the
ruck
pack.
Inflation
would
also
be
useful,
so
there's
definitely
room
to
make
the
components
more
interoperable
and
to
think
about
how
that
should
be
pluggable
on
in
the
flux
project
in
the
rock
pack
project.
C
Great
well,
it
sounds
like
that's
a
near-term
technical
need
that
the
ruckpack
team
will
be
exploring,
and
let's
keep
that
conversation
going
if
there
are
like
potential
collab
or
even
just
like
design
considerations
that
we
want
to
make
together
in
the
cluster
adams
channel.
D
Yeah,
I
think,
as
we
start
merging
stuff
over
the
next
like
week,
or
so
sorry-
maybe
not
this
week,
but
over
the
next
three
weeks,
I'll
try
to
keep
posting
updates
when
things
get
in
and
also
some
of
the
sort
of
technical
docs,
where
we
kind
of
outline
different
options
for
things
I'll
try
to
share
them.
More
broadly.
C
C
Well,
well,
yeah,
the
past
few
times
we've
used
this
meeting
as
a
place
to
discuss
and
vet
like
some
of
the
content
of
the
technical
issues,
it
sounds
like
right
now.
We
probably
don't
have
any
more
things
to
flesh
out
through
our
agendas
anything
else
that
people
need
to
be
able
to
bring
up.