►
From YouTube: OKD Working Group Meeting 05-09-2023
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
to
the
okd
working
group
meeting
for
May
9th
of
the
Year
2023
and
we've
got
the
agenda
in
the
chat
and
take
about
30
seconds
to
take
a
look.
Let
me
know
if
there's
anything
else,
that
we
need
to
add
change,
modify,
move
around.
B
Okay,
let's
start
out
with
okd
release
and
cicd
updates
with
Christian
welcome
back
Christian.
C
Hey
everybody,
so
we
have
new,
okay
on
S
cross
releases
and
yeah,
quite
a
swath
of
of
them.
Actually,
we've
been
doing.
The
team
that
Shireen
is
in
has
been
doing
an
excellent
job
in
doing
sprinkly
releases.
Now
so
with
okd
and
escals.
We
can
expect,
which
is
three
weekly
releases
now,
for
both
our
stable
stream,
which
is
part
of
12,
and
also
the
next
stream,
which
is
currently
Fallout,
13.
C
and
once
followed.
D
C
Our
next
stream
will
likely
then
pivot
to
Fallout
14.,
yeah
I,
don't
think
we
have
had
another
okb
on
fcos
released
since
the
last
working
group
meeting.
So
I
can't
comment
on
that
yeah!
That's
it
for
me.
B
Nope,
okay,
moving
on
now
to
fcos
updates,
Timothy
can't
be
here,
so
we
don't
really
have
any
updates
for
that.
So,
let's
move
along
well
overall,
as
cost
updates
anything
anything
else
that
we
should
know
just
in
general,.
D
C
With
Timothy
here
in
Hamburg
last
week,
he's
on
on
holidays,
and
so
there
are
some
exciting
plans
in
in
corowa
Westland
that
we
will
also
I
think
try
to
to
use,
or
at
least
to
to
experiment
with
in
in
okd,
which
is
they
are
going
to
essentially
split
the
base
OS
and
that
that
is
less
F
cost
and
more
our
cause,
and
and
also
s-cos
they're.
C
Gonna
split
that
into
a
core
that
essentially
doesn't
carry
any
of
the
kubernetes
and
openshift
or
okd
dependencies,
and
then
layer
on
top
the
things
like
the
cubelet
cryo
and
those
things
that
are
essentially
bound
to
a
specific
kubernetes
version
and
with
that
and
and
those
different.
C
C
And
things
like
iot
comes
to
mind
and
certainly
other
as
well,
so
we
want
to
do
that
in
in
s-cos,
in
order
to
also
have
have
clear
ownership
of
the
base,
which
will
be
well,
it'll
still
be
us
in
the
working
group
and
the
okay
screams
team,
but
we
want
to
make
that
an
official
Centos
artifact
this
core
Central
stream
core
OS,
and
then
we
as
the
the
okay
working
group,
can
essentially
layer
of
the
openshift
clients,
RPM
the
openshift,
hyper
Cube,
RPM
and
cryo
on
top
and
test
that
together
for
for
each
version
of
kubernetes
and
still
deliver
all
the
different
streams
into
different
versions
and
I.
C
Think
having
that
split
kind
of
supported
by
the
the
team
that
that
maintains,
Fedora,
coros
and
also
Rel
coros,
which
is
Timothy's
team,
is
going
to
help
us
achieve
that
way
way
more
quickly.
I
think
now
that
they
are
essentially
doing
the
same
there,
so
I
think
that's
something
to
look
forward
to
there's
also
new
tooling,
coming
in
from
from
the
ROI
side,
there's
boot
c
and
things
they
are
experimenting
with
for
for
image.
C
Creation
which
boot
C
can
I
think
could,
in
in
that
case,
replace
RPM
or
S3
eventually,
so
that
that
is
certainly
something
we
will
also
experiment
with,
not
not
immediately,
but
that
is
kind
of
on
the
on
the
radar.
If
there's
any
interest
in
the
community
to
to
help
with
that-
or
you
know
just
if
you're
interested
in
that,
let
me
know
and
then
finally
on
as
a
kind
of
maybe
I'm
getting
ahead
here
on
on
the
agenda
as
a
quick
update
on
on
S
cross.
C
Also,
we
have
now
wow
yeah.
We
are
in
the
process
of
deploying
a
cluster
to
Moc
Massachusetts,
open
Cloud,
and
so
our
current
build
system
on
operate
first
is
essentially
unmaintained.
Now,
so
it's
starting
to
Decay
a
little
bit.
We,
we
don't
have
anybody
working
on
on
the
maintenance
for
that
cluster
anymore
and
now
CI
is
broken,
I,
don't
think
anybody's
there
to
fix
it.
So
we.
D
C
Working
on
on
setting
up
a
new
shop
on
on
Massachusetts,
open
Cloud
and
having
our
own
cluster
there
as
a
build
Farm,
which
we
will
then
one
of
these
our
build
Farm
cluster,
we
will
onboard
into
Centos
infrastructure
as
kind
of
an
official
builds
as
again
making
it
part
of
the
of
the
official
centers
infrastructure
and
that
way
enabling
us
to
do
these
official
artifact
builds
there,
and
there
is
still
interest
from
from
a
few
other
folks
to
also
have
have
their
own
clusters,
so
we'll
likely
also
have
a
secondary
cluster
for
kind
of
deploying
Community
projects.
C
That
is
something
they
operate
first
folks
would
like,
and
they
are
in
turn
going
to
help
us
with
with
Administration
or
the
former
operator,
for
instance,
I
should
say
with
administration
of
these
build
farms.
I
think
that's
a
good
deal
and
it'll.
D
B
A
E
Okay,
so
we
had
our
meeting
last
week,
there's
a
link
in
the
pack
MD
with
the
planning
board,
so
I
said
we're
now
being
driven
by
cards
on
that
board.
So
we
went
through
all
the
work
in
progress.
We
now
have
Luigi's
detailed
single
node
install,
so
that's
doing
it
manually
using
sort
of
building
it
up
slowly
so
that
that's
there
now
I'm
I
need
some
help
to
get
the
assistant
installer.
E
Neither
the
s-cos
or
the
f-cos
version
works
running
exactly
the
same
commands
that
the
red
hat
hosted,
one
does
for
ocp,
so
it's
either.
The
assistant
installer
isn't
quite
compatible
with
what
we're
doing
with
okd
or
we're
building
the
image
of
a
slightly
different,
but
neither
cluster
works.
The
s-cost
doesn't
even
get
to
the
rotate
and
where
the
F
cost
does
bring
up
a
cluster.
But
a
lot
of
The
Operators
are
just
broken.
E
I've
actually
documented
in
the
in
the
issue.
I've
posted
that
into
the
chat
here.
So
it
is
issue
number
two
on
the
planning
board.
It's
in
the
in
progress.
There
is
a
muscular
log
for
the
f-cos.
So
please.
If
anybody
has
time
ping
me
on
I'm
on
Slack,
either
in
the
open
shift
users
or
the
kubernetes
users
and
I'd
like
to
get
that
up
so
we've
then
got
single
node,
install
and
documentation
available.
E
E
So
next
week
we're
going
to
look
at
the
home
page
and
try
and
refactor
the
home
page
and
the
week
after
we're
going
to
go
back
to
just
reviewing
all
of
the
items
on
the
planning
board
and
then
the
week
after,
if
people
can
attend,
we
want
to
do
a
kickoff
for
the
operators,
the
okd
operator,
catalog,
so
again,
I'm
hoping,
maybe
Christy
and
Luigi.
We
can
get
you
in
on
that
one,
because
I
think
it's
important
that
we
do
have
some
of
the
technical
Red
Hat
folks
on
that
one.
E
So
next
week
is
going
to
be
the
home
page
for
okd.io
week
after
we'll
do
a
normal
go
go
through
the
board
and
then
the
week
after
will
be
the
operators
kickoff
session
I.
Don't
think,
there's
really
anything
more
to
say,
because
I
notice
you've
got
the
other
item.
As
the
next
item
on
the
agenda
that
we
discussed
the
the
survey
for
the
the
time
change.
B
Yeah,
the
only
thing
that
I
would
add
is
so
here's
the
link
I'm
putting
this
in
the
in
the
chat
there.
That's
an
issue
that
I
opened
that
has
the
2023
survey.
If
anyone
has
any
comments
of
questions
that
you'd
like
to
see,
I
updated
it
a
little
bit,
but
if
folks
have
any
other
questions
that
they
think
would
be
helpful,
as
we
put
this
out
just
add
a
comment
in
that
issue.
B
You
know
the
working
group
and
and
give
us
a
little
bit
of
guidance
and
whatnot
yeah,
and
then
that,
as
speaking
of
of
surveys,
that
leads
us
to
the
survey
for
the
time
change
possible
time
change.
Well,
it
turns
out
we
have
17
responses.
B
The
majority
of
the
responses
are
to
are
vote
to
move
it
back
permanently
one
hour
and
we
did
get
some
suggested
feedback
like
don't
do
it
on
Tuesdays
for
the
other
option
we
had
just
a
handful
of
others.
Some
were
to
do
it
even
earlier.
B
One
was
to
keep
it
at
4
pm
UTC,
which
would
be
an
hour
earlier,
but
would
change
obviously
for
everyone
continuously
over
time,
so
the
majority
appears
to
be
having
this
be,
for
example,
it
would
be
noon
Eastern,
always
for
me,
Christian
does
that
work
for
you.
We
haven't
had
a
chance
to
talk
to
Luigi
yet,
but
so
it
would
be
an
hour
earlier,
like
all
the
time
basically
than
it
is
currently.
D
For
me
and
for
Alex
I
think
you
works,
don't
know
for
Luigi,
but
probably
as
well,
because
we
are
sharing
a
lot
of
the
meetings.
These
days
sure.
E
Yeah
I
I
think
when
we
had
the
conversation
when
Luigi
was
here,
he
wanted
it
to
go
back
so
so
we're
going
to
follow.
Daylight,
Saving,
basically
I
know
it's
going
to
be
a
two
week
when
we're
out
of
sync
Europe
America,
but
basically
we're
going
to
follow
daylight
savings.
So
we're
going
to
go
back
an
hour
which
is
going
to
be
what
six
o'clock
a
year
or
five
o'clock
UK
midday
noon,
Eastern
and
then
we're
going
to
stay
basically
at
that
time.
B
B
Does
that
Define
to
do
that
next
week?
All
right,
so
hereby
next
meeting
will
be
one
hour
earlier
and
we'll
follow
this
process
of
just
keeping
it
at
that
time
and
deal
with
the
the
slight
bit
of
confusion
for
that
that
brief
period
and
between
the
time
changes,
but
all
right.
So
I'll
update
the
calendars,
probably
tomorrow
or
later
today
and
I'll,
send
something
out
on
the
working
group.
B
Google
group,
email
and
I'll
also
post
something
in
the
slack
channel
so
that
it's
it's
visible
to
everybody
all
right.
Is
there
anything
else?
We
need
to
talk
about,
we
kind
of
whipped
through
today's
agenda
since
we're
missing
a
couple
of
votes
and
that
might
actually
end
up
helping
I
know
you
know
Timothy's
on
vacation,
but
I
know
other
folks
haven't
been
able
to
attend
because
of
this
sort
of
later
hour.
So
maybe
that
we'll
get
more
attendance
back
in
the
meetings
by
moving
in.
C
There
is
two
things
actually
Brian
since
you
tried
to
get
the
assistant
installer
working,
I,
think
internally.
We
are
also
looking
at
that
specifically
for
for
s-cross
but
yeah.
Maybe
we
can
kind
of
invite
you
to
to
our
work
and
just
you
know
be
prepared
if
you're
open
to
that
to
get
an
invite.
E
C
Okay
and
then,
secondly,
there
is
a
shift
week
coming
up
for
us
internally
at
in
openshift
engineering,
meaning
there
is
one
week
where
the
engineers
and
it's
I
think
mid-june
I
think
it's
the
week
of
defconf
actually
so
well.
Many
of
us
won't
actually
be
there
of
this
group
at
least
Korean
Alex
Bruno.
C
But
there
is
kind
of
this
we
want
to
drive
as
many
of
our
internal
redhead
folks
to
to
okd
and
there's
essentially
two
two
onboarding
things.
There
is
one
of
the
internal
kind
of
onboarding
I
want
to.
You
know,
really
change
a
part
of
of
openshift
or
okd
and
want
to
want
to
get
a
build
out
of
that.
How
do
I
do
that
and
then
there
is
kind
of
the
people
just
want
to
try?
C
Okay,
you
know
they
just
want
to
install
and
instead
of
kind
of
providing
another
guide
for
that.
We
essentially
because
there
is
going
to
be
both
of
those
groups.
We
will
have
some
some
internal
docs
for
what
what
people
can
maybe
play
with
and
then
for
those
who
just
want
to
install-
and
you
know
give
it
a
try.
C
We
really
want
to
point
them
to
the
working
group
and
also
to
just
you
know,
okd.io
and
yeah,
hopefully
providing
kind
of
an
installed
guide
for
specifically
okd
on
s-course,
so
you
know
running
up
to
to
that
it's
in
a
month,
so
we'll
we'll
likely
kind
of
want
to
add
something
to
the
to
the
website.
C
C
Yeah
and
I
I
haven't
checked
the
okd.io
docs,
but
I
am
sure.
There's
some
onboarding
there
and
we
will
kind
of
want
to
tweak
it
a
little
bit
to
point
two
to
ask
us.
E
Some
of
that,
some
of
that
we'll
be
looking
at
next
week,
just
in
terms
of
how
how
does
the
site
present
itself
to
a
new
user
and
then
what
is
the
flow
but
also
I,
think
if
we
can
get
I
mean,
are
you
looking
at
full-blown
clusters
or
single
node
clusters
or
any.
C
Of
everything
I
mean
those
folks
will
have.
You
know
the
the
internal
rethaters
first,
if
we
will
have
access
to.
E
There
have
been
a
couple
of
bugs
flying
around
or
discussions
flying
around
about
gcp,
so
that
might
be
one
that
we
I.
That's
a
cloud
I
don't
use,
but
I
think
there
were
some
issues
around
the
configuration
on
gcp
I've
seen
a
few
things
being
raised
in
the
last
couple
of
weeks.
So.
C
Okay,
yeah
and
then
we
definitely
also
want
to
and
and
there's
gonna
be
interest
for
for
Sno
and
agent-based
driven
installed.
So
that
is
something
we
we
also
want
to
get
working
for.
All
of
our
you
know
streams
that
we
have,
of
course,
as
well
as
s-cross
yeah.
So
hopefully
we
will
kind
of
get
that
done
before
shift
week,
so
we
can
really
have
people
just
you
know,
trying
trying
it
all
out
for
us.
A
Hi,
my
name
is
Andrea
fazano,
probably
we
didn't
met
before
I
work
on
the
agent-based
installer
I
joined
this
team,
because
I
was
interested
also
and
collaborating
with
Sharon
could
now.
A
I
was
interested
also
in
in
enabling
it
for
deploying
okd
right
now,
I
think
internally,
its
course
could
be
it's
something
that
seems
to
be
to
be
working
and
I.
Think
maybe
the
agent-based
installer
could
be
interesting,
given
the
fact
that
it's
particularly
easy
to
to
set
up
to
deploy
I
think
so
the
app
you
also
to
help
on
that
side.
A
C
Awesome
welcome
to
the
group
Andrea.
Thank
you,
yeah
I
think
we'll
definitely
use
you
and
the
contact
to
your
team
to
get
this
done
and
I
know
that
Shireen
has
already
yeah
talked
to
you.
So
welcome
to
the
working
group.
Two,
it's
good
to
see
you
here.
B
Excellent,
oh
and
while
we
talk
about
access
or
I,
should
access
for
people
to
get
into
the
okd
world
I'm
going
to
highlight
that
the
working
group
is
looking
for
someone
to
help
with
CRC
folks
within
Red.
Hat
have
been
generously
sort
of
providing
time
to
keep
things
updated
so
that
that
can
run,
but
there's
a
little
bit
of
work.
That
needs
to
be
done
on
a
continuous
basis
to
keep
that
going
and
they're.
Looking
for
someone
in
the
community
to
volunteer.
A
B
E
E
Sure
yeah
yeah
is
it
possible
to
have
someone
Community,
because
my
thoughts
is
what
we
really
need
to
do
is
have
a
pipeline
F,
cross
and
s-cos.
So
when
there's
a
build,
it
kicks
off
a
pipeline
to
build
the
new
CRC
artifacts.
So
that's
where
we
want
to
get
to,
but
so
the
new
the
new
build
infrastructure
would
be
useful
there
as
well.
B
C
B
No,
it's
pretty
straightforward
because
they
last
year
I
think
it
was.
They
did
some
work
basically
to
sort
of
module
modularize
it.
So
it
is
a
much
more
straightforward
process,
but
there's
still
the
issue
of
I
think
it's
updating
the
certificate
in
it,
and
so
that's
like
the
one
thing.
That's
that's
like
a
manual
task
that,
ideally
someone
would
take
over
here.
You
know
is
update
and
this
reassemble
you
know
and
as
an
as
an
input.
What
what
does
a
CRC
Bill.
B
C
Yeah
we
couldn't
already
like
put
that
on
on
operate
first,
even
for
for
the
time
being,
you
know
if,
if
somebody
were
to
create
a
a
tact
on
task
that
does
you
know
whatever
is
needed,
we
could
easily
deploy
it
so
definitely
yeah,
yeah,
very
cool,
definitely
something
to
kind
of
keep
in
mind.
We
can
all
of
these
kind
of
artifacts
that
are
derived
from
openshift
or
you
know.
C
Okay,
we
can
easily
make
that
because
we
have
access
to
the
well
to
our
releases
and
our
release
control
controller.
So
we
can
yeah
just
we
I
I'm
not
familiar
with
the
CRC
build
process,
so
yeah.
B
I
have
a
ticket
to
explore
it
more.
So
it's
that's
actually
on
my
agenda
for
the
next
week.
So
the
other
thing
I
wanted
to
highlight
is
that
we
do
have
the
in
The
Matrix
Channel
Shireen
did
the
awesome
work
of
getting
the
okd
bot
set
up.
B
So
if
anyone
watching
this
video
wants
to
see
build
results
and
where
they're
you
know
the
success
or
failure
on
these
different
builds
and
sort
of
watch
the
process,
you
can
go
to
the
okd
channel
in
our
Matrix
channel
and
watch
that
all
unfold
there.
So
no
details
are
on
the
okay
website
and
then
I'm
getting
notes
all
right.
Anything
else.
B
All
right,
I
think
we're
good
okay,
so
not
the
same
bet
time
next
week,
a
hour
earlier,
but
same
bat,
Channel
and
thanks
so
much
folks
for
participating
and
we'll.