►
From YouTube: OKD Working Group Meeting 07-04-2023
Description
The OKD Working Group's purpose is to discuss, give guidance to, and enable collaboration on current development efforts for OKD, Kubernetes, and related CNCF projects. The OKD Working Group includes the discussion of shared community goals for OKD 4 and beyond. Additionally, the Working Group produces supporting materials and best practices for end-users and provides guidance and coordination for CNCF projects working within the SIG's scope.
https://okd.io
A
B
Welcome
folks
to
the
okd
working
group
meeting
for
July
4th
of
the
Year
2023,
please
take
a
few
minutes
to
look
at
the
agenda
and
see
if
there's
anything,
that's
been
missed.
Anything
that
you
want
to
move
around
anything
that
you'd
like
to
postpone
discussion
on
or
anything
like
that,
we'll
take
about
30
seconds.
For
that.
B
D
Yeah
it's
okay!
Last
week
we
stabilized
our
pipelines
in
mlc,
so
there
is
only
one
step
that
we
need
to
make
sure
it's
working,
but
basically,
probably
by
the
end
of
this
week,
we
will
be
able
to
do
the
okd
releases
again
and
so
I
think
that's
it
do
you.
Do
you
have
anything
else?
Luigi.
C
D
E
Hey
folks,
how's
the
Sun
good,
all
right,
so
after
a
couple
of
ideas,
three
items
be
precise,
but
all
of
them
are
about
upcoming
changes,
so
things
that
haven't
landed
yet
in
federal
Aquarius,
but
things
that
we're
looking
at
doing
so.
E
This
one
is
about
ipv
support,
so
you
might
not
need
it
right
now,
if
you
want,
unless
you
want
to
run
firaco
versus
LTD
on
ipv
on
top
of
windows,
but
it's
something
that
is
interesting,
for
example,
for
if
you
want
to
run
things
like
Jupiter
us
or
or
systems
like
single
node
okd,
for
example,
on
top
of
hyper-v
directly
on
top
of
windows,
so
support
five
review
has
been
coming
nicely
and
should
happen
soon
in
federal
course,
and
then
we
can
get
sport
in
okg,
then
we
can
get
single
node,
okay,
keep
working
on
Windows
directly
on
top,
so
that's
one.
B
E
Is
about
firmware
update
metadata,
so
this
is
the
change
that
we're
doing
in
federal
course
as
well.
That
will
come
as
well
to
okg
is
that
we
enable
the
Refresh
on
the
firmware
update
metadata
by
default.
So
what's
all
this,
the
the
firmware
so
film
web
D
is
is
a
tool
that
lets
you
update
the
firmware
of
Hardware
devices
on
your
system.
E
It's
using
the
lvfs,
the
new
expander
Computer
Service,
to
get
the
firmware
from
the
internet
from
vendors
there,
and
so
what
this?
What
we're
not
doing
here
is
the
metadata
fishing.
So
it's
actually
fetching
the
list
of
people
updates
from
the
network
and
getting
them
to
your
computer.
And
if
you
have
a
few
more
update,
then
it
will
show
you
a
nice
message
on
the
login
screen
and
tell
you:
hey.
You've
got
like
two
pieces
of
Hardware
that
have
a
firmware
update
available.
E
E
A
E
Local
storage
modes
in
plugins
in
kubernetes
do
that
they
use
lvm
to
create
a
pool
of
storage
for
approach
and
continuous.
So
if
you
create
ldm
devices,
it's
actually
right
now,
they're
just
created
there
and
they're,
not
particularly
managed
and
particularly
listed
anywhere
with
this
chain.
Here
we're
going
to
introduce
an
option
of
going
to
enable
an
option
in
the
event
configuration
that
essentially
has
makes
a
distinction
between
manage
and
manage.
Lvn
blocks,
editing
devices,
and
so
that
changes
slightly
to
the
behavior
so
that
it
works
better
with
kubernetes.
E
So
this
has
potentially
an
impact
if
or
you
already
using
lvm
and
you
we
can
have
to
make
sure
that
this
channel
doesn't
break
your
current
existing
lvm
setup,
so
we're
trying
to
do
likely
to
do
a
migration
script
or
something
to
make
sure
that
existing
system
that
they
are
already
using
and
then
keep
working,
but
as
always
help
with
testing.
This
would
be
greatly
appreciated.
E
E
B
Right
that
seems
pretty
straightforward
and
now,
let's
move
on
to
any
S
cost
updates.
I,
don't
know
Timothy.
If
you
have
anything,
Christian
isn't
here
anything
in
terms
of
yeah
right
yeah
it's
summer,
so
everything
is
quiet.
Yeah.
C
If
there
is
sorry
there
is,
one
thing
that
we
have
added
to
our
build
pipeline
is
a
layering
that
that
Christian
has
created
a
PR
for
so
we
build
the
so
it
splits
the
the
build
into
s-cos
base,
and
then
we
add
in
the
okd
layer
afterwards.
So
there's
this
notion
of
a
layered
approach
build.
C
We
actually
should
have
mentioned
that
that
we've
implemented
that
the
pr
have
been
implemented
in
a
popular
I'll.
Add
the
pr
in
in
the
pipeline
section.
Thank.
E
B
Yeah,
that's
true,
you
weren't
at
the
last
one.
Well,
basically,
we
went
through
the
cards
and
there
was
a
little
more
discussion
and
actually
it
brings
us
down
to
the
to
the
item
after
this.
But
basically
there
was
a
little
bit
more
discussion
on
doing
sessions.
B
That
would
allow
us
to
sort
of
bootstrap
community
members
into
supporting
okd
and
not
everything
falling
on
the
shoulders
of
the
deem
basically
in
the
community,
and
we
had
broken
it
down
previously
in
discussion
to
fcos
MCO
and
must
gather
being
the
three
things
for
people
to
really
understand
to
be
able
to
provide
community
help.
B
So
a
little
bit
of
discussion
at
the
community
development
last
week.
It
came
up
that
it
would
be
good
if
these
were
half
an
hour
sessions,
a
piece
for
each
topic
pre-recorded
and
then
that
would
allow
us
time
to
to
find
out
if
people
need
to
know
more
and
then
do
like
a
live
session
that
we
invite
community
members
to
if
there's
something
that
needs
to
be
filled
in.
B
If
there's
you
know
anything
that
that
they
that's
not
appearance
in
the
videos
that
would
be
pre-created
Mike
thinks
that
he
can
get
someone
from
the
MCO
team
Timothy.
Would
you
be
willing
to
do
like?
Maybe
you
know
20
minutes,
you
know
half
an
hour,
video,
that's
just
demonstrating.
B
You
know
F
costs
and
sort
of
how
it
fits
into
the
overall
okb
picture
and
I
know
that
there
are
some
tweaks
that
get
made
and
obviously
we'd
have
Vadim
I.
Think
chip
in
on
that.
So
maybe
it
would
be
something
that
you
and
Vivian
could
work
on
together
to
just
provide
something
that
would
be
like.
Okay,
this
is
f
cost,
and
then
this
is
how
it
cost
gets
tweaked.
You
know
in
a
in
an
okd
install.
E
Yeah
we're
cool,
we
think
we
could
do
that.
Or
do
you
have
a
better
idea?
Is
there
is
a
kind
of
like
a
script
or
like
a
list
of
things,
people
who
want
us
to
talk
about
and.
B
B
Sure
to
give
you
give
you
some
Foundation
stuff,
I
think
what
we
could
do
is
over
the
next
week
or
two.
The
Community
Development
Group
could
go
through
slack,
go
through
the
working
group
mail
list
and
try
to
get
a
sense
for
what
types
of
questions
people
are
asking
and
things
that
people
would
need
to
know
to
understand.
How
have
cost
fits
into
okay,
so
I
think
we
can
get
back
to
you
with
a
list
of
bullet
points
that
need
to
be
addressed
for
for
people's
understanding.
E
E
B
Less
yeah,
that's
true,
I
mean
there
is
a
lot
of
stuff.
That's
already
been
done.
You
know
the
past
like
what
five
years
or
whatever
videos.
If
would
you
be
able
to
maybe
collate
the
top
three
presentations
that
are
out
there?
If
you
want
us
to
take
a
look
at
those
and
see
if
those
are
sufficient
enough,
you
know
maybe
just
drop
some
links
into
the
into
the
meeting
notes
sometime
in
the
next
two
weeks.
B
You
know
or
a
week
or
so,
and
then
you
know,
let
us
know
which
youth
which
presentations
that
you
think
are
best
representative
or
best,
showing
I
guess
of
the
of
the
core
ideas
so
to
speak.
E
You
know,
could
we
pay
some
links,
I
have
and
but
yeah
it's.
E
For
sure,
but
I
I
think
it's
fine.
If,
if
we
have,
if
we
have
like
a
list
of
things
that
that
we
want
to
talk
about,
then
it
helps
us
create
something
together
for
that,
and
actually
you
know,
let
me
speak
just
about
those
topics
and
and
yeah
we'll
make
a
short
video
about
those
that
works.
For
me.
B
F
No
I
actually
did
watch
the
what's
the
playback
that
you
put
up,
I,
think
that
makes
sense
having
a
number
of
smaller
ones,
but
I
think
it
is
the
whole
process
of
understanding
core
OS
the
ignition
system.
It's
it's
immutable.
You
can't
go
in
and
just
edit
it
so
understanding
what
it
is
about
core,
OS
and
then
I.
F
F
I
I
mean
the
one
thing
I
would
add
on
to
that
is:
are
there
any
significant
differences
between
s-cos
and
fcos,
because
we've
got
the
two
difference?
We've
got
the
two
different
platforms.
So
just
if
there
are
any
differences,
I
know
we
build
them
very
differently
today,
and
one's
done
proud,
internal
ones
done
pipeline
sort
of
on
the
MC
with
mass
Cloud.
D
B
Right
so
this
sounds
good
and
then
I
think
we'll
have
Vadim
ship
in
maybe
on
must
gather
with
like
if
Mike,
because
he
says
he's
going
to
try
to
get
around
to
it,
but
he
also
I
know
has
time
constraints
as
well,
but
if
we
can
maybe
have
people
work
together,
they
might
maybe
could
record
a
zoom
session
between
the
two
of
them
or
something
like
that.
That
might
be
an
efficient
way
to
get
these
collaborative
ones
done,
and
that
way
it
doesn't
fall
all
on
someone's
plate.
B
Okay,
I'm
trying
to
think
if
there's
anything
else
from
the
community
development
meeting,
that's
relevant.
Other
than
survey
is
moving
forward.
We're
getting
responses,
I
think
we're
close
to
like
20
responses
now,
which
doesn't
sound
like
a
lot,
but
please
promote
it
to
folks
and
we'll
try
to
get
people.
You
know
filling
that
out
so
that
we
can
get
squared
away
for
for
direction
and
understanding
I
did
want
to
Riff
on
something
I
think
that's
it
from
Community
Development
updates
any
questions
on
any
of
the
Community
Development
stuff.
B
Okay,
one
thing
I
wanted
to
throw
on
before
we
end.
The
meeting
is
multiple
release
streams,
so
this
has
come
up
in
a
variety
of
contexts
over
the
past
couple
years,
we've
said:
well,
we
just
because
of
resources.
We
only
support.
You
know
the
the
current
latest,
basically
of
ocp
and
Vadim.
Actually,
in
a
conversation
with
someone
in
slack
the
other
day
said
well.
If
we
had
Cincinnati
support,
then
we
could
actually
have
a
graph
with
multiple
streams
and
we
could
support
like
a
13
and
a
14.
B
Is
there
any
desire
to
open
this
Pandora's
Box
can
of
worms?
However,
you
want
to
phrase
it
we
would
have
to.
You
know:
there's
links
there
for
the
technical
background
to
Cincinnati,
and
it's
really
do
we
want
to
go
there.
Do
we
want
to
go
to
you
know
supporting
two
releases
and
again
Community
Support
right,
not
even
official
support.
This
is
community
help
and
and
dedicated
Community
Resources
to
multiple
major,
really
or
multiple
minor
releases.
Us.
E
E
Version
one:
if
you
want
to
go
to
version
two,
it
tells
you
hey
do
then
those
versions
are
okay
to
update
to
version
two,
and
so
it
tells
you.
How
do
you
update
conversions
and
so
every
time
you
make
a
new
version,
you
have
to
decide:
hey
I'm,
not
going
to
refer
this
to
the
previous
stream
or
node
as
it
working.
If
the
is
the
update
working,
something
like
that,
so
this
is
work
that
somebody
has
to
do
to
maintain
this
graph.
This
is
domain.
F
I
I
guess
the
other
thing
Jamie.
Are
you
looking
at
F
cost
based
s-cost
base
or
both
of
them
because
again,
I
think
the
for
the
F
costs
release.
That's
all
going
to
fall
on
probably
radim
or
Christian,
because
daily
noise
that
can
actually
make
the
releases
and
I
don't
know
Luigi
Alex.
Do
we
have
any
any
sort
of
idea
or
concept
of
multiple
sort
of
streams
of
the
pipeline
with
different
against
different
branches
or
anything?
F
B
Well,
and
just
to
be
clear,
that's
what
Vadim
actually
said
was
he
was
like
if
you
want
to
volunteer,
because
it
was
one
of
our
regulars
was
actually
asking
about
it.
It
was
Ryan,
Monroe
and
Vadim
was
saying.
Well,
if
you
want
to
volunteer
to
get
this
going
and
support
it
and
help
support
it
in
the
community,
then
then
we
could
do
this,
but
otherwise
you
know
we're
at
where
we're
at.
B
E
C
C
and
then
obviously
there
are
no
Z
streams,
so
yeah
I
don't
know
if
the
Cincinnati
graph
will
be
that
complicated.
But,
like
you
say,
Brian,
someone
has
to
maintain
that
someone
has
to
make
sure
it's
right
and
that
you
have
some
form
of
upgrade
path.
If
you're
going
from
413,
let's
say
for
15
comes
out,
you
need
to
maintain
that
upgrade
path
and
and
make
sure
the
edges
are.
F
Because
because
I
mean,
as
we
look
at
the
previous
versions,
We
Don't
backboard
everything,
so
we
Cherry
picks
certain
things
which
then
was
going
to
add
even
more
work
to
it,
because
it's
not
just
taking
sort
of
like
the
headstream
or
the
dev
stream.
You've
got
to
cherry
pick
stuff
in
which
may
have
dependency.
I
can
just
see.
There
is
quite
a
lot
of
work
to
actually
back
ports
fixes
into
previous
versions.
F
C
Well,
internally,
for
the
release
for
Red
Hat
for
Red
Hat
releases,
there's
an
internal
Cincinnati
graph
server
and
that
is
totally
red
hat
within
the
red
hat
VPN.
No
one
will
be
able
to
access
it.
So
I,
don't
know
if
there's
a
Reaper
available
for
that
and
I
suppose.
The
other
question
is
so
and
I'm
talking
now
I'm
talking
from
a
red
hat
perspective,
okay,
Red
Hat
will
say
guys.
If
you
want
a
okay
Dean
production,
then
just
use
red
hat
use
RCP
create
some
subscriptions
buy
the
product.
B
But
that's,
but
that
actually
is
what
the
the
framing
that
this
came
out
of.
Was
someone
actually
saying
we
don't
have
the
money
to
buy
ocp
because
it
costs
you
know
x,
amount
of
dollars
and
I
think
it
was
the
estimated
out
that
it's
like
half
of
the
cost
of
their
Hardware
every
year.
Right,
so
you
know,
okv
sounds
palatable
until
they
start
thinking
about
those
operators
right.
C
If
they
have
internal,
so
internal
sres
and
and
and
support
for
that,
then
it
does
make
sense.
Then
then,
I
think
that
the
work
that
we
or
the
drive
that
we're
going
through
the
initiative
or
operators
could
help
towards
that
and.
C
A
E
A
So
we
already
kind
of
have
our
own
graphs
of
it's
just
nobody's
actively
maintaining
that
graph
data.
It's
just
a
copy
of
what
what
gets
built.
E
F
That
I
I
guess
the
thing
of
that
is.
We
only
have
a
single
stream
and
like
ocp,
we
don't
have
multiple
streams,
we
don't
have
fast,
we
don't
have
sort
of
a
stable,
we
don't
have
and
we
just
have
the
single
stream.
You
know:
okay,
the
so
the
graph
is
really
just
that
single
stream
and
I'm
guessing
that's
part
of
the
build
process
where
we
just
do.
The
updates.
F
I
I
know
that
we
do
run
some
testing,
because
you
can't
see
that
but
I'm
guessing
that's
part
of
what
happens
in
prowl
for
the
F
cost
version.
What
happens
in
a
pipeline
for
we'll
just
do
that?
We'll
actually
just
create
the
update
path.
F
B
All
right,
well,
I,
just
want
to
get
the
conversation
going.
I
I
the
dean
brought
a
good
point,
which
is:
if
folks
we
can
get
Community
Support.
Then
we
can
support
multiple
versions.
It
would
be
great
to
see
okd
used
in
production
more
and
I
think
that
not
always
being
on
the
edge
like
that
would
be
helpful
for
folks
to
embrace
it
for
production.
What
reasons
but.
F
But
I
do
think
it.
It
raises
the
question
who
is
going
to
decide
what
gets
back
talks
is
to
the
previous
versions,
because
we
don't
have
that
function
anywhere
within
okay
at
the
minute.
I
know,
ocp
does
because
they
support
for
18
months
and
the
longer
versions,
but
we
don't
have
that
notion,
so
somebody
will
have
to
work
out
for
each
fix.
E
I
think
it's
it's
slightly
different
than
that
in
the
sense
that
we
get
the
fix
up
kit
by
putting
too
so
people
get
them
into
a
video.
So
essentially
the
the
version
2
would
get
all
the
fix
from
openshift
Also.
Let's
take
real
numbers,
4
14
412,
all
the
412
bills
will
get
all
the
backboards
for
for
12.
and
they
will
happen
in
the
next
bills
and
they
will
blend
in
that
stream
same
thing
for
4
13
and
they
were
done
there.
So.
B
B
Yeah
all
right,
well
and
I
think
that
this
boils
down
to
volunteers
right,
and
that
was
another
thing.
That's
come
out
of
the
community,
particularly
in
the
last
meeting.
Is
we
really
need
to
make
a
push
for
volunteers,
and
if
someone
is
interested
in
volunteering,
this
is
a
great
opportunity
right
and
sort
of
a
great
issue.
I
think
to
to
get.
We
would
need
a
handful
of
volunteers,
I
think,
but
if
people
are
really
interested
in
getting
this
in
production
and
and
getting
that
multi
minor
version
support
we
or
it
releases
I
should
say.
B
E
Yeah
this
is
you
know.
This
is
the
whole
point
about
how
I
felt
around
the
okt
streams,
so
the
streams
with
with
an
S
as
that.
Essentially
it's
much
easier
for
anybody
to
post
their
own
Pipeline
and
do
their
own
start
doing
their
own
builds.
And
so
because
we
published
the
whole
thing
as
a
detect
on
the
pipeline
and
you
can
sell
Force
to
run
so
we
could
have
community
members
taking
over
builds
for
older
editions
of
the
okd
discuss
and
hosts
themselves
on
their
own
infra
or
somewhere.
E
B
All
right
any
last
thoughts
as
we've
made
it
through
the
agenda
and
we
can
close
the
meeting
anything
else
that
folks
want
to
talk
about.
B
All
right
well,
in
that
case
you
want
to
say
goodbye.
No
all
right
in
that
case
I
think
we're
good,
and
so
we
will
talk
about
this.
This
particular
item
and
also
the
videos
I
think
at
the
community
development
meeting.
B
B
B
A
Anything
I
personally
find
them
useful,
as
I
haven't
been
able
to
attend.
The
last
couple
of
community
meets
regularly,
so
I've
been
going
up
and
catching
up
on
the
videos.
Okay,.
B
F
B
B
B
B
Just
gonna
go
through
the
board
again
and
I
think
talk
about
these
videos
and
and
sort
of
nail
down
time
frames
and
the
suggested
things
in
terms
of
f
costs.
What
some
bullet
points
that
we
think
that
folks
would
definitely
want
to
know
about
in
terms
of
understanding.
B
You
know
F
costs
And
in
regards
to
okd
yeah,
so
because
it's
summertime
and
a
lot
of
folks
are
sort
of
scattered
right
now
so,
and
you
know
it
was
pretty
slim
meeting
last
meeting
last
couple
of
both
meetings
actually
been
kind
of
slim
attendance,
so
yeah
all
right.
Well,
thanks!
So
much
folks
appreciate
your
time,
yeah
all
right
thanks
so
much,
and
we
will
talk
to
you
next
time.
Bye.