►
From YouTube: OKD Working Group Meeting 05-23-2023
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
23rd
of
the
Year
2023,
what's
gaze
at
the
agenda
for
a
few
seconds,
let
us
know
if
there's
anything
that
is
missing
or
that
you'd
like
to
move
around
or.
B
All
right,
everyone
good
on
the
agenda
right
I,
see
a
lot
of
Thumbs
Up
nodding
heads.
Then,
let's
jump
into
it.
Okay,
releasing
cicd
updates,
Luigi,
I,
think
you're.
The
only
one
here
so
take.
D
It
away
yeah
yeah,
yeah,
yeah,
yeah,
so
yeah
we
had
a
discussion
early
on
and
with
our
engineering
manager.
D
So
we
are
looking
at
building
four
4.14
for
the
okd.next
version
and
4.13
for
our
stable
version
because
of
the
well
basically
we're
in
the
throes
of
having
413
G8.
So
that
was
our
thinking
anyway.
I
still
have
to
just
confirm,
with
Christian
I've,
been
aware
on
PTO,
so,
but
that
was
the
the
general
thinking.
A
B
And
anything
yeah
Neil
did
you
have
a
question
or
just
saying.
E
Yeah
yeah
I
have
a
couple
of
questions.
Actually,
the
first
is
so
with
opd413
coming
down
the
pipeline,
so
I
saw
with
the
openshift
413
talk
that
we
are
going
to
have
C
group
V2
by
default
and
C
run
as
available
as
a
non-default
ga
for
whatever
that
means.
E
That
makes
me
think
that
we
we
actually
should
probably
go
a
step
further.
Okay,
you
know
KD.
E
Can
we
actually
like
kind
of
hit
a
home
run
here
and,
like
do
see
run
by
default,
for
for
at
the
very
minimum
new
okd
deployments,
like
we
did
for
secret
V2,
several
releases
black
I
think
it's
like
at
least
a
year
or
so
ago
that
we
did
that
because
Fedora
defaults
to
it,
and
we
wanted
to
make
sure
that
everything
worked
with
that
I
would
like
us
to
be
doing
all
the
cool
first
and
I
want
us
to
to
be
able
to
Pro
and
I
think
that's
an
Avenue
for
us
to
provide
value
to
to
the
ocp
team
as
well
and
yeah.
E
So
that's
kind
of
like
where
my
head
isn't
on
this
is
like
there's
a
couple
of
things
that
were
announced
with
openshift
413
that
are
like
either
preview
or
or
non-default
GA,
which
is
a
weird
State
I've
never
heard
of
before,
and
we
should
just
do
those
things
by
default
out
of
the
box
for
okd.
Can
we
do
that?
E
B
Technically,
if
at
a
technical
level,
if
we
can,
it
would
just
involve
obviously
the
working
group
taking
a
vote
to
say
yeah,
let's
do
this,
what
is
the
general
sense?
Let
me
well
first,
let
me
ask
Luigi:
if
you
are
you.
D
Yeah
yeah,
so
just
on
the
sea
run
thing
I
have
been
playing
around
with
it,
especially
for
wasm
for
running
Western
binaries
in
serum,
because
you
can
actually
compile
sea
run
with
the
with
wesm
enabled.
So
we
were
experimenting
with
that.
So
I
I
I'm,
with
Neil
here
I,
would
really
like
to
I
would
really
like
to
see
that
only
because
it's
it's
a
it's.
D
Yeah
and
and.
D
That
we
should
be
playing
with
so
I
I,
don't
know,
as
should
we
try
and
look
at
it
for
shift
week
and
I
mean
I
know.
This
is
a
very
red
hat
specific
thing,
so
we
might
exclude
a
lot
of
the
community
folk,
but
it
could
be
maybe
something
to
Pilot
and
then
drive
forward.
I,
don't
know
I'm
just
a
suggestion.
D
So
basically,
what
happens
is
open
so
in
the
open
shift,
the
operation
of
larger
Community,
all
the
devs
have
a
week
where
we
can
experiment
complete
week
to
do
whatever
you
want
to
do
on
related
to
openshift.
That's
why
they
call
it
shiftwick.
So
what
happened?
Last
year
we
played
around
with
the
pipelines
we
we
built
a
couple
of
pipelines
and
that
type
of
thing,
but.
G
D
E
F
E
Yeah
and
that
that's
like
part
of
the
motivation
here
is
because
introducing
making
a
Tri-State
for
GA
not
on
and
default,
not
is
implies
that
somebody
is
not
confident
about
our
about
the
tech
that
we
have
and
and
I
don't
share
in
that
same
feeling.
But
the
only
way
to
prove
it
is
to
do
it,
and
so
we
should.
F
This
is
also
something
that,
by
the
way,
I'm
Jeremy
Peterson
I'm
a
Staff
engineer.
Oh.
E
F
On
the
openshift
team,
I've
been
asked
to
Liaison
with
this
group
on
behalf
of
the
staff
Engineers.
The
staff
Engineers
have
asked
us
explicitly
to
do
this.
What
you're
asking
so
it's
a
good
idea
yay
for
my
approval.
C
C
G
Me
just
make
one
point:
the
reason
why
it
may
not
be
default
is
if
no
one's
been
using
it
and
you
said
it
to
default,
and
maybe
they
just
don't
want
to
break
anyone's
environment
by
then
bringing
this
in
and
exchanging
functionality.
It
may
not
be
that
they
don't
have
confidence
in
it.
They
just
don't
want
to
break
people
because
they're
not
already
using
it.
E
So
the
the
reason
so
I
thought
of
that
Amy
I
was
just
a
little
skeptical
of
the
argument
because
they
G
ate
it
just
not
bite
yeah.
E
G
H
H
Pipelines
and
s-cos
is
all
about
that.
We
can
create
a
a
shadow
release
that
has
this
on
by
default
and
we
want
people
to
play
with
it,
but
it
it's
then
the
consumer's
choice
on
whether
they
want
the
version
with
it
on
by
default,
or
they
want
to
be
a
little
bit
safer
and
have
diversion
With
It
Off
by
default,
but
it
but
I
thought
that
was
the
reason
we
had
this
pipeline
in
the
open.
E
The
the
amount
of
entry
questions
I
see
go
on
when
people
talk
about
prowl
makes
me
think
that
that
is
less
understood
than
people
think
it
is,
but
no
like
my
thought
process
around
this
is
that
if,
if
there
was
a
reason
for
someone
to
think
that
it
might
break
something
because
it
is
new
and
may
be
different,
maybe
they
were
burned
by
something
else.
E
Like
I
know
that
the
Pod
man
cni
change
to
net
of
Arc
broke
a
ton
of
container
workloads
and
I
know
that
that
spooked,
a
bunch
of
people
around
that
I,
don't
consider
the
change
of
run
C
to
C
run
to
be
in
the
same
class
of
potential
breakage,
but
the
only
way
to
really
demonstrate
that
is
to
do
it
and
prove
it
and
I
think
that
okd
is
the
place
that
we
should
be
doing.
That
sort
of
thing.
B
B
Can
we
maybe
formulate
a
async
discussion
in
the
discussion
section
about
benefits
and
downsides
and
then
from
there?
If
we
decide
if
the
group
decides
yes,
this
is
where
we
want
to
go
then
start
formulating
a
plan
for
how
we
promote
it.
How
we
make
you
know
how
the
technical
change
comes
about,
how
we
promote
it
and
what
not,
because
my
sense
is
that
the
the
general
tenor
is
that
we're
a
yes
on
this,
but
it
would
be
good
to
actually
document
this
in
a
discussion.
B
B
F
B
B
E
Exactly
so,
okay,
no
I
totally
get
it
like
go
ahead
and
like
make
a
discussion
thing
and
and
add
a
link
to
the
agenda
to
the
agenda,
and
then
we
can
then
I'll
we'll
part
we'll
throw
some
async
comments
and
stuff
in
there.
E
General
feeling
is
that
I
think
we
should
just
do
it
for
everything
and
then
just
see
what
happens
going
from
there,
because
I
suspect
that,
with
the
data
that
we
can
have
on
doing
this,
it
will
make
it
a
lot
easier
for
this
technology
to
show
up
sooner
across
the
entire
openshift
portfolio
and
actually
could
also
benefit
Upstream
kubernetes.
E
Folks
too,
because
I
don't
think
anyone
in
Upstream
kubernetes
has
seriously
thought
about
their
recommendations
of
how
the
container
interface,
the
the
container
runtime
interface
stack,
is
structured
in
God
knows
how
long,
and
so
it
would
be
very
useful,
I
think
for
us
to
leverage
our
experience
here
to
provide
some
useful
information
to
even
just
vanilla,
kubernetes
folks
Upstream,
because
I
think
that
that
is
definitely
something
we
can
provide
value
to
them
as
well.
F
One
of
the
Visions
for
okd
that's
been
communicated
to
me
is
for
okg
to
be
an
upstream
of
openshift
right
now.
The
Upstream
of
openshift
is
kubernetes
if
okd
were
really
to
become
the
Upstream
of
openshift.
That
would
imply
that
kubernetes
would
be
the
Upstream
of
okd
and
okd
would
be
the
Upstream
of
openshift,
meaning
that
all
of
the
features
of
kubernetes
would
be.
F
You
know
demoed
and
you
know
exercised
in
okt
first
and
so
we'd
be
getting
all
that
experience
in
OCD
first
and
you
know
experiencing
all
that
pain,
and
you
know
before
the
openshift
people
do.
This
is
obviously
a
massive
change
to
the
way
things
are
done
right
now,
but
maybe
it's
worthwhile.
E
E
There
was
just
no
real
appetite
to
getting
that
built
out
and
it
finally
started
happening
what
what
mid
last
year,
Jamie
and
and
we're
we're
starting
to
be
in
a
position
where
we
can
do
this
for
the
first
time
and
I'm
and
honestly
I
think
Cron's
going
to
be
a
great
Proving
Ground
for
this,
because
I
actually
expect
nothing
to
be
bad
to
happen,
and
that
means
that
if
something
bad
happens,
it'll
probably
be
very
mild
and
it'll
be
resolvable,
and
we
can
use
it
as
a
way
to
qualify
a
process
that
we
do
have
for
for
doing
this
further
down
the
road
with
much
more
devastating
or
much
more
invasive
features,
because
because
there's
this
is
not
going
to
be
the
end
of
the
road
there's
way
more
stuff
that
that's
going
to
go
on
in
the
future.
E
Like
things
around
I,
don't
know
if
anyone
saw
what
happened
to
Grand
Summit
today,
but
we
got
a
whole
bunch
of
new
things
coming
out
that
are
open
shift
Centric
and
not,
and
some
of
those
are
going
to
trickle
into
okd,
and
we
got
to
think
about
how
we
want
to
have
those
part
part
of
the
story
too,
and
and
then,
of
course,
there's
building
relationships
with
Upstream
kubernetes,
so
that
we
can
be
a
valuable
partner.
H
Okay,
so
I'm
I'm
going
to
be
sort
of
a
little
bit
of
a
naysay
here.
I
just
think
we
have
to
be
very
careful.
We
are
a
small
community
and
at
the
minute
we
don't
actually
have
a
separate
code
base.
We
build
off
the
openshift
code
base.
So
what
what
you're
talking
about
here
is
quite
a
substantial
change
and
it's
going
to
require
that
all
the
internal
openshift
people
that
currently
contribute
things
are
going
to
have
to
come
and
contribute
into
okd
to
get
something
into
ocp,
potentially
so
I
I.
Think
there's!
H
Quite
you
are
talking
about
a
sort
of
Fairly
major
shift
here
that
we
can't
just
slide
in
by
accident.
There's
going
to
be
quite
a
lot
of
governance,
conversation
organization,
conversation
and
us
as
a
community
I
think
it
it
will
bring
a
big
change
on
to
us
and
because
we'll
now
have
a
sudden
flood
of
of
contributors
want
to
contribute
to
our
code
base,
which
we
actually
don't
have
at
the
minute.
So
I
think
there
is
quite
a
lot
of
change
here.
I
think
it's
the
right
direction,
I
I,
I!
H
F
I
mean
it's
it's
by
no
means
there's
no
by
no
means
a
consensus
that
it's
even
in
the
right
direction,
because
I
mean
it's
a
downstream
right
now,
but
I
think
by
proving
out
that
we
can
get
value
by
having
people.
E
There
is
a
community
place
to
do
this
stuff
yeah
right,
like
the
problem
right
now.
That
openshift
has,
to
put
it
bluntly,
is
they
don't
have
anywhere
to
try
anything
except
for
in
prod
in
in
the
product,
and
that's
not
a
great
place
to
be
when
you
want
to
do
bigger
shifts?
F
Yeah
I
mean,
ultimately,
you
know,
testing
of
riskier
configurations
has
to
happen
somewhere,
so
it
makes
sense
for
groups
and
individuals
and
entities
which
are
more
comfortable
with
those
configurations
to
be
empowered
to
do
so.
So
that's
what
that's!
What
that's?
What
this
exists
for
right,
we've.
E
Already
been
accidentally
doing
it
for
a
couple
years
now,
because
we
were
forced
to
turn
on
C
group
B2
for
bro
for
okd,
when
Fedora
did
because
that's
like
we
had
to
be
able
to
do
that.
So
we've
been
running
in
C
group
V2
for
a
long
time
and
that's
why
I
was
a
little
surprised
that
we
weren't
even
doing
serums
like
oh
okay,
so
we're
doing
C
group
B2
with
run
C.
That
was.
H
H
E
Yeah,
we
probably
do
need
to
pull
fcos
out
of
the
in
out
of
that
and
into
into
our
Loop,
because
I
mean
frankly,
we've
had
problems
with
with
the
F
cause,
okd
integration
and
I
kind
of
lay
the
feed
at
the
fact
that
none
of
us
actually
understand
what
the
hell's
going
on
when
they
get
put
together
and
and
without
being
able
to
influence
and
improve
that
I.
Don't
know
how
that
was
supposed
to
get
better
well.
B
And
it
was
also
you
know
for
a
time
I
was
attending
the
the
F
cost
working
group
meetings
and
their
intent.
Their
direction
is
to
not
be
tied
to
a
particular
kubernetes
or
any
type
of
container
orchestration
system
and
for
people
to
be
able
to
run
containers
individually
for
people
to
be
able
to
use
kubernetes
for
people
to
be
able
to
use
whatever
like
and
not
be
tied
to
a
particular
not
be
opinionated,
basically
in
that
regard,
and
so
for
fedoric
Ros.
B
That
was
great,
but
for
us
not
so
great,
because
it's
not
as
maybe
tightly
bound
as
we
might
want
it
to
be
right
right,
it's
yeah!
So,
okay,
it
sounds
like
we
have
two
different
things
or
three
different
things
going
on
here:
number
one:
I'll
I'll
start
a
thread
for
this
particular
item
for
C
run
number
two
start
a
discussion
about
in
general,
some
of
the
things
that
have
been
coming
out
of
ocp
land
that
we
would
like
to
follow
that
overall
process
of
turning
on
now
and
being
forerunners.
B
H
B
D
E
D
Access
the
reaper,
build
it
and
then
implemented
it's
a
it's.
A
change
in
the
in
the
in
the
cryo
in
the
config.
E
H
E
And
also
the
existing
binaries
might
actually
might
actually
already
have
all
the
stuff
set
up.
We
just
need
to
flip
the
flag
and
then
future
builds
will
have
it
too
yeah.
C
B
B
Let's
see
next
up
is
fcos
updates,
which
we
don't
have
either
of
them
here.
S-Cos
updates,
I,
don't
know
if
you
have
anything
Luigi.
B
B
Okay,
next
up
Community
Development
updates,
Brian.
H
Okay,
so
last
week
we
had
our
first
of
the
new
format
meetings
which,
where
we
actually
use
this
time
in
the
community,
we
want
to
actually
do
some
work.
So
we
actually
reviewed
the
home
page
I've
got
to
actually
release
the
work
that
we
did
and
we
actually
had
like
a
joint
working
session
to
actually
go
in
change.
The
content
on
the
home
page
I
think
that
worked
really
well.
H
It
actually
allows
us
to
move
stuff
on
where
we
want
a
group
conversation
rather
than
sort
of
giving
someone
the
work
to
do
and
not
having
that
collaboration.
So
the
next
one
we're
going
to
do
is
not
the
next
Community
call,
which
will
be
reviewed.
The
work,
the
project
board
again,
but
the
week
after
that,
we
are
going
to
have
a
conversation
on
operators.
H
So,
ideally
we
do
want
to
line
up
a
few
people
to
be
at
that
meeting
such
as
Christian,
so
I'll
I'll,
either
Luigi
or
I,
can
sort
of
reach
out
and
see
if
he's
going
to
be
available.
So
that
is
what
three
weeks
from
today
yeah
three
weeks
from
today
and
that's
really
to
come
up
with
the
okd
operator
catalog.
H
So
we
can
actually
release
some
of
the
operators
that
are
only
available
with
a
subscription
on
ocp
to
actually
add
some
of
that
functionality
and
back
into
okd
things
like
the
pipelines,
things
like
gitups
and
those
operators,
just
one
other
thing,
I
want
to
mention
the
work
around
the
Sno
this
for
okd
the
documentation.
H
I
did
see
that
on
okd43
there
was
a
conversation
about
the
agent
installer,
which
is
needed
for
the
assisted
installer,
so
I'm
wondering,
if
that's
what
was
causing
the
problem,
because
I
couldn't
get
the
assistance
installer
working
on
either
escort
or
f-cos
for
okay,
the
versions
of
okd
and
self-hosted,
assisted
installer,
so
I'm
good,
when,
when
we
get
a
release
of
413,
now
we'll
have
a
replay
of
that
and
see
if
that's
actually
fixed
some
of
the
problems
or,
if
there's
something
else
going
on
there
as
well.
I
Hi
Brian
about
the
Asia
based
installer
I've
been
able
to
use
it
with
s-cos
4.13
I
I
have
a
pocpr
open
to
to
show
it
up.
H
Yeah,
that's
when
I
saw
okay,
so
hopefully
that
will
fix
our
sing,
our
Sno
documentation
and
a
process
for
getting
that
working
that
we
can
actually
put
that
back
in
the
docks
foreign.
H
B
Group,
a
little
bit
of
a
a
little
bit
of
detail
about
about
what
we
talked
about
in
terms
of
the
website
changes
just
so
that
folks
in
this
group
are
aware
who
might
not
attend
those
meetings.
H
Yeah,
so
what
what
we
looked
at
is,
if
you
go
to
okd.io,
the
home
page
has
a
lot
of
words
on
it,
but
very
little
useful
content.
There's
a
lot
of
sort
of
fluff
and
buzzwords
on
there
and
we've
come
across
a
number
of
cases
where
we,
where
we've
talked
to
people
at
this,
the
summit
or
we've
we've
met
people
and
they
say
I'm,
never
sure
what
okd
is
and
whether
I'm
allowed
to
use
it,
whether
I
have
to
get
a
subscription
or
whether
I
have
to
get
a
permission
or
anything
like
that.
H
So
what
we
wanted
to
do
is
we
want
to
simplify,
reduce
the
amount
of
work,
the
amount
of
words
and
just
be
very,
very
brief
and
succinct,
just
to
actually
say
what
is
okd
and
we
had
quite
a
good
conversation
around
getting
that
get
getting
those
words
right
and
what
we
actually
wanted
to
include
and,
more
importantly,
what
we
didn't
want
to
include
what
we
wanted
to
get
rid
of
from.
What's
on
opd.io,
so
I've
got
some
wordsmithing
to
do
we'll
release
it
in
a
branch
on
the
okd.io
and
git
repo.
H
B
Any
other
thoughts
or
feedback
on
Community
News
in
development
news
all
right.
One
thing
that
I
would
add
is
the
the
survey
has
been
updated
with
tweaks
and
I've
gotten
no
feedback
that
anything
needs
to
get
changed.
So
after
the
U.S
holiday
next
week,
I'll
release
the
survey,
and
so
that'll
be
our
2023
survey
and
ideally
next
year,
I'll
we'll
do
it
a
little
bit
earlier,
maybe
in
February
or
something
like
that,
but
at
least
we're
getting
one
out
this
year
and
then
we
can
start
that
process
of
doing
it
regularly.
H
H
Is
it
to
be
more
of
a
because
I
know
there
are
some
people
use
okd
in
production
and
we've
had
we've
had
comments
before
where,
especially
on
the
c
groups
V2,
where
we
lost
a
little
bit
of
stability,
we
had
quite
a
few
quite
a
bit
of
pushback
from
people
that
were
using
it
in
a
production,
environment
and
so
I
think
it
would
be
a
useful
question
just
to
get
a
a
feeling
from
the
community.
H
B
D
E
B
Wordsmithing
another
time
but
yeah,
we'll
I'll,
put
a
question
in
and
then
we'll
word
Smith
it
for
sure.
I
think
that
that's
a
great
idea
and
that'll
help
us
really
know
what
the
community
wants,
as
opposed
to
just
people
that
come
to
the
working
group
meetings,
all
right,
any
other
questions
or
thoughts
about
Community
Development
stuff
that
we
can
take
back
to
the
meetings
and
just
by
attendance
here.
B
I
can
assume
that
the
time
change
folks
are
are
happy
with
we've
got
some
old
faces
that
are
finally
able
to
attend
again,
and
some
people
who
don't
look
at
as
nearly
worn
out
is
so.
E
C
I
B
A
E
H
B
Timothy
was
on
PTO
I'll
check
and
see
if
he's
back,
Christian
was
moving
and
then
had
some
other
stuff
come
up
like
right
after
he
he
moved
and
then,
of
course,
this
week.
Summit
is
happening
right
now,
so
I
want
to
extrapolate
too
much,
but
we'll
we'll
see
I'll
send
something
out
and
just
take
the
temperature
of
folks
beginning
of
next
week
once
they're
back
so
for
sure
all
right.
B
Next
up
on
the
agenda
just
operator,
next
steps
I
wanted
to
tease
this
out
to
make
sure,
because
we
got
another
question
in
the
channel
in
slack
yesterday
about
operators
and
sort
of
where
things
are
going.
So
I
want
to
tease
this
out
again.
So
in
three
weeks,
we're
gonna
have
the
community
development
meeting,
focus
on
operators
and
moving
forward
with
operators
and,
at
the
same
time,
ideally
in
next
week's
or
sorry
in
two
weeks,
the
technical
meeting.
B
This
meeting
I
want
to
be
able
to
Ping,
Christian
and
and
anyone
else
from
the
red
hat
technical
side
to
see.
Do
we
have
all
of
our
ducks
in
a
row
to
be
able
to
pull
this
off
in
terms
of
the
catalog?
B
Does
that
sound
about
right
for
the
for
our
plan
moving
forward
to
address
operators
yep
excellent,
all
right,
and
that
actually
brings
us
to
the
end
of
the
agent
a
little
bit
earlier
than
I
thought
we
were
going
to
be
any
other
questions
or
comments
or
things
that
folks
want
to
address
in
this
meeting.
A
Yeah
Jimmy
yeah
one
thing
we
might:
if
we're
going
to
go
along
the
road
to
Upstream,
then
we
might
need
to
get
multiple.
You
know
version
streams.
You
know
like
yeah,
like
multiple
release
streams,
which
we
don't
now
have
sure
sort
of
like
f-cost.
B
All
right
cool,
well
thanks,
so
much
folks
for
your
time.
I
will
create
those
three
discussion
items
and
then
in
the
working
group
Google
and
then
also
maybe
on
slack
I'll,
shoot
out
the
links
to
those
so
that
folks
can
start
adding
stuff.
Please
do
follow
up
in
this
async,
because
I
think,
ultimately,
follow-up
is
going
to
be
crucial
to
pulling
this
off
all
right.
Thank
you
all
and
see
you
next
time.
Thanks.