►
From YouTube: OKD Working Group 2020 12 08 Full Meeting Recording
Description
OKD Working Group 2020 12 08 Full Meeting Recording
https://okd.io
https://github.com/openshift/community/projects/1
A
So
here
I
am
christian's
having
problems
and
I'm
getting
in
so
he's.
His
tethers
connection
is
too
slow
is
what
he's
saying
in
chat
so
hi
welcome
everybody,
I'm
I'm
not
supposed
to
be
here,
but
I'm
here
so
happy
to
to
to
drive
this
and
the
recording
is
on.
So
that's
good,
so
vadim,
while
I'm
getting
the
page
set
up,
you
want
to
give
an
update
on
the
four
six
release.
If
you,
if
you
don't
mind,
if
you
haven't
already.
B
Yeah
sure
so
the
telemetry
says
it
wasn't
a
complete
flop,
because
we
have
about
140
clusters,
upgraded
and
170
four
five
clusters
still
on
45..
B
Apparently
the
biggest
problem
is
this
fear,
which
has
hit
all
possible
issues
like
fedora,
suddenly
changing
a
hostname
from
default,
localhost
to
fedora
and
systemd
resolve
the
problems,
and
there
was
oh
and
there
was
also
a
problem
with
oc
mirror.
B
So
at
this
point,
fedora
has
reverted
a
change
of
the
host
name
back
to
localhost.
We
are
waiting
for
this
to
hit
our
repost
and
we'll
be
able
to
test
the
sphere
more
actively
system.
The
result
is
apparently
fixed,
we're
disabling
that
and
disabling
it
in
in
a
switch
conf
and
in
network
manager.
So
it
should
be
working
great.
So
there's.
C
Supposed
to
be
another
fix
for
that,
because
there's
a
sc
linux
issue,
that's
preventing
one
of
the
processes
from
working
properly
supposed
to
do
a
symbolic
link.
So
hopefully
that
will
get
fixed
in
florida
in
fedora
and
f
coast,
so
that'll
that'll
fix,
hopefully
the
the
patch
that
we're
doing
for
in
this
way,
yeah.
B
C
B
B
We
have
finally
gotten
to
the
very
end
of
this
just
yesterday
and
it's
gonna
take
I
I
don't
have
estimates
right
now,
probably
a
week
to
rebuild
the
whole
images
so
that
we
could
mirror
that
it's
pretty
serious
for
okay,
but
we're
still
debating
how
serious
that's
for
our
customers.
B
B
B
B
B
Most
of
the
installer
folks
are
on
ptos
right
now,
so
we'll
chase
them
this
week,
because
the
next
friday
is
the
future
freeze
for
openshift
and
soon
we
have
kubernetes
120
rebase
landing
in
4.7.
This
will
be
available
in
the
night
list.
That's
where
we
work
right
now
and
hopefully,
4
7
release
would
go
much
smoother
than
this
one.
A
So
I
apologize
that,
because
I'm
jumping
in
in
late
here
today
and
I'm
a
little
out
of
out
of
sorts,
the
so
the
4.6
we
have
a
build
of
it,
that
people
can
use,
should
we
be
publicizing
it
yet,
and
that
was
a
conversation
we
were
having
in
slack
the
other
day.
So
should
we
wait
until
all
of
these
things
are
re
resolved
before
we
start?
B
Think
so
blogging
and
everything
so
yeah.
We
are
it's
an
official,
stable
release,
it's
just
not
recommended
for
vsphere
folks,
and
these
fear
folks
are
approximately
30
of
our
user
base.
If
the
telemetry
is
correct,
so
that's
pretty
large
chunk.
B
Okay,
I
also
would
like
to
have
some
more
time
with
brush
updates
of
samples
operator
where
we
ship
to
us
and
ubi
things
and
get
some
experience
from
that.
That
would
be
if
we
would
be
publishing
a
blog
post.
They'll
be
very
interesting
to
mention
this,
that
it
now
has
community
samples
and
we
can
extend
them
in
a
bit
faster
fashion
than
ocp,
because
we're
pretty
much
independent
from
from
this
part
of
ocp.
B
That
should
be
an
interesting
part
in
the
blog
post,
probably
yeah.
Okay,.
A
Which
is
good
because
I
haven't
written
the
blog
post,
so
I
kind
of
I
kind
of
was
listening
to
charo
in
some
of
the
back
room
who's.
Not
here,
I
don't
think
charo
managed
to
make
it
today
either
it.
This
is
just
a
week
from
hell,
I
think,
for
for
meetings
clashing
and
he
was
having
difficulty
making
the
crc
or
for
okd
4.6
to
work.
I
think
he
got
it
going,
but
I'm
and
I
think
he
got
the
the
images
up
and
running.
A
I
don't
know
if
anybody
else
on
this
call
has
has
tested
those
and
played
with
them.
Yeah.
Is
there
anybody
who's?
Taking
a
look
at
those,
we
could
use
some
feedback
on
that
to
make
sure
that
they're
working
correctly.
A
D
E
One
thing
I
saw
vadim
and
other
folks
can
chime
in
on
this-
is
one
of
the
things
we're
seeing
a
lot
in
the
slack
is
people
who
are
confused
about
the
32
versus
33f
costs.
Is
there
a
way
that
we
can?
Is
it
notate
that
or
better?
Let
people
know
not
to
panic
or
to
know
that
they
should
grab
32
and
that
it
will
be
updated
to
33?
How
can
we
manage
that
so
that
it
isn't
a
recurring
point
of
confusion?
I
think
that
that
hinders
adoption.
B
I
think
we
need
some
more
documentation
on
this
in
ipi.
The
starting
image
is
managed
by
you
and
it's
going
to
pull
the
latest
stable
f32
in
upi.
People
are
free
to
do
whatever
they
want.
We
just
tell
them
use
stable
fluorocrs
or
you.
If
you
have
some
special
use
case,
you
can
use
a
different
starting
point,
but
it's
they
are
just
confused
by
the
release
page,
which
shows
which
door
cross
release
you
end
up
with,
but
it's
managed
by
mco.
You
don't
have
to
start
with
the
same
one.
B
Another
problem
is
that
some
folks
are
thinking
that
if
they
start
with
the
very
same
release,
the
upgrade
won't
go
will
go
faster,
but
it's
not
true,
because
we
mix
in
our
own
binaries
there
install
packages,
it's
just
gonna,
take
the
very
same
approximate
time.
Perhaps
we
should
know
this
on.
We
should
give
more
attention
to
faq
page
we
have
on
github,
and
probably
documentation
is
the
only
answer
I
have
right
now.
E
A
Yeah,
if
you
make
a
pull
request,
I'll
make
sure
it
gets
merged.
If
you
don't
have
proofs,
the
other
thing
is
to
put
a
note
on
the
okd.io
site
as
well
somewhere
on
the
somewhere
somewhere
loud
on,
probably
on
the
download
page.
So
if
you
make
a
pull
request
there
I'll
merge
that
as
well.
So
that
would
be
great.
I
just
I
I
know,
and
I
so,
if
that's
the
questions
I'm
just
looking
at
who's
here
is
juliana
or
mike
from
acm.
Here
I
think
they
were
going
to
come.
A
I
wanted
to
make
sure
I
didn't
forget
to
say
that
there
I
don't
see
either
of
them
or
maybe
in
the
chat
yeah.
No,
the
folks
that
came
last
meeting
and
go
back
to
the
share
screen
here
from
ocm,
the
open
cluster
management
initiative,
otherwise
known
as
acm
inside
of
red
hat
they're
gonna.
I
think
on
thursday
they're
having
their
first
meeting
and
I'll
post
the
link
to
that
in
the
the
google
group,
so
that
if
people
want
to
join
in
that,
that
would
be
great.
A
A
So
I'd
like
to
rinse
and
repeat
that
so
just
I'll
I'll
put
a
note
on
the
google
group
with
that
as
well.
So
anybody
who
are
other
people
here
besides
me
planning
on
going
to
dev
comp
cz
is,
I
guess,
my
quick
question.
A
Okay,
so
I
will
tap
on
you
and
I
will
try
and
get
christian
to
come
because
I
think
it's
his
time
zone
so
there's
one
at
least
one
chair
from
from
that
and
we'll
we'll
host
that
pretty
much
the
same
way
that
we
did
just
a
quick
five-minute
overview
of
what
okd
is,
and
maybe
a
demo
of
the
code
ready
container
or
some
other
new
aspect.
That's
coming
up
by
february
that
we
want
to
chat
about
and
then
leave
the
rest
of
it
open
for
q.
A
and
conversations
that's
pretty
easy
to
do.
A
That
is
a
new
thing
and
they
probably
will
come
soon
to
one
of
the
working
group
meetings,
but
I
thought
I
would
flag
that
today
and
throw
it
into
the
stop
sharing
again.
I
love
blue
jeans
for
this.
The
link
to
that
there
are
some
already
set
up.
Oh,
this
is
cool.
A
So
take
a
look
at
those
and
I'm
going
to
invite
john
bohannon
from
github
and
one
of
the
red
hatters
who's
been
working
with
them
to
come
to
a
future
thing.
So
start
look,
look,
take
a
look
at
it
and
if
it's
of
interest
we'll
we'll
give
them
some
time
on
the
agenda.
A
Okay,
so
I
thought
it
was
interesting,
so
that
was
that
was
good
and
the
the
other
thing
that
I
have
been
besides
the
blog
post
and
and
besides
the
fedora
magazine
thing,
outreach
and
and
on
that
side
of
the
fence
is
doing
thinking
about
doing
an
okd
summit
or
okd
con,
or
something
like
that
for
contributors
and
working
group
members
and
end
users,
because
that's
what
everybody
needs
is
one
more
virtual
event,
but
trying
to
make
my
virtual
events
be
more
very
specific,
so
that
people
have
a
purpose
as
opposed
to
generic,
so
I
think
and
a
reason
to
come
and
the
good
folks
at
cern.
A
I
don't
know
if
iago
is
not
on
the
call
here.
He
didn't
think
he
would
come
in
yeah.
I've
already
got
access
to
hop
in,
but
thanks
berkus,
that's
a
wonderful
yeah.
I
I
have
open
is
good
yeah.
No,
it's
really
good,
so
I
I
was
thinking
there
since
devconf
is
in
february,
and
summit
is
doing
something
in
april.
That
maybe
march
would
be
a
good
time
to
do
it.
The
folks
at
cern
and
is
joseph
meyer
on
the
call
here
from
rhodian
schwartz.
A
F
D
A
A
A
I'm
pretty
sure
is
using
it
on
openstack
so
and
they
have
they're
just
migrating
from
four
five
to
four
six,
so
they're
going
to
be
one
of
the
first
production
non-working
group
members,
because
I
know
some
of
you
bruce,
I
see
your
face
and
there
was
a
couple
other
folks
who
are
using
it
in
production.
A
Okay
and
sorry,
I'm
not
recognizing
yours
and
which
john,
which
market.
C
A
Okay,
so
I
would
you
know,
I'd
take
everybody,
and
maybe
four
or
five
end
users
talk
about
their
journeys.
What
they're
doing
give
us
our
feedback
and
have
the
working
group
members
listening
to
that
with
as
much
or
as
little
confirmation
bias
as
possible,
but
just
listening
to
that
and
then
since
josh
is
on
here
and
amy
is
here.
One
of
the
things
that
I'd
like
to
do
is
work
a
bit
more
on
our
contribution
ladder
and
onboarding
people
into
the
working
group
and
figuring
out.
I
know
vadim.
A
One
of
the
things
that
you've
been
muttering
in
the
background
is:
how
do
we
get
more
like
code
actual
code
donations
from
the
community
as
opposed
to
waiting
for
vadim
to
get
off
pto
and
tell
us
what's
in
the
latest
release.
A
So
I
would
like
to
spur
that
side
of
the
conversation
in
2021
is
to
try
and
figure
out
how
to
because
it's
very
it's
difficult,
because
openshift
the
product
really
the
main
driver
engineering
resource
wise
for
okd.
A
But
there
are
other
places
where
we
could
do
better
and
we
I've
been
preaching
externally
about
wonderful
contribution,
ladders
and
how
to
become
a
maintainer
and
all
that
stuff
about
other
projects,
and
I
would
like
to
apply
some
of
that
to
okd
itself
and
figure
out
how
that
is
so.
That
would
give
us,
from
now,
until
march,
to
really
have
conversations
about
what
that
would
look
like
and
vedim.
A
I
I
really
value
your
insights
into
how
to
how
people
can
do
that
and
help
so
that
you
and
christian,
and
and
even
even
the
charo
building
the
code
ready
container
images
on
his
home
lab
and
figuring
out
that
part,
but
that
I
I'm
think
if
people
are
on
board
for
that
and
I'm
just
looking
in
the
chat.
I
I'm
thinking.
C
C
Everybody's,
probably
tired
of
seeing
my
email
inside
of
github
at
this
point,
so
I'm
probably
too
prolific.
B
Now
it
definitely
needs
to
be
discussed.
The
problem
is
if
an
outside
users
have
some
low
hanging
fruits
in
okd.
That
means
we're
doing
our
jobs
very
very
poorly,
since
we
base
our
stuff
on
no
cp
and
it's
supposed
to
get
properly
tested
by
ci
and
then
trickles
down
into
okd,
you
are
not
supposed
to
to
find
issues
immediately
on
installation
or
right
after
update.
That
means
we
on
our
ci
side
need
more.
B
I
don't
know
more
tests
more
attention
to
that.
There
is,
however,
a
huge
area
of
contributions
to
ocd.
As
in
how
do
we
extend
it?
How
do
we,
which
components
community
would
use
more
actively
than
enterprise
consumers,
for
instance,
something
related
to
let's
encrypt
something
like
certbot
included
in
okd
payload
itself
would
be
more
interesting
for
the
community.
B
Ocp
doesn't
need
that,
so
these
extension
points
are
probably
where
we
should
be
contributing
code.
There
is
a
bunch
of
interesting
optional
operators,
for
instance,
folks
at
adversary
organization
in
openshift
part
have
developed
their
own
routes,
monitor
basically
a
black
box
exporter,
which
looks
that
all
of
your
routes
are
available
and
served
correctly
that
won't
ever
get
into
cp
code
base.
Originally,
for
instance,
it
probably
belongs
to
olm,
so
that
discussion,
which
operators
or
components
should
live
as
optional
another
lamb
and
which
should
be
part
of
ocd
itself.
B
That's
an
interesting
discussion.
We
probably
want
to
minimize
our
our
surface
of
what
we
maintain
and
put
more
stuff
into
olam,
but
tweaking
okay
as
a
distribution
and
getting
some
pre-installing
some
subscriptions
from,
for
instance,
that's
an
interesting
topic
to
discuss,
but.
A
There's
also,
the
the
other
end
of
the
thing,
too,
is
that
I'm
getting
inquiries
from
the
product
management
team
a
lot
a
lot
of
people
who
are
coming
to
openshift
for
the
first
time
that
are
potential
users
of
that
come
to
the
open
source
side
of
things.
So
many
of
you
maybe
didn't
come
in
through
the
openshift.com
world.
You
came
in
through
the
open
source
side
and
making
that
more
of
a
seamless
experience
for
people.
A
So
I
know
I've
been
talking
to
the
developer.com
folks
about
having
okd
be
one
of
the
tools
that
they
use
when
they're
doing
demos
and
things
like
that
and
making
that
a
smoother
thing
so
and
also
hearing
from
people
who
are
are
in
the
community.
And
how
do
we
get
that
our
feedback,
more
fluidly
into
the
product
management,
epics
and
things
like
that,
I
think,
is-
is
something
that
I'd
like
to
figure
out.
A
We
do
it
by
sharing
our
stories,
and
you
know-
market
america
and
rhodium,
shorts
and
cern-
and
you
know
putting
yourselves
on
stage-
is
one
way
because
the
product
managers
and
the
engineers
all
listen
to
those
talks
over
and
over
again,
which
is
great,
but
I
think
there
needs
to
be
an
even
more
sort
of
a
feature
request
process
of
some
ilk
that
we
build
out
so
that
it
gets
up
and
in
and
there's
mike
all
right,
mike
ing
joined
a
minute
ago-
and
I
did
mention
this-
and
this
is
mike
ing
from
the
oc
m
folks.
A
The
open
cluster
management
folks
is
just
put
in
dropped
into
the
chat
that
this
thursday,
which
I
think
is
the
10th
at
15
30
utc.
There
is
going
to
be
their
very,
very
first
open
cluster
management
community
project.
So
if
you're
interested,
they
were
the
ones
that
spoke
at
the
last
meeting.
A
So
that's
so
that's
what
I
think
2021
more
more
seamless
into
the
cicd
workflow
and
the
testing
stuff.
I
think
that's
a
big
goal
for
us
to
get
that
done,
as
well
as
the
build
process
for
code
ready
containers
to
be
more
automated
from
my
perspective,
but
I
would
really
like
john
and
joseph
and
people
who
are
actually
using
it
out
there
in
the
world.
A
What
what
are
the
features
that
you're
looking
for,
that
we
don't
have
and
where
and
where
we
we
need
to
be
working
and
to
push
that
up
the
pipeline
to
more
visibility
to
the
product
managers
is
really
kind
of
their
key
and
how
you
can
you
know?
How
can
we
get
your
contribute
if
you're
willing
john?
Where
can
we
put
you
to
work
besides,
you
know
making
changes
to
the
the
documentation
and
the
okd
site
and
creating
recipes.
C
Honestly,
because
that
I
mean
for
the
last
two
weeks,
that's
been
my
pain
point
and
I
and
I'm
more
than
willing
I've
actually
gone
through
the
process
of
building
my
own
packages
and
stuff
and
making
changes
and
then
funding
vocally
to
make
sure
that
they
work.
So
I
become
a
lot
more
familiar
with
that
process.
B
B
Rsa
keys
are
not
being
being
rejected,
so
internally
we're
discussing
that
around.
G
B
Yeah
the
way
we
work,
we
don't
touch
clusters
manually.
We
just
ask
ci
to
build
it.
If
it
passes,
then
it's
good
if
it
doesn't
pass.
There
is
no
way
we
can
override
this,
and
that
takes
quite
a
lot
of
time
and
really
hampers
our
productivity
right
now,
but
that's
an
important
decision
we
need
to
make.
But
meanwhile
there
is
a
bunch
of
proposed
patches.
We
need
to
test
like
fresh
systemd,
freshest
network
manager.
B
At
least
they
won't
hit
us
at
the
same
time
just
like
they
did
the
months
ago
when
it
comes
to
open
shift
codes
like
sending
code
straight
to
openshift,
it's
a
bit
tricky
because
you
have
to
start
at
4.7
and
if
you
really
want
to
to
try
it
out
in
4.6,
you
would
have
to
file
a
bug
that
bug
needs
to
be
triaged
by
folks
and
approve
that
we
want
this
to
be
backboarded.
That's
a
lot
of
bureaucracy.
B
You
probably
don't
want
to
get
involved
in
this,
but
if
you
really
want
to
and
you're
hitting
a
specific
bug
will
definitely
help
with
that.
That's
that's
not
an
issue.
It's
just
not
trivial,
because
we
don't
have
like
a.
I
cannot
think
of
some
low
hanging
fruit.
We
need
to
fix
right
now
and
we
could
we
could
bring
people
through.
B
Although
that
sounds
like
a
it
sounds
like
something
we
should
have.
We
should
have
a
list
of
bugs.
We
would
like
to
teach
folks
to
how
to
work
with
with
oakley
and
go
through
the
whole
process.
Yeah.
A
I
think
is
part
of
the
contributor
summit
con
thing.
We
might
even
have
a
like
a
little
hackathon,
where
we
make
even
just
walk
people
through
fixing
a
bug,
and
so
there
well.
C
I
tell
you
from
my
experience
like
yesterday.
You
know
I
went
through
and
there's
a
couple
of
bugs
out
there
that
I
wanted
to
test
for
the
fixes.
So
I
I
took
the
code
from
github
and
then
applied
the
fixes
and
then
built
mco
and
then
applied
it
to
my
cluster.
So
but
that
was
not
a
greatly
documented
process.
C
You
know
on
how
to
do
that.
Some
of
it
there's
one
page
that
was
okay,
it's
sort
of
there,
so
I
figured
it
out,
but
you
know
those
are
pieces
that
you
know
if,
if
you
want
people
to
help
develop
or
help
test
patches
and
stuff,
that
probably
need
a
little
bit
more
guidance
on
in
order
to
get
there.
Because
there's
a
lot
there
to
understand,
and
then
you
know
how
do
you?
C
G
Yeah,
I
wanted
to
add
a
comment
on
the
whole
bug:
triage,
debugging,
kind
of
informational
side
of
things
too.
It
might
be
worthwhile
to
kind
of
go
through
some
of
the
hot
topics
that
are
coming
up
in
bugzilla
at
one
of
these
virtual
events,
to
kind
of
show,
I
guess
more
of
the
community.
G
What
are
the
issues
that
are
hitting
the
head
of
master
right
now
or
the
you
know
the
tip
of
our
main
trunk
development,
so
we
could
help
to
propagate
those
issues
further
out,
because
you
know
vadim
was
pointing
this
out.
You
know
when,
when
we're
working
on
ocp
as
a
product,
you
know
a
lot
of
the
stuff
we're
looking
at
is
you
know
coming
out
for
the
plus
one
version
and
that's
probably
a
version
or
two
away
from
what
we're
seeing
in
okd
so
like.
G
A
E
One
of
the
things
I
think
might
be
helpful
is
for
us
to
somehow
collate
a
list
of
folks
that
aren't
testing
and
debugging
and
what
particular
platform
they're
using
so,
for
example,
if
I
knew
all
of
the
vsphere
folks,
we
could
all
get
together
and
and
have
a
little
chat
about.
Okay,
did
you
notice
this?
Did
you
notice
this
right
now,
everything's,
sort
of
scattered,
and
you
don't
necessarily
unless
you're
on
a
call
with
someone,
and
you
hear
them
talking
about
what
particular
platform
they're
using
you
don't
really
know
like.
E
I
know
now
know
john,
is
you
know
from
this
and
some
conversation
online
he's
vcr
as
well?
I
think
it
would
be
helpful
if
we
somehow
attract
that
so
that
we
could
say:
okay,
here's
something
I'm
seeing
in
vsphere.
It
seems
specific
to
v's
here
who
are
the
other
vco
folks
that
can
bounce
this
off
yeah.
A
And
I'd
be
happy
to
arrange
like
a
zoom
or
a
blue
jeans
for
v
v
sphere,
like
a
sub
working
group
of
people,
but
I
think
we
we
started
doing
that
with
when
we
did.
The
okd
marathon
started
like
trying
to
get
the
content
created
for
that
back.
A
That
was
kubecon
eu
in
august
and
started
to
identify
who,
who
were
the
people
who
were
willing
to
like,
say:
hey
I'll,
do
digitalocean
or
I'll
do
vsphere
or
I'll
do
this,
but
then
to
go
the
next
step,
and
you
know
self-identify,
maybe
in
the
group
which
one
you
are
you're
willing
to
be
a
tester
for
and
then
creating
a
space
for
that
in
in
the
repo
somewhere
and
posting
you
know,
maybe
a
bi-monthly
all
the
vsphere
spokes
or
like
when
vsphere
is
a
problem
like
it
is
right.
A
Now,
I'm
sorry
joseph
you
came
in
late,
so
yeah
we're
we're
having
some
issues
with
the
vsphere.
But
I
think
that
I
mean
I'm
happy
to
facilitate
that
to
and
to
do
that
and
host
that.
But
I
think
we
also
have
to
would
have
to
have
one
of
the
technical
co-chairs,
like
vedim
or
charo
or
christian,
to
come
and
be
the
facilitator
for
that
conversation,
or
at
least
the
listener,
and
maybe
not
the
answer
of
all
questions,
but
to
have
one
of
the
openshift
engineering
team
I
am
mike.
A
A
Do
you
think
next
week,
at
the
same
time
by
then
we'd
have
one
that
we
could
do
a
vsphere
triage
session?
Maybe
what
would
be
what
I
would
call
it
at
this
point.
B
B
I
don't
have
an
estimate
for
that
and
I
don't
want
to
break
another
stable
release
for
the
folks
who
are
doing
mirroring,
so
the
nightlys
in
vsphere
should
be
fixed
very
soon,
but
stable
release
we'll
probably
want
to
hold
it
off
until
we
fix
the
build
from
problem
again
I'll
file,
a
tracking
bug
where
we
will
track
all
the
problems
and
discuss
the
stuff
yeah.
A
If
that,
if
you
think
that's
better
buddy,
we
wait
longer
or
should.
Are
you
okay
with
doing
something?
Maybe
next
week
for
vsphere
just
to.
B
Let's
give,
let's
give
it
a
try
next
week
and
you
know
holidays,
are
tricky
and
see
how
it
goes.
We
might
not
have
to
meet
every
single
week
just
to
just
put
out
the
fire.
I'm
thinking.
A
Alternate
like
next
next
in
between
do
vsphere
then
do
a
work,
a
regular
working
group
meeting
and
then
the
following
one
pick
one
of
the
other
platforms.
You
know
and
you
know,
do
a
bare
metal
one.
You
know
vsphere.
G
Sorry,
like
I
I
mean
I'm,
I
am
totally
happy
to
get
involved
and
help
like
liaise
as
much
as
I
can
between
what's
happening
in
the
machine
api
team
and
what's
happening
here.
I
think
you
know
it's
interesting
to
see
it's
interesting
to
me
at
least,
to
see
all
the
discussion
about
vsphere,
because
I
think
we
have
an
internal
impression,
at
least
that
we,
you
know
we're
not
totally
happy
with
where
we're
at
with
the
vsphere
status.
G
Right
now
you
know
we'd
like
it
to
be
more
stable
and
I
think
we
see
a
lot
of
problems
with
vsphere
and
I
would
love
to
be
able
to
see
a
situation
under
which
we're
taking
community
patches
and
incorporating
them
into
the
installer
into
the
machine
api.
G
Whatever
it
happens
to
be
so
yeah
like
if
there's
a
vsphere
triage
meeting,
and
we
come
up
with
a
list
of
issues
or
whatnot
like
I'm,
certainly
willing
to
help
like
try
and
match
those
against
either
what
we're
looking
at
internally
or
how
we
could
coordinate
a
patch
coming
from
the
community
and
landing
in
one
of
the
ocp
repos,
because
I
think
it
it's
completely
doable.
We've
taken
community
patches
before
and
I'd
love
to
see
that
kind
of
like
accelerate
in
the
future
yeah.
A
So
mike,
if
you
and
maybe
vadim
can
come
to
next
week,
I
will
post
something
on
the
google,
the
google
group
and
just
use
this
exact
same
time
slot,
because
it
works
for
my
schedule.
Now
that
I've
adjusted
for
time
zones
and
and
then
just
use
that
and
then
and
maybe
just
call
it.
The
vsphere
tree
update
or
triage
conversation
and
I'll
call
it
triage,
because
that's
what
came
out
of
my
mouth
and
now
you're
all
thinking,
triage
and
then
I'll.
A
Let
mike
and
vadim
mostly
drive
it
and
we'll
have
a
notes
document
like
we
have
for
the
meeting
notes,
and
we
can
share
that
with
the
engineers
and
the
product
managers,
and
I
can
let
you
guys,
maybe
mike,
if
you'll
own,
that
facilitating
the
conversation
with
the
engineering
teams,
because
I
I
yeah
the
other,
the
other
part
of
it
is.
Is
the
product
managers
are
looking
for
this
feedback
and
and
the
engineering
teams
are
looking
for
this
feedback.
G
Yeah
yeah,
I'm
totally
happy
to
help.
I
mean
you
know.
I
don't
say
unfortunately,
but,
like
my
my
sphere
of
influence
is
kind
of
limited.
In
this
respect
I
mean
I'm
I'm
working
with
the
cloud
team
and
machine
api.
So,
like
I'm
looking
a
lot
of
the
cloud
ap,
the
cloud
providers
and
our
team
is
owning
that
code
for
ocp,
but
we
don't
necessarily
dive
deep
into
the
installer
stuff.
G
A
Okay,
be
good,
I'm
just
looking
at
the
so
that
I'm
happy
to
facilitate
that,
and,
and
and
and
I'm
gonna
float
a
few
dates.
Probably
not
this
working
group.
Do
we
when's
the
next
meeting
that
we
have
scheduled?
I
think
we
have
one
for
some
obscene
time
close
to
correct
to
to
the
holidays.
A
Maybe
the
22nd,
the
22nd.
Are
we
really
gonna,
I'm
gonna
miss
that
one
yeah.
That
was
the
the
other
question
I
had
is
that.
D
I
think
we
can
all
just
kind
of
say
no
or
not.
We
should
probably.
D
A
Don't
know
if,
unless
there's
like
something,
everybody
needs
to
do
that.
Maybe
we'll
do
this
vsphere
triage
to
make
sure
that
the
vsphere
folks
are
all
happy
and
have
something
to
test
with
next
week
and
they
can
play
on
their
holidays.
G
Something
john
brought
up
earlier.
You
know
when
it
comes
to
hacking
and
kind
of
like
replacing
components
inside
of
openshift
and
whatnot.
Are
you
know
we
are
aware
of
this,
at
least
in
terms
of
some
of
the
engineering
teams
and
we're
trying
to
write
more
documentation
and
blogs
kind
of
addressing.
G
You
know
how
do
you
get
in
here
and
start
hacking?
These
things
we've
started
to
add
this
to
the
machine
api
repo.
So
if
you
look
at
the
machine
api
operator,
you'll
start
to
see,
we
have
like
some
hacking
documents
there
and
we're
starting
to
add
more.
But
you
know
one
of
the
things
that
our
team
is
taking
on
is
we
would
like
to
document
showing
our
users.
You
know
how
to
hack
on
these
components.
You
know
how
to
replace
components,
how
to
add
your
own
cloud
provider,
these
kind
of
things.
A
Well
I'll
give
you
a
space
for
it
yeah.
So
I
actually-
and
I
know
amy
merrick's
on
here
and
so,
and
she
is
a
documentation,
person
par
excellent
and
a
community
person
part
if
we
can
coerce
her,
maybe
into
helping
us
coach
people
on
how
to
do
better
documentation
and
to
contribute
to
the
company.
That
would
be
an
excellent
thing
to
do
too,
to
set
up
a
a
session
on
that.
A
That
would
be
wonderful,
amy.
I
I
would
love
that
help,
and
that
would
be
great
and
then
so
we
could
do
that
and
I
I'm
thinking
you
know
the
more
we
can
do
to
get
the
contributor
ladder
to
be
clear.
Hacking
guides
how
to
contribute.
We
have
that
one
page
that
vadim
put
the
link
in
there
for
on
contributing,
but
that
is
good,
but
it's
not
sufficient.
Obviously,.
C
A
And
jamie
did
I
I
did
if
you
put
that
recipe
in
as
a
request,
a
pull
request
against
okd.io
I'll,
add
it
to
the
recipe
page
for
okd.
That
I
owe,
and
even
the
and
I
know,
charo
was
going
to
do
some
documentation
around
making
that
clear.
How
to
add
your
recipes
into
up
there.
So
there's
a
lot
of
room
for
improvement
is
what
I'm
saying,
and
I
think
2021
is-
is
what.
A
Want
to
focus
on
that
making
sure
that
we
have
all
that
documentation
aligned
there
and
if
anyone
I
was
looking
at
the
calendar
for
things
in
march,
circling
back
to
doing
like
an
okay,
dude,
end
user
summit,
contributor
summit
kind
of
thing.
A
If
anyone
has
I
looked
at
my
calendars
and
it
looked
like
march-
was
the
emptiest
month
to
do
something
and
to
try
and
do
it
in
using
blue
jeans
or
zoom
and
making
it
highly
interactive
as
opposed
to
entrato
and
maybe
use
hoppin
as
as
josh
was
pointing
out
to,
but
something
that's
that
would
give
us
the
ability
to
do
like
to
do
hack
things
and
to
actually
collaborate
and
work
on
some
of
this
documentation
live.
G
Dan,
I
know
I
know
you.
You've
talked
a
lot
about
like
our
virtual
meetups
and
everything
and
all
that
stuff.
What
like
any
thoughts
about
doing
like
an
okd
like
hackathon
for
a
couple
days
next
week
or
something
like
we
could
just
get
together,
triage
some
issues
and
then
work
through
them
over
the
course
of
a
day
or
two
or
just
allow
people
to
experiment,
see
what
they
come
up
with.
That
kind
of
stuff.
A
Yeah,
so
what
I'm,
what
I'm
always
happy
to
what
I
think,
what
what
I'd
like
to
do
is
do
this
triage
session
for
v
vsphere
and
then,
if
we
have-
and
we
can
go
all
day,
you
know
as
long
as
people
want
and
keep
the
keep
the
lines
open
to
do
that.
I'm
just
looking
at
the
calendar
here.
Excuse
me
a
second
and
see
what
next
week
looks
like
before
I
put
my
foot
in
my
mouth
and.
A
Yeah
I
I
think
that
if
we
did
the
triage
meeting
and
maybe
made
it
longer
like
the
first
part,
would
be
actually
setting
up
listening
to
people's
issues
and
stuff,
like
that,
and
I'm
happy
to
extend
the
time
period
for
as
long
as
people
want.
I
might
not
be
the
the
best
person
to
to
run
the
hackathon,
but
but
let's,
let's
do
the
first
triage
thing
there
and
it
may.
We
can
also
do
it
the
next
day
or
yeah.
It
gets.
A
It
gets
easier
as
we
get
close
to
christmas,
because
people
are
leaving
and
leaving
me
alone
so,
but
yeah
I'm
happy
to
set
up
a
you
know
an
interactive
zoom
account
or
if
zoom
is
your
preference
or
blue
jeans,
whichever
is
zoom
allows
us
to
do
breakouts
a
little
bit
nicer
than
blue
jeans
has
nothing
for
that
so,
and
I
have
a
zoom
account
now
that
lets
me
do
that
so
yeah.
That
would
be.
That
would
be
good
and
maybe
the
doc
stuff.
A
But
let's
do
the
first
triage
session
next
week
see
what
comes
from
that
who
shows
up
and
what
issues
we
have
so
and
then
I'm
happy
to,
and
then
I'm
happy
to
do
like
a
reoccurring,
hack
stuff,
so
that
we
do
more
than
just
talk
in
the
working
group,
which
is
good.
We
need
to
see
your
faces
and
you
can
get
your
feedback
but
yeah.
G
I
was
just
thinking,
maybe
like
you
know,
I'm
looking
optimistically
towards
next
year
when
we
can
hopefully
travel
again
and
whatnot,
and
we
get
back
to
like
kubecon
and
openshift
comments
and
whatnot
that
just
it
kind
of
makes
me
wonder
like
it
would,
it
would
be
really
cool
if
we
could
do
like
an
okd
focused,
get
together
or
something
where
we
all
kind
of
are
able
to
meet
up
and
and
do
some
hacking
and
whatnot,
and
you
know
just
yeah
like
have
fun.
You
know
hackathon
kind
of
stuff.
A
Yeah,
I
think
optimistically,
I'm
seeing
people
in
vancouver
open
source
summit
said
in
august
that
they're
going
to
try
and
host
something
in
person
in
august
for
open
source,
the
linux
foundation
thing
and
then
kubecon
north
america
is
scheduled
for
los
angeles
in
october,
which
I
think
is
is
more
viable.
You
know,
I
I
don't
have
a
crystal
ball
and
I
haven't
been
vaccinated,
yet
my
name's
not
william
shakespeare
and
I'm
not
in
the
uk,
but
I'm
hoping
that
I
that
we
do
optimistically.
A
But
I
think
that
we
could
do
some
stuff
well
before
that,
virtually
using
zoom
in
the
breakout
stuff,
because
basically
that's
the
three
day
class
that
I'm
in
right
now
that
I'm
gonna
have
to
hop
out
at
is
is
is
all
done
in
zoom
and
is
all
with
breakouts
and
zoom.
And
it's
working
pretty
nicely
an
inner
source
summit
or
apac
did
a
nice
thing
with
breakouts
and
and
hacking.
A
So
a
group
of
people
could
be
assigned
to
go
to
a
room
work
on
something
like
led
by
amy
to
fix
the
you
know:
the
onboarding
documentation,
for
example,
so
yeah
and
joseph
you're
always
nervous.
A
So
that
that's
sort
of
what
I
was
thinking
is
there
anything
else
that
I
I've
been
talking
a
lot
that
other
people
have
that
we
should
have
talked
about
today.
A
Any
problems
people
are
seeing,
I
saw
a
finger
go
up
by
neil,
was
that
just
a
random?
No,
that
was
just.
A
C
C
C
D
Well,
unfortunately,
slack
is
the
place
that
we're
that,
in
theory,
we're
supposed
to
potentially
be
doing
it.
The
problem
with
slack
is
that
it's
slack
but
other
than
that
yeah.
That's
that's.
Typically,
the
place.
A
There's
a
new
set
of
features
and
I
haven't
played
with
them
yet
maybe
somebody
else
has
in
github
called
github
discussions
that
I
haven't
played
with
at
all
that
that
might
be
something
we
could
use
as
well,
and
we
could
always
you
know
the
github
actions
team
will
be
coming
sooner
than
later,
but
there's
another
whole
group
of
people
over
at
github
that
have
been
trying
to
promote
using
their
discus
discussions
capabilities.
A
D
A
I
I
I
would
like
to
keep
stuff
in
the
open
as
much
as
possible
and
tied
to
like
the
github
community
projects
agenda
and
that
just
so
that
there's
one
central
location.
But
that's
me.
A
G
Have
to
be
able
to
get
into
the
slack
in
the
first
place.
I
mean
there
are
a
couple
devs.
If
we're
talking
about
the
kubernetes
slack.
There
are
a
couple
openshift
devs
who
do
hang
out
in
that
openshift
dev
channel.
I
mean
I'm
one
of
them,
so
you
know
for
those
who
wanted
to
start
a
more
developer
oriented
conversation
like
you
could
certainly
try
there.
You
know.
C
And
then,
if
the
conversation
goes
about,
where
right
now
are
basically
the
openshift
dev
opened
well
three
places,
I
guess:
openshift
dev
openshift
users,
but
also
github
and
github,
doesn't
seem
like
a
great
place
to
ask
questions
about
how
to
do
something
and
yeah,
and
I
don't
want
to
do
that.
G
A
Yeah,
it's
also
there's
the
fine
line
of
customer
support
versus
and
technical
support
for
customers
versus
you
get
what
you
pay
for.
So
no,
I
understand.
C
A
Yeah,
no,
I
know
I
mean
and
that's
what
it
what
we
yeah.
So
I
think
right
now.
What
I'm
trying
to
do
is
make
sure
that
I
know
that
vadim
and
christian
pay
attention
and
charo
pays
attention
and
mike
you
pay
attention
but
keep
the
visibility
of
okd
in
the
openshift,
dev
and
user
in
that,
so
that
the
engineers
are
cognizant
that
there's
a
lot
of
good
feedback
coming
from
okd
and
the
okd
working
group.
A
E
And
one
of
the
things
I
think,
oh
sorry,
diane
go
ahead.
E
One
of
the
things
that
I've
seen
just
in
the
past
couple
weeks
is
in
terms
of
conversations
on
github
or
conversations
in
the
channel
or
whatnot.
It
can
help
for
folks
that
are
in
the
working
group
to
participate
in
them
or
or
folks
who
are
friendly
to
to
okd
and
and
to
to
christian
and
vadim
etc.
E
In
terms
of
I've,
seen
a
couple
of
tickets
come
in
where
folks
were
like
really
sort
of
demanding
to
vadim
and
christian
and
other
folks,
and
I
think
if
you,
if
you
participate
in
the
channels
and
are
friendly
and
supportive,
it
sort
of
lifts
the
conversation
up
a
little
bit
more
so
that
folks,
it
reinforces
the
notion
that
vadim
and
christian
are
busy
and
they
need
people
to.
E
Actually,
you
know
get
the
documentation
into
their
bug
reports
when
they're
submitted-
and
things
like
that,
like
so,
I
think,
having
conversations
out
in
public
helps
sort
of
reinforce
a
positive
culture
in
dealing
with
with
the
folks
who
are
actually
doing
a
lot
of
development
on
this.
Just
a
side
thought.
A
One
of
the
things
that
that
I'm
trying
to
do
internally
at
red
hat
is
to
raise
the
visibility
of
okd
within
the
product
managers,
conversations
and
so
get
your
feedback
back
to
them
so
that
they
realize
that,
like,
I
think,
john
you're,
using
the
commercial
version
and
the
the.
A
Okay,
perfect
so
there's
I
guess:
rhodium
schwartz
is
there's
a
cup
there's
a
number
of
people
who
have
both
ocp
and
opd,
and
so
there's
that
that
conversation
that
I'm
trying
to
make
sure
people
realize
that
this
is
one
of
the
on
ramps
to
being
a
customer,
and
you
know
raise
that
visibility,
and
it's
that's
so
that's
part
of
part
of
one
of
the
things
I'm
trying
to
do
this
year,
too,
is
to
really
focus
on
end
users
and
what
they're
doing
and
what
your
workloads
are,
because
I
think
there's
also
like,
with
the
open
data
hub
team,
they're,
very
interested
in
hearing
people
who
have
ai
and
ml
and
the
edge
folks
and
each
of
the
other
silos
and
the
telcos
and
stuff
that
are
market
specific.
A
And
then
we
once
we
once
they
know
that
someone's
using
okd
and
maybe
using
it
in
a
an
edge
scenario.
Then
we
can
get
some
of
the
resources
from
the
edge
team
to
help
us
as
well
and
be
aware
of
it
and
and
the
more
visible.
We
are,
the
more
more
likely
they'll
they'll
come
and
pay
attention
and
take
your
feedback
yeah.
There
are
many
roads
to
paying
customers.
G
And
you
know
this
is
probably
something
I
imagine
resonates
with
with
vadim
and
christian
and
whatnot.
You
know
for
those
of
us
who
are
like
red
hat
employees
and
we're
working
on
ocp,
but
we
also
really
interested
in
okd
or
we're
working
on
okd.
It
can
get
tough
for
us
to
like
turn
it
off.
You
know
because
it's
like
we're
going
100
trying
to
make
the
product
really
good,
and
then
we
turn
to
the
community,
and
it's
like
the
community's
got
a
lot
of
asks
and
it's
difficult
sometimes
for
us
to
say
you
know.
G
Well,
I'm
gonna,
I'm
gonna
put
more
effort
into
the
product
or
community.
It's
like
we
don't
we
don't
necessarily
look
at
it
that
way.
So
I
would
love
to
see
us
have
a
stronger
okd
community
that,
like
traditional
red
hat
products,
there's
a
bigger
communication
between
the
community
and
the
product.
Folks,
like
I
would
love
to
see.
A
That,
personally
yeah
no,
I
think
I
think
that's
my
goal
really
with
and
and
raising
up
the
visibility
of
the
end
users
in
this
community.
A
So
jamie
you'll
get
a
umich
request
to
talk
at
something,
and
you
know
every
everybody
will
at
some
point
and
dado
when
you're,
where
you're
doing
stuff,
neil
and
and
others
and
and
bruce
at
bcit,
so
really
showcasing
this
coming
year.
You
know
some
of
the
stuff
that
you're
doing
and
by
recording
the
videos-
and
you
know,
there's
you
know
the
guy
from
well.
It
says
waleed
from
armaco
in
saudi.
You
know
those
folks
are
you
know.
A
This
is
really
a
great
group
of
folks
who
show
us
what
some
of
the
problems
are,
and
you
know
what
problems
you're
trying
to
solve
and
that's
really
key
to
us
innovating
period.
You
know,
that's
been
my
been
my
sort
of
mantra
the
past
six
months
here.
Is
that
really
the
end
users
are
driving
the
innovation
these
days
in
open
source,
so
you'll
hear
me
pontificate.
A
I
do
have
to
end
this
call
now
because
I
am
in
a
class
and
supposed
to
be
learning
something,
but
I
always
learn
when
I'm
hanging
out
here
with
you
guys.
So
thank
you
for
today
and
I
will
go
into
the
working
group
mailing
list
and
we'll
get
you
guys
an
appointment
for
next
week
to
do
a
vsphere
triage
and
I'm
hoping
vadim
that
you'll
be
available
to
come
to
that.
If
that's
good,
yeah
cool
and
mike
and
I'll
I'll,
let
you
guys
mostly
lead
it
and
I'll
just
record
it.