►
From YouTube: OpenShift Commons Briefing: CoreOS Integration Strategy Update - Jeff Ligon and Ben Breard (Red Hat)
Description
OpenShift Commons Briefing: CoreOS Integration Strategy Update with Jeff Ligon and Ben Breard (Red Hat)
A
Well,
hello,
everybody
and
welcome
to
another
openshift
Commons
briefing.
This
one
is
about
the
core
OS
integration
strategy,
I
think
it's
a
very
timely
conversation
and
when
a
lot
of
people
have
been
asking
me
about
so
I've
thought
and
Jeff
and
then
to
come
on
to
give
us
an
update
on
the
progress
and
the
road
ahead.
So
I'm
not
going
to
spend
a
lot
of
time
introducing
them
or
talking
about
them,
I'm
gonna.
Let
them
introduce
themselves
and
get
right
into
the
meat
of
the
thing.
B
Saying
hey
guys
so
my
name
is
Ben
Briard
I'm,
the
product
management
team
here
at
Red,
Hat,
I
kind
of
work
with
a
lot
of
different
technologies
that
are
kind
of
like
think
of
them
as
like
low-level
enablement
on
both
the
Linux
side,
container,
runtimes
and
things
like
systemd
and
then
how
we
make
that
into
a
useful
container
host.
So
you
know
moving
for
just
really
focusing
on
everything
we're
doing
around
Red
Hat
core
OS,
so
yeah.
That's
me,
I
kind
of
sit
in
between
the
rel
and
OpenShift
sides
of
the
house.
B
D
B
At
the
kind
of
what
we're
calling
the
immutable
host
layer
of
the
stack
I
want
to
kind
of
give
a
little
bit
of
context
for
for
the
presentation
today.
Just
when
we
say
core
OS
I
want
to
make
sure
that
we're
all
kind
of
speaking
the
same
language
using
terms
in
the
same
way
so
that
you
know
I'm
not
talking
in
circles
here,
but
just
a
really
brief
look
at
what
core
OS
offered.
B
So
obviously,
tectonic
was
the
the
flagship
from
core
OS,
which
is
you
know,
full-blown
kubernetes
platform,
which
they
did
a
phenomenal
job
on
the
actual
provisioning
and
sheathing
the
cluster,
and
then
all
these
concepts
from
updates
you
management
with
things
like
operators
and
things
like
that.
It
was
a
really
robust
platform.
Everybody
ends
call
hopefully
by
this
point,
knows
that
there's
you
know,
there's
a
lot
of
commonalities
there
with
open
ships,
obviously
both
being
kubernetes
offerings
that
we're
really
we're
basically
combining
the
two
right
so
we're
getting
it
we're
gaining
all
of
these
cool
capabilities.
B
That
tectonic
brings
on
like
the
the
operation
side
of
the
house,
to
open
ship
where,
as
open
ships,
as
always
had
this
great
developer
story
for
onboarding
applications.
Now
we
were
just
really
excited
about
about
everything
that
tectonic
brings
to
that
story
to
round
it
out
now.
The
next
one
some
Kouros
container
linux
is
the
it
was
originally
called
core.
Last
and
fun
story,
I,
don't
know.
B
If
anybody
here
knows
this
or
not,
but
the
kind
of
the
way
this
got
started
is
I
am
Leonard
to
to
longtime
Red
Hatters
I
gave
a
talk
at
a
plumbers
conference
and
I,
don't
know
2012,
maybe
about
this
concept
of
the
core
OS,
and
it
was
a
basically
a
system.
V
talk
and,
and
at
the
end
of
that
talk,
Alex
came
up
and
said:
hey.
Where
can
I
download
the
coral
ask
and
they
said
well,
it's
concept
doesn't
really
exist.
B
B
So
anyway,
the
the
OS
core
OS
was
rebranded
a
couple
years
ago
as
container
of
Linux.
So
you
know
throughout
this
talk
that
we
say
container
lives.
We're
talking
about.
You
know
that
version
of
the
operating
system,
that
kind
of
grew
out
of
Chrome
OS
and
we're
going
forward
with
this
is
that
name
didn't
really
seem
to
catch
on
with
all
the
user
base.
A
lot
of
people
still
call
core
OS
and
moving
forward
we're
becoming
a
red
hat
core
OS.
B
A
B
B
B
Off
at
summer
2013-
and
it's
been
a
super,
successful
distribution
and
kudos
to
those
guys,
we
always
really
really
like
this
vision.
At
Red,
Hat
I
stole
a
few
slides
from
Brandon
Philippine
I
did
a
talk
at
the
summit,
so
definitely
want
to
give
him
credit
for
the
the
next
three
or
four
slides
here.
We
kind
of
went
a
little
more
in-depth
on
a
retrospective
looking
back
technology.
B
We
basically
separated
that
from
the
app
from
the
underlying
operating
system,
and
with
that
separation
the
the
result
is,
we
can
just
then
automate
updates
for
the
system,
and
this
is
this
is
very
true
and
proves
now,
you
know
quite
effective
with
the
user
base.
Even
people
with
you
know
poker
chips
or
Braille
systems
and
everything
else
have
had
experienced.
The
the
you
know
the
ease
of
upgrading
the
underlying
OS
from
the
actual
containers
themselves.
B
Now,
a
couple
other
things
on
the
premise
so
as
it
evolved
got
to
this
like
simple
architectures,
this
kind
of
simple
layer
cake,
if
you
will
always
kind
of
latest
bits
right,
do
you'd
have
a
really
a
new
kernel,
a
new
system,
be
you
know,
a
pretty
new
docker
engine
and
cool
it
right,
and
then
the
the
theory
was
that
all
of
these
will
kind
of
have
fairly
clean,
reasonable
separation
and
interact
well
with
each
other
again.
This
is
Brandon
slide.
I.
Really
it's
funny
answer,
I.
Think
I.
B
B
D
B
B
Now,
when
we
look
at
the
ecosystem,
side
and
kind
of
how
this
is
over
the
past
four
plus
years,
right
from
the
CN
CF
website,
I
think
it's
good
all
right,
I,
if
I
was
drawing
this.
The
kubernetes
logo,
which
is
kind
of
hidden
here
on
the
left
side,
would
be
about
eight
times
bigger,
but
anyway,
it
still
makes
a
point,
and
you
know
with
the
with
the
ecosystem.
That's
growing.
This
is
fast
with
so
many
so
many
projects
and
working
in
so
many
different
levels
and
layers
of
the
stack
right,
I
think
I.
B
Think
a
lot
of
people
have
this.
They
kind
of
see
this
and
it
always
has
that
sushi
menu
active
on
with
this
and
with
that
and
with
this
I'm
gonna
run
on
here
now,
I
have
my
dream,
stacked
and
I'm
ready
to
go
and-
and
that's
that's
challenging
right.
We
have
11
releases
of
open
ship
under
our
belts
and
I
think
the
reality
is.
Is
you
can
make
a
lot
of
these
pieces
pluggable,
but
it's
it's
really
really
hard
to
do,
and
and
I
think
the
reality
and
and
the
realization
like
we've
come
to.
B
Coupled
right,
yes,
there
are
API
interfaces
to
define
a
lot
of
these
interfaces,
but
especially
when
you
get
into
low-level
plumbing
of
the
stack
you
know,
containers
can
update
independently
of
the
OS,
but
separating
the
OS
from
platforms
is
actually
a
whole
nother
thing.
It's
far
more
complicated,
all
right,
and
so
with
that
is
a
backdrop.
I'm.
C
So
we've
we've
got
the
great
stability
of
the
rel
kernel
and
the
backing
of
the
support
organization
of
Red
Hat
that
can
really
be
delivered
and
a
little
bit
more
of
an
automated
and
controlled
fashion.
That
really
preserves
the
philosophy
of
container
Linux
while
giving
our
customers
the
best.
What
we've
been
delivering
for
years
right
and
next
slide.
C
Some
of
the
stuff
that
that
made
a
little
bit
of
room
for
on
the
right
hand,
side
is
the
stuff
that
we're
no
longer
considering
as
important
to
our
strategic
direction,
but
we
still
know
that
the
community
cares
about,
and
we
don't
want
to.
Just
you
know,
become
the
big,
bad
red
hat
and
just
kill
or
anything.
C
We
want
to
make
sure
that
we're
treating
it
responsibly
and
acting
nicely
for
the
community,
so
rocket
is
continuing
to
be
a
CN
CF
project
and
we're
trying
to
make
sure
that
the
people
that
want
to
use
it
continue
to
use
it
and
not
do
anything.
Drastic
fleet
we're
deprecating
and
that's
been
the
kind
of
model
for
the
proper
way
of
staying
hey.
This
is
something
we
don't
see
a
lot
of
future
for
so
we
want
to
make
sure
that
we
get
announcements
out
early
and
we
get
plenty
of
warning
even
with
plenty
of
warnings.
C
Sometimes
people
wind
up
surprised,
and
so
we
try
and
make
sure
that
we're
having
good
honest
conversations
about
locksmith
were,
as
it
says,
we're
moving
the
coordination
to
operators
and
running
this
a
little
bit
more
as
a
kubernetes
model,
letting
the
cluster
kind
of
drive
it
and
we're
trying
to
provide
alternatives
in
Fedora
chorus
for
non
kubernetes
environments
as
well,
because
we
thought
it
was
a
great
idea.
We
don't
want
to.
You
know,
get
rid
of
it.
C
It's
just
a
matter
of
what
should
really
own
the
cluster
and
what
should
really
own
the
nodes
on
the
cluster
matchbox
continues
to
be
a
community
project
for
pixie
and
ignition
configs,
but
we
don't
really
have
many
plans
for
things
that
we're
going
to
be
doing
with
it
the
same
with
torx
we're
not
really
keeping
the
update
model
that
we've
had
with
container
linux,
so
there
it's
just
not
as
relevant,
but
we're,
certainly
not
getting
rid
of
it
by
any
means
and
flannel
again
maintenance.
Only.
C
So
the
way
that
we
really
wanted
to
bring
this
together
is
to
make
sure
that
the
node
is
a
little
bit
more
managed
by
what
the
cluster
needs
at
that
time
right
and
the
way
that
we're
really
trying
to
deliver.
That
is
to
make
sure
that
we've
got
the
the
known
kernel,
systemd
and
selinux
that
you
get
from
just
a
regular,
real
host
and
add
in
exactly
the
version
of
the
cubelet
and
the
runtime
and
ignition
that
you
need
to
do
the
things
that
accrue
Burnette
each
cluster
wants.
C
We
want
to
kind
of
maintain
some
of
the
great
things
that
came
about
from
container
linux,
with
the
user
experience
and
just
trying
to
make
it
the
minimal
host
possible
and
totally
integrated
with
what
the
higher
levels
need,
and
we
want
to
deliver.
Automated
updates
and
CVE
remediation
through
the
automated
update.
C
Modeled
that
were
we're
kind
of
driving,
with
the
stream
of
updates
that
we're
going
to
be
delivering
to
the
cluster
we're
keeping
the
ignition,
because
it's
just
an
amazing
technology
that
does
a
lot
of
really
good
stuff
and
really
just
that's
the
turn
for
how
we're
going
to
get
nodes
up
and
running
right,
and
this
is
gonna,
be
fully
supporting
the
rail
API
and
the
ecosystem.
So
if
it
runs
on
row,
it's
going
to
run
on
RedHat
core
OS
a
container
right
next
slide.
C
So
some
of
the
things
that
we're
doing
differently
we're
dropping
cloud
init
in
favor
of
the
ignition.
So
we
know
that
there's
a
lot
of
tooling
already
kind
of
around
that,
and
so
we're
trying
to
be
mindful
of
how
we're
going
to
get
everything
going
correctly
with
ignition
and
make
sure
that
everybody
knows
that
this
is
kind
of
the
direction
we're
going.
C
C
And
we've
got
this
great
mechanism
for
doing
that
with
our
PMO's
tree,
where
it
pulls
down
a
container
it
boots
up.
The
container
pulls
down
the
updates
that
it
needs
to
the
node
and
then
reboots,
and
if
everything
goes
smoothly
and
everything
comes
up
and
then
we
go
and
sit
closer
than
great
and
if
it
doesn't,
then
we've
got
some
rollback
protection
with
our
pmos
tree,
where
it
will
roll
back
automatically
and
try
and
rejoin
the
cluster
and
make
sure
that
the
cluster
knows.
What's
going
on
with
that,
node
right
next
slide.
C
So
the
transactional
updates
that
used
to
happen
in
container
linux
were
more
along
the
lines
of
an
AV
partition
switch
where
they
would
downloading
image
to
the
other
partition
and
then
reboot
and
hope
that
it
comes
up
and
rejoins
for
what
we've
got
with
our
pmos
tree.
We're
using
what
we
feel
is
a
little
bit
better
of
a
mechanism
for
delivering
those
updates.
Our
pmos
tree
has
always
had
the
ability
to
be
in
the
middle
of
an
update,
and
you
could
just
pull
the
power
on
that
node
and
it
wouldn't
completely
just
destroy
the
node.
B
Like
well,
while
Jeff's
needles
pop
in
here
for
a
second,
so
we
when
looking
at
how
both
container
Linux
and
atomic
host
handle,
update
or
update
excusing
the
the
transactional
side
is
actually
one
of
the
biggest
of
these
of
these
hosts
right
that
you
know
one
of
the
things
on
like
a
Rd
invariant
system.
Obviously
it's
incredibly
robust
that
that's
all
good,
but
with
so
neat
about
this
is
you
you
pull
the
updates
in
the
back
crap
right,
so
it
never.
B
There
there's
never
a
point
on
the
system
where
you're
in
transaction,
so
there's
never
a
runtime
change
of
the
bits
right.
So
really
we
bring
all
of
that
immutability
of
the
what
we
experienced
for
containers
right.
We
bring
that
down
to
the
host
operating
system.
So
this
is
one
of
the
biggest
successes
of
this
model
and
absolutely
something
we're
going
to
be
carrying
forward.
C
C
We
we
really
want
the
cluster
to
be
able
to
know
when
it's
time
to
upgrade
that
we
can
move
to
the
next
version
of
kubernetes
or
even
a
minor
or
major
bump,
and
just
have
that
ability
to
have
that.
One
button
push
and
everything
just
rolls
out
and
everything
succeeds
and
provide
a
little
bit
of
stability
and
security
on
the
side
of
knowing
that
we've
done
it
in
advance
internally
before
it
gets
out
and
that
everything
is
going
to
go
really
seamlessly.
C
C
C
That's
out
there.
The
host
will
pull
that
computer
image
down,
pull
the
updates
out
of
that
container
image
and
then
do
the
reboot
and
if
things
don't
work,
it'll
roll
back
right,
we're
gonna
feature
pod
man,
mobi
docker
in
spawn
and
pretty
much
anything
else
that
somebody
wants
to
put
on
there.
We've
we've
got
a
mechanism
for
getting
feedback
from
the
community,
and
anybody
can
always
just
come
and
ask
us
and.
C
Next
slide,
we've
got
a
lot
more
technical
information
that
is
always
constantly
being
updated.
We've
got
talks
that
Colin
and
Benjamin
and
dusty
have
done
in
the
past
and
we'll
be
doing
again
in
the
future,
and
we've
got
the
Fedora
core
OS
tracker.
That
has
a
list
of
all
the
things
that
are
going
on
on
the
fedora
side
of
the
house
for
core
OS
and
I.
Think
that's
all
got
today's.
We've
got
some
questions
in
the
chat,
so
Ben
do
you
wanna
yeah,.
B
B
D
B
B
What
we
really
want
to
keep
those
use
cases
consistent
and
I
mean
it
should
be
almost
a
drop
and
then
for
anybody
on
the
open
ship
side
of
the
house
right,
if
you,
if
you
open
ships
right,
because
the
experience
of
everything
from
how
the
cluster
is
provisioned
to
just
this,
this
immutable
lifestyle,
I
think,
is
the
right
way
to
say,
then
all
of
that
is
being
brought
in,
and
so
the
the
entire
cluster
is
just
this.
This
self-driving
model
right
where
we
just
we
just
instantiate
it
and
boom
it.
B
B
D
B
Right
sergeant,
so
first
one
is
metering
in
charge
of
it
going
to
be
available
without
platforms,
and
the
answer
is
yes,
so
our
tectonic
had
this
beautiful
administrative
console
that
is
moving
into
open
ship,
so
it
does
have
our
back
on
it
right.
So
if
you're
developer
you'll
still
hit
the
current
or
you
know
these,
the
open
console
we've
had
for
a
while.
That
has
a
very
good
just
a
task
oriented
workflow
for
deploying
applications.
B
Now
that
said,
cloud
still
has
powerful
capability
in
to
you
know:
that's
not
changing,
but
yeah
prometheus
is
there
and
running
and
interacting
everything
from
cluster.
So
hopefully
that
answers
that
question
by
the
way
as
I
go
through
this
list,
if
I
screw
something
up,
just
I,
don't
feel
free
to
ask
another
comment:
I'll
try
to
pick
that
up.
B
Okay,
so
nice
little
Dennis.
We
have
coop
control,
yes
control,
so
we
always
have
OC,
which
basically
wraps
coop
control
and
by
the
way
that
is
the
canonical
pronunciation,
is
boob
control.
So
please
scold
anybody
who
says
coop
that's
technically
incorrect
from
the
guy
who
started
I'm
just
kidding.
We
would
never
scold
people
when
in
the
open
source
world
but
yeah
so
coop
coop
control
is,
is
there
everything
you'd
expect?
Is
there
and
again
OC
has
all
those
commands,
so
it's
fully
compatible.
It
just
adds
a
few
more
so
they're.
B
B
C
B
D
D
B
Can't
tell
you
that
Claire,
you
know
Claire
continues
to
be
an
investment
point
and
isn't
going
away.
So
you
know
that
will
give
us
from
above
scanning
on
the
registry
all
right,
Richard,
I
love.
This
quote
kubernetes
has
done.
That
would
be
a
world
record
because
I
think
that
would
be
the
first
piece
of
software
to
actually
be
complete
and,
of
course
say
that
happened.
B
Yeah
I
think
a
lot
of
a
lot
of
the
design
probably
could
be
considered
as
done,
but
I
I
would
not
consider
who
complete
by
any
means,
I.
Think
I
think
this
is
probably
one
of
those
things
work
who
you
talk
to.
You're
gonna
get
a
different
answer
on
where
I
just
depending
on
their
perspective
and
how
close
they
are
to
the
technology.
B
As
far
as
having
problems
with
updates
these
these
problems
kind
of
manifest
themselves
for
different
reasons
right,
sometimes
it's
sometimes
it's
changes
in
the
core,
but
sometimes
it's
changes
on
on
other
other
parts
of
the
stack
right
like
a
networking
tear
or
you
know,
for
example,
the
part
that
Jeff
and
I
work
on
all
the
time
is
it
contain
a
runtime
side
of
this.
All
these
projects
and
open
source
have
their
own
cadence
their
own
life
cycle
writing.
B
So
they
don't
always
line
up
with
with
Kubb
it's
one
of
the
big
reasons
we
started
the
whole
cryo
project.
Was
we
really
wanted
a
runtime
to
iterate
perfectly
anyway?
I
do
think
we're
actually
we're
definitely
closing
the
gaps
on
all
the
upgrade
stuff.
So
when
and
when
v4
comes
out,
it's
totally
a
game
changer
on
that
side
of
house.
So
all
of
us
are
really
excited
about
that,
but
yeah
I
don't
know.
B
C
So
no
Red,
Hat
chorus
is
not
going
to
have
a
10-year
Support
Lifecycle.
It's
going
to
be
supported
with
openshift,
though.
As
long
as
that
openshift
release
is
supported,
Red
Hat,
core
OS
running
and
that
release
will
be
supported.
Right,
I,
don't
know
the
current
support
terms
for
every
open
ship
release.
Ben,
do
you
know
those.
B
D
B
Right,
it's
basically
a
quarterly
cadence
and
I.
Remember
almost
a
year
ago,
coop
con
when
Tim
Hawkins
talks,
he
asked
he
has
a
packed
room.
You
know
how
many
people
want
that
Cades
too
slow
to
something
like
six
months
and
like
every
single
hand
in
the
room
was
so
that
was
interesting
feedback.
That,
of
course,
is
no
happened
and
has
been
changed,
but
you
know
what
the
work
we're
doing
around
upgrades
and
everything
is
really
targeted
to
where
customers
can
move
from.
You
know,
let's
just
say,
112
113
effortlessly
right.
B
Just
it's
one
click
the
note
cycle
through
oh
you're,
you're
up
and
running
right
and
it's
a
non-event.
We
really
want
this
to
work
so
so
with
all
the
effort
and
resources
around
this
getting
to
like
the
seamless
update.
So
we
really
want
them
to
be
a
non-factor
for
the
environment.
Sorry
Jeff
that
probably
wasn't
exactly
what
you
look
for,
but
yeah.
C
So,
let's
see
the
next
question
terraform
instead
of
ansible
to
bootstrap.
I
believe,
though,
I'm
not
sure
that
it's
still
terraform,
but
I
think
right
now.
The
installer
work
that
we're
working
on
in
public
is
moving
in
that
direction,
but
thanks
for
the
Red
Hat
chorus
version,
not
necessarily
the
bring
your
own
host
version,
we're
still
keeping
some
ansible
around
and
we
still
have
play
books
that
will
need
to
be
run
if
you're
running
unreal
and
both
paths
are
planning
on
being
supported.
Is
that
correct,
then
yeah.
B
C
C
C
Recreate
exactly
what's
going
on
in
a
customer's
cluster
and
know
with
certainty
what
the
capabilities
of
the
node
are
right,
so
it
shouldn't
affect
users
too
much.
The
the
same
functionality
and
same
features
should
be
there,
and
the
ability
to
customize
things
with
rel
will
still
exist,
but
it
may
require
a
little
bit
more
effort
and
work
on
the
part
of
the
operator
right.
Yes,.
B
D
B
D
D
B
C
B
C
Say
Chen
is
not
that
open
shift
on
role
is
still
remain
option.
Open
shift
on
rel
will
still
remain
an
option.
Yes,
we
will
still
continue
to
support
installing
open
shipped
on
the
rel.
That
is
a
big
focus
of
ours
and,
yes,
both
paths
are
definitely
very
important
to
us
any
details
about
the
integration
with
satellite.
D
B
That
part
of
the
process
to
be
involved
in
the
cluster
we're
very
quickly
moving
to
the
world
of
auto
scaling
clusters,
both
in
the
pile
and
on
Prem,
and
you
know,
scaling
out
and
scaling
down
with
registration
becomes
a
you
know,
almost
a
sticking
point
in
some
way.
Now
how
these
things
can
work
together
is
around
the
actual
update
streams
themselves.
You
know,
as
we've
said,
that
all
over
the
content
for
opiates
is
delivered
via
container
images
now
and
and
satellite.
D
B
A
really
great
job
marrying
container
repos
on
Prem,
and
so
we
see
that
as
one
of
the
biggest
ways
that
satellite
will
actually
tie
into
the
value,
but
it
actually
won't
be
that
mechanism
actually
go
apply
patches
to
the
cluster
right
that
is
driven
by
an
operator
in
kubernetes,
in
conjunction
with
some
of
the
other,
so
the
other
tooling,
like
Cincinnati
and
these
other
pieces.
So
so
it's
it's
a
I'm,
gonna
I
may
use
it
overloaded
term.
B
B
B
Okay,
the
next
question
I
see
sachin
by
the
way,
I
think
you'd
all
for
the
good
questions.
This
is
awesome
would
be
a
migration
patrick
customer
drink.
That
was
just
on
rel
to
coral.
Ask
we
are
not
planning
on
flipping
clusters
over
Tokoro
and
I
I'm,
not
gonna,
say
it's
something
that
we
we
could
look
at
adding
for
the
future,
but
it's
not
something
that
we're
gonna
have
right
away,
they'll,
be
interesting
to
hear
feedback
or
that's
something
that
you
guys
would
see,
value
and
I
think
with
with
the
new
model.
B
As
an
OVA
great
question,
we
we're
gonna
have
several
the
long
term
how
we
the
various
ways
we
release
ran
a
core
OS,
will
look
very
similar
to
the
format's
we
have
available
for
tacos.
Today,
though,
the
like
atomic
we've
got,
you
know
I,
so
we
have
a
cute
cow
being
DK
like
all
that
stuff.
All
of
these
different
platforms
are
gonna,
get
lit
up
for
rent
at
core
OS.
We
just
can't
do
it
all
at
once,
so
we're
really
trying
to
nail
a
specific
environment
in
the
end
user
experience.
C
B
C
B
B
It
is,
it
will
be
easiest
if
you're
running
like
rel
on
a
3d
environment,
we
will
be
able
to
take
you
to
rel
on
a
four
dot
X
environment.
That
is
no
problem.
If
you
want
to
get
a
zoo,
you
know
that's.
Basically,
if
you
move
that
way,
you
used
to
manage
and
control
the
operating
system.
What
he
always
had
with
rel,
which
is
you
know,
that's
almost
an
advantage
for
many
of
our
customers
and
they're
on
print
environments.
B
A
We're
getting
close
to
the
end
of
the
hour
here
and
I
just
backed
everybody's
time
and
hopefully
invite
you
guys
back
when
you're
closer
to
the
the
beta
release
of
core
OS
in
the
coming
months.
I'll
pin
you
to
a
date
today,
but
I
will
be
posting
this
video
up
on
our
YouTube
channel
RH
openshift
on
YouTube
and
along
with
the
slides,
which
I'll
ask
you
guys
to
email
me
or
as
a
PDF
or
just
send
me
Lily,
and
it
was
incredibly
informative.
A
So
I
can't
thank
you
enough
for
sharing
the
vision
and
the
details
around
this
because
it's
always
been.
You
know.
It's
been
an
interesting
Road,
merging
the
to
work,
work
pass
together
and
making
everybody
happy,
but
this
has
been
great
content
for
us,
so
I'm
sure
we'll
get
a
lot
of
people
emailing
you
questions
and
asking
asking
for
more
information
as
well
as
we
get
closer
to
the
beta,
the
thanks,
Jeff
and
Ben,
and
everybody
for
your
amazing
questions
very
good
and
yeah.
Well,
thanks.
A
There
will
be
a
3
311
release,
open
ship
Commons
briefing
coming
up
a
tip
on
the
18th
of
October.
The
blog
post
on
311
is
out
today
on
the
pod,
Alicia
calm.
So
there's
a
little
detail
there
to
take
up,
take
a
look
at,
and
we
hope
to
see
you
soon
back
on
another
briefing,
I'd
open
ship
Commons
or
at
the
upcoming
gathering
on
December
10th
in
Seattle.
If
you're
coming
to
kook
on,
please
join
us
there.
The
day
before.