►
Description
Virtual machines have been a part of OpenShift since version 4.5, when OpenShift Virtualization was released as a fully supported feature for all entitlements. As OpenShift Virtualization has matured it has become more and more integrated with other features of OpenShift, such as Pipelines, Service Mesh, GitOps, and more.
This stream we’re joined by Chandler Wilkerson from the OpenShift integration team, to look at how we can use VM-based application components hosted in OpenShift in a more container-like way. Not all applications can be containerized, but that doesn’t mean we can’t use, consume, and manage them as if they are!
Want to learn more?
- https://cloud.redhat.com/learn/topics/virtualization/?sc_cid=7013a000002qG1LAAU
- https://docs.openshift.com/container-platform/4.10/virt/about-virt.html?sc_cid=7013a000002qG1LAAU
- https://kubevirt.io/
Try Red Hat OpenShift: https://red.ht/3uVxkya
Follow us:
Andrew Sullivan https://twitter.com/practicalandrew
Jonathan Rickard https://twitter.com/jrocktx1
Chandler Wilkerson https://twitter.com/rhcwilkers
OpenShift https://twitter.com/openshift
Subscribe to Red Hat's YouTube channel: https://www.youtube.com/redhat/?sub_confirmation=1
About Ask an OpenShift Admin:
Do you have questions? We want to answer them! Every Wednesday at 11am ET, hosts Andrew Sullivan and Jonathan Rickard sit down to answer your questions about Red Hat OpenShift.
About OpenShift:
Red Hat OpenShift is an open source container application platform based on the Kubernetes container orchestrator for enterprise application development and deployment.
OpenShift on redhat.com https://red.ht/3rYz5ZS
OpenShift on GitHub https://github.com/openshift/
#RedHat #OpenShift #Kubernetes #ContainerPlatform
A
B
A
Yeah,
it's
we
were
just
chatting
a
minute
ago
in
in
slack
we're
both
kind
of
discombobulated.
Today,
it's
like
normally
on
stream
days
like
the
morning
I
block
off
my
whole
morning
and,
and
you
know
it
gives
me
time
to
focus
and
collect
my
thoughts
and
organize
everything
and
all
that
other
stuff,
and
today
it's
been
just
packed
with
meetings.
I've
been,
and
normally
I
can
blow
off
those
meetings,
but
it's
like
today
for
some
reason:
it's
been
no,
you
you
have
to
be
there.
You
cannot
skip
these
meetings.
So
that's
right.
C
A
Yeah
yeah,
you
know
it's
it's
a
beautiful
day
at
red
hat
right,
but
I
I'm
not
overly
concerned
about
that,
because
our
guests
today
are
our
two
awesome
gentlemen.
So
we
are
joined
by
chandler
wilkerson,
who
is
a
part
of
the
openshift
virtualization
integration
team,
as
well
as
john
herr,
who
they
are
they're
both
on
the
openshift
virtualization
integration
team.
A
So
this
is
something
that's
near
and
dear
to
my
heart.
I
spent
a
lot
of
time
playing
with
and
experimenting
with
and
just
testing
out,
openshift
virtualization,
because
I
just
think
it's
really
really
cool
to
be
able
to
to
put
vms
into
into
virtual
into
openshift
right,
so
yeah
all
right.
So,
oh,
I
forgot.
I
almost
forgot
john
chandler.
If
you
would,
if
you
wouldn't
mind
introducing
yourselves
I'll
start
with
john.
D
Yeah,
hello,
I'm
john
hurr,
I'm
a
senior
software
engineer
at
red
hat
been
working
with
openshift
for
several
years
now.
What
we
do
is
our
little
team.
We
we
integrate
different
products
and
different
technologies,
as
best
we
can
to
you
know,
show
how
to
better
use
the
product
and
all
the
great
features
of
it.
A
A
E
Okay,
yeah,
as
you
say,
my
name
is
chandler
wilkerson.
I've
also
been
working
with
red
hat
and
openshift
for
a
number
of
years,
most
recently
in
the
openshift
virtualization
and
just
kind
of
you
know,
spending
the
last
couple
years
doing
blogs
and
videos
on
kind
of
bringing
different
parts
of
the
whole
red
hat
portfolio
to
bear
on.
E
Big
engineering
org
that
does
all
the
actual
hard
stuff
in
virtualization,
so
we
just
take
that
nice.
You
know
puzzle
piece
and
then
try
to
fit
it
in.
You
know,
with
a
little
bit
of
boom
into
the
other
pieces
of
the
portfolio,
just
to
make
sure
that
everything
works
nicely
and
we
tell
people
how
easy
it
is.
A
Yeah,
I
find
your
team
to
be
really
interesting
because,
like
my
my
official
title
will
used
to
be
technical
marketing
manager.
Right-
and
you
know
you
talk
to
christian
one
of
the
other
hosts
for
you
know,
he's
the
host
or
one
of
the
hosts
for
the
get
ops
guide
to
the
galaxy
right
most
of
the
hosts,
jafar,
etc.
We're
all
technical
marketing
managers.
A
You
know,
alongside
product
management
and
product
marketing
and
kind
of
do
exactly
what
you
described,
which
is
you
know
we
we
take
the
technical
aspects,
we
make
flashy
demos
and
we
map
it
over
into.
You,
know
business
value,
but
you
all
sit
in
engineering
and
have
this
completely
different
view
of
the
world
which
to
me
is
really
awesome
because
you
have
that
kind
of
direct
line
into.
So
how
is
this
supposed
to
work?
And
I
really
want
to
showcase
this,
and
I
want
to
make
this.
D
A
A
All
right,
mara,
fetts,
openshift,
virtualization
or
red
hat
openstack,
especially
for
telco
workloads.
Oh
that's
an
interesting
question:
johnny!
Let's
let
we
can
talk
about
that
briefly
before
we
get
to
the
top
of
mind
topics,
so
I
can
tell
you
andrew's
opinion
here
so
law.
I
think
it
depends
on
your
long-term
strategy
of.
Are
you
moving
from
vnf's
virtualized
network
functions
to
cnf's
containerized
network
functions
so
with
a
a
vnf.
You
know
if
everything
that
you're
doing
is
vnfs,
there's
no,
you
know
goal.
A
There's
no
plan
to
move
to
cnfs,
then
you
know.
Openstack
certainly
makes
a
lot
of
sense.
Even
if
it's,
I
think
a
small
number
of
cns
openstack
probably
still
makes
sense,
because
you
can
deploy
openshift
into
openstack,
absolutely
nothing
wrong
with
that.
We
have
customers
doing
that
all
of
the
time.
A
But
if
your
goal
is
to
be,
you
know
more
cnf
oriented.
If
you
want
to
use
all
of
the
other
benefits,
you
know
all
of
the
or
should
say
benefits
because
they're
both
very
capable
platforms
that
have
pros
and
cons.
So
it's
not
necessarily
openshift
versus
openstack,
but
it's.
If
you
want
one
feature
set
or
one
type
of
set
of
capabilities,
kubernetes
oriented,
openshift
oriented
versus
you
know
openstack.
A
Then
you
know
that
certainly
makes
sense.
I
will
say
that
stack.
You
know
it's
10,
plus
years
old,
when
it
openstack
come
out
2011,
while
yeah
10
2011.,
and
it's
always
been
in
the
telco
space.
So
there
is
that
heritage
there
is
that
deep
expertise?
You
know
trust
in
what
it's
doing,
whereas
open
shift-
and
you
know
openshift
virtualization
for
vnfs
it's
a
little
bit
newer
from
a
I
will
say
from
a
manageability
perspective,
the
core
technology,
as
I'm
sure,
john
and
chandler,
can
point
out.
A
E
C
E
E
Right,
I
was
going
to
point
out
also
the
there's
kind
of
a
migration
aspect
to
this
as
well.
So
you
have
a
number
of
groups
that
are
looking
to
bring
the
microservice
architecture
into
their.
You
know,
edge
offerings
and
into
the
like
the
5g
network
kind
of
space
and
all
that
kind
of
stuff.
And
sometimes
you
know
all
your
development
work
is
into
the
latest
version
of
the
product,
but
you
also
have
to
carry
the
older
standards
as
well.
A
All
right
so
refit,
I
see
your
message
there.
So
hopefully
we
answered
your
question.
If
not,
please
feel
free
to
keep
asking
there
in
chat.
You
can
also
reach
out
to
us
social
media,
practical,
andrea,
on
twitter,
on
reddit
on
all
the
places
johnny
is
at
jrocktx1
or
you
can
reach
out
via
email,
andrew.sullivan
reddit.com.
I
don't
mind
at
all,
so
I'm
gonna
mike
and
mario
and
a
few
others,
I'm
let's
delay
until
after
the
top
of
my
topics
and
then
we'll
dive
into
those
questions.
A
So
for
anybody
else
who
happens
to
be
watching
us
or
listening
to
us,
I
will
put
out
a
reminder
that
this
is
our
one
of
our
office
hours.
Series
of
live
streams
here
on
red
hat,
live
streaming,
which
means
that
we're
here
to
answer
your
questions,
kind
of
regardless
of
whatever
they
may
be
about,
although
we
generally
try
and
keep
it
openshift
focused,
as
you
would
expect,
so
don't
hesitate
to
ask
any
questions.
Anything!
That's
on
your
mind.
We'll
do
our
best
to
answer
those.
A
If
we
don't
have
the
answer
and
johnny
is
brilliant,
but
he
doesn't
know
all
the
answers.
We'll
we're
happy
to
reach
out
to
the
appropriate
folks
inside
of
red
hat,
whether
that's
you
know,
product
management
or
engineering
or
or
whoever
that
happens
to
be,
and
then
we'll
follow
up
with
those
questions
at
either
the
next
stream
or
maybe
in
the
blog
post
that
we
put
out
after
each
one
of
these
streams.
A
C
Yeah
so
a
couple
days
ago,
I
think
it
might
have
been
monday.
Google
announced
that
they're
going
to
donate
istio
to
the
cncf,
which,
to
be
honest
with
you,
I
thought
it
had
already
been
done,
but
you
know
I
was
kind
of
shocked
to
hear
that
this
year
wasn't
part
of
cncf
but
yeah,
so
it
it
donated
the
project.
So
it
should
be
going
through
like
the
incubation
process
and
stuff
like
that.
A
A
A
Yeah
very
cool,
so
yeah
moving
up
in
the
world
right
becoming
an
official
cncf
project
and
all
that
or
getting
closer
to
becoming
a
full-on
cncf
project,
which
is
awesome
moving
down
our
list
here.
A
couple
of
upcoming
things
so
summit
is
happening,
may
10th
and
11th.
So
for
anybody
who
isn't
familiar
with
summit,
it
will
be
this
year
will
be
a
hybrid
event.
So
there
will
be
some
some
number
of
folks
on
site
and
it's
boston
right.
Johnny,
yes
yeah,
but
but
the
vast
majority
of
folks
will
be
virtual.
A
So,
for
example,
I
will
be
attending
virtually.
I
won't
be
there
on
on
site,
so
I
would
definitely
recommend
and
keep
an
eye
out
on
the
cloud.redhat.com
blog.
I
think
that
they
are
planning
on
posting
a
an
an
open
shift,
specific
list
of
of
information-
that's
happening
at
summit.
So
if
you
only
care
about
openshift
things
right,
cloudreddit.com
blog
there'll
be
a
blog
post
there.
That
says
you
know:
hey
here's
openshift
things
at
summit,
which
leads
me
into
my
my
next
one,
which
is:
where
did
my
window
go?
A
So
if
we
I'm
going
to
share
my
window
here
because
I
happen
to
have
the
blog
post
up
so
kubecon
emea
happening
in
valencia,
spain,
you
can
see-
is
the
week
after
summit
the
16th
through
20th
of
may,
so
they
went
ahead
and
published
this
blog
post
and
I'll
probably
push
this
into
our
chat
here,
so
that
oh
stefanie
is
already
on
top
of
it.
Thank
you
stephanie.
So
if
we
look
through
this
there's
a
huge
list
of
you
know,
red
hat
related
open
shift
related
sessions
that
are
happening
at
kubecon.
A
Honestly,
I
was
extremely
surprised
at
this.
I
was
expecting
like
a
dozen.
You
know
different
things
listed
here.
You
can
see,
there's
like
30.,
so
lots
of
stuff
going
off
going
on
here,
so
you
can
see
kubecon
itself
kicks
off
on
wednesday,
but
I
know
that
we
have
several
day
zero
day
negative
one
events
that
are
happening
so
openshift
commons,
which
you
can
see,
is
called
out
here.
There's
also
get
ops
con,
there's
also
gosh.
A
What's
the
other
one,
there's
at
least
one
more
that's
happening
as
well,
so
lots
of
stuff
that's
going
on
when
I
registered
for
kubecon
the
other
day
I
saw
like
openshift
commons
was
free
to
to
register
so
yeah,
there's
always
lots
of
stuff,
that's
happening
there.
I
think
they're
going
to
be
reviewing
the
roadmap
and
some
other
things
that
comments.
A
I
haven't
looked
at
the
full
list
of
sessions,
that's
happening
there
either,
but
just
an
fyi
for
our
audience,
those
two
weeks
in
order
to
not
compete
with
your
attention
for
either
summit
or
for
kubecon.
We
will
not
be
streaming
those
weeks
so
next
week
we
will
be
streaming.
We
have
the
migration
toolkit
folks
joining
us,
so
marco
berube
and
oh
gosh
names
are
hard.
Today.
I.
A
What
happens
when
I
don't
get
my
thoughts
organized
beforehand,
but
anyways
the
migration
toolkit
folks
will
be
joining
us,
we'll
be
talking
about
migration
toolkit
for
containers,
moving
from
openshift
3
to
openshift4,
moving
between
openshift
versions
using
using
the
my
creation
toolkit
so
as
stephanie
so
helpfully
put
it
on
the
screen
there.
Just
now,
please
do
subscribe
to
the
channel.
So
that
way
you
can
be
alerted
whenever
we
do
go
back
onto
the
air.
A
Yeah,
I
think
what
you
said
there
is
it's
it's
the
right
topic
at
the
right
time,
which
is
yeah
in
case
you
didn't
know,
or
if
anybody
is
still
on
openshift
3.
Remember
that
this
year,
I
believe
it's
july
is
when
open
shift
3
moves
into
the
last
stage
of
its
life
cycle,
so
now's
the
time
to
be
very
seriously
investigating
and
earnestly
pursuing
moving
off
of
openshift
three.
If
you
still
happen
to
be
on
it
all
right,
just
real
quickly.
Here,
I've
got
two
other
things
to
talk
about.
A
A
This
was
the
reason
that
upgrades
between
openshift
4.8
and
openshift
4.9
were
blocked
for
a
while.
So
that
has
been
rectified
as
of
openshift
version.
4.9.28,
which
you
can
see
is
the
current
version
in
stable
and
4.10.9
in
4.10,
so
upgrades
are
unblocked,
are
now
available
from
4.8
to
4.9.
They
were
never
blocked
and
4.9
to
4.10,
but
I
would
all
of
that
being
said.
I
would
definitely
recommend
that
everyone
update
to
either
4.9
or
4.
A
Please
do
update,
because
this
is
where
the
mitigations
for
that
issue
is
addressed.
So
if
we
look
down
here
in
the
bz
potential,
etsy
inconsistent
revision
and
data
occurs,
so
they
added
in
some
additional
checks.
Where
did
my
thing
go,
so
they
added
in
some
additional
checks
in
those
versions
to
prevent
that
from
happening,
the
eventual
fix
which
is
to
update
the
lcd
version
to
4.5.
A
or
4.3.
gosh.
I
keep
clicking
around
in
tabs
and
look
what
happens
to
me.
4.5.3
is
coming
coming
eventually,
but
short
term.
Please
update
to
those
latest
z
streams
inside
of
the
4.9
and
4.10
channels
to
to
protect
against
that
all
right.
Last
one
real
quick.
So
if
you-
and
we
have
had
this
come
up
a
few
times
internally,
if
you
are
interested
in
moving
your
virtual
m,
your
vmware
virtual
machines
hosting
an
openshift
cluster
between
datastores
there's,
a
couple
of
things
that
you
want
to
be
aware
of.
A
So
this
kcs
and
let
me
post
it
into
here,
so
stephanie
can
post
it.
So
this
kcs
can
walk
through
some
of
the
things
that
you
need
to
be
aware
of
some
of
the
process
that
you
want
to
follow.
So
this
usually
comes
up.
You
know
wholesale
migrating,
a
cluster
between
data
stores
when
folks
are
decommissioning,
a
data
store
inside
of
open
or
inside
of
vsphere.
A
You
know
moving
the
pvcs,
reintroducing
them
or
doing
a
migration
pvc
to
pvc
migration
using
the
migration
toolkit,
so
the
kcs
walks
through
all
of
that
stuff.
It's
a
great
read
if
you
happen
to
need
to
do
that,
but
if
you
just
want
to
expand,
you
know
grow
your
your
existing
openshift
cluster,
into
new
data
stores
or
into
new
drs
clusters.
You
can
do
that
as
well.
Just
using
machine
sets.
B
A
A
It's
all
right,
we'll
get
it
sorted
out.
Let's
see
I'm
scrolling
back
in
chat
here,
so
I
can
address
some
of
our
questions
so
mike
murphy
on
episode,
65,
it
was
mentioned
that
resource
requests
are
resources
being
held
by
the
scheduler.
What
would
the
scheduler
be
holding
those
resources
for
so
essentially
the
scheduler
doesn't
know
and
doesn't
really
care
what
it's
holding
them
for.
It's
just
what
the
pod
has
in
its
request.
So
basically
whoever
created
the
pod
definition.
Presumably
a
human
has
said
this
pod.
A
I
want
to
ensure
that
it
has
one
cpu
worth
of
resources
or
one
gigabyte
of
ram
whatever
it
happens
to
be,
and
the
scheduler
just
says.
Yes,
I
got
it
I'm
going
to
make
sure
that
there
are
that
many
resources,
it's
not
a
true
reservation,
but
available
to
it
when
I
schedule
it
to
a
particular
node.
So
it's
it's
really
an
arbitrary
thing
determined
by
the
pod
definition
and
whoever
created
it
and
I'll
john
or
chandler.
Well,
johnny
sorting
out
his
robotness.
E
Well,
yeah,
pretty
much
I
mean
I
came
from
a
high
performance
computing
background.
So
definitely
you
see
that
with
like
job
schedulers
in
the
hpc
cluster,
where
you
want
to
make
sure
you,
like
some
jobs,
may
spend
like
three
hours
with
you
know
a
tiny
ram
requirement,
but
then,
at
the
very
end,
put
everything
together
and
then
spike
up
to
something
huge
and
just
knowing
that
workflow
in
advance
and
asking
for
the
right
numbers
so
that
you're,
you
know
the
that
the
job
doesn't
get
killed
by
the
scheduler
for
exceeding
its
requirements.
E
A
Yep
yeah,
it's
funny.
You
say
spike
like
that.
I
way
back
when
I
had
a
developer
who
they
were
writing
python
scripts
and
the
apparently
the
story
that
they
told
me
was
the
dba
was
was
mean
to
them
and
wouldn't
give
them
a
database,
so
they
just
decided
to
serialize
everything
onto
disc,
as
you
know,
basically
take
a
dictionary
and
and
serialize
it
right
so
so
anytime.
This
application
would
fire
up
every
time
they
had
to
reload
it.
A
A
So
yeah
developers
sometimes
do
some
wonky
things,
mario
openshift
virtualization,
as
an
alternative
to
vmware.
So
we
get
that
question
quite
a
bit.
From
from
what
I
understand-
and
I
I
can
tell
you
andrew's
perspective-
and
that
is
openshift
virtualization
and
vmware
vsphere
are
they're
similar
and
they
have
similar
capabilities
in
that
they're.
Both
hypervisor
platforms,
they're,
both
capable
of
hosting
you,
know
pretty
much
any
virtualized
workload
that
you
would
like
the
difference
is.
You
know
in
particular
how
you
plan
on,
managing
and
consuming
those
virtual
machines.
A
So
the
kind
of
core
and
important
thing
to
remember
about
openshift
virtualization
is
it
is
first
foremost
and
inseparably
a
part
of
openshift.
So
if
you
don't
intend
to
use
openshift
for
all
of
the
things
that
openshift
does
without
virtualization,
then
it's
really
not
the
best
fit
right.
You
have
a
lot
of
features,
a
lot
of
complexity,
a
lot
of
things
that
you're
paying
for
that.
You
know
you
might
not
you
know,
might
might
just
get
in
the
way.
A
In
addition
to
that,
it's
also
a
dramatically
different
experience.
You
know
I
was
a
vmware
admin
for
a
long
time.
It
took
me
a
little
bit
to
wrap
my
head
around
these
concepts
and
map
them
into
you
know
the
concept
of
a
data
store
or
a
storage
domain
in
rev
terminology
into
vm
disks
are
now
stored
in
pvcs
and
there
is
no
monolithic
data
store
or
you
know
things
like
storage,
emotion,
aren't,
aren't
a
reality
or
storage
life.
Migration
aren't
a
thing
as
a
result,
so
there's
other
things
like
high
availability
is
handled
differently.
A
It's
handled
in
the
kubernetes
way,
not
in
the
way
you
would
expect
with
something
like
vmware,
so
you
really
have
to
understand
the
requirements
of
your
applications,
irrespective
of
virtual
machine
and
hypervisor
platform,
to
really
be
able
to
make
a
decision
on
that
so
I'll
I'll.
Let
the
experts
weigh
in
here
john
chandler.
If
you
have
any
any
thoughts.
D
So
yeah,
so
so
the
way
the
the
virtual
machine
start
is:
there's
actually
a
pod
tied
to
them.
So
when
you
start
a
virtual
machine,
there's
actually
a
pod
gets
generated.
That
gives
you
a
lot
of
the
control
and
access
into
the
into
the
virtual
machine,
so
it
is
not
similar.
To
I
mean
it
is
a
different
experience
as
you,
as
you
said,
andrew
so
so
those
are
things
to
keep
in
mind,
not
always
as
straightforward
as
you
would
think.
A
That's
something
that
kind
of
blew
my
mind
when
I
first
started
digging
into
openshift
virtualization
we're
putting
vms
in
containers
and
then
we're
basically
managing
that
container
through
the
pod
paradigm,
just
like
any
other
container
and
that
just
it's
one
of
those
like
but
wait.
We
put
pods
or
containers
in
vms,
not
vms
and
containers
right,
and
you
know
it
so
that
was
really
interesting.
I
don't
know
if
there's
anything
more,
that
you
want
to
or
or
can
say
about
that,
but.
E
E
And
so
you
you
get
the
pod
in
in
the
container
in
the
pod,
in
the
container
and
in
the
vm
go
on
down,
and
I
think
we
have
seen
some
attempts
to
install
openshift
on
vms
in
openshift
virtualization,
and
you
know,
however,
far
down.
You
can
take
that
of
course,
at
some
point:
you're
you're
virtualizing,
the
the
hardware
you're
not
doing
hardware
virtualization
anymore
you're
having
to
emulate
so.
B
A
Now,
I'm
putting
you
know,
containers
and
pods
on
containers
and
pods
running
on
the
physical
server
and
everything
that
I've
seen
and
heard,
and
I've
talked
with
with
him
about
the
customer.
That's
doing
that
they're
very
happy
with
it.
It
performs
exactly
the
way
that
they
expect
it
to
and
and
does
all
the
things
that
they
wanted
to
it's
just
today.
I
think
the
integration
isn't
there,
it's
a
non-integrated,
install
so
there's
no
csi
provisioner
for
the
underlying
physical
cluster
type
of
thing.
E
Yeah
there
there
was
a
a
topic
in
this
year's
cooper
summit,
which
was
in
february
on
doing
basically
kubernetes
on
coopert
and
that
kind
of
set
up.
So
I
know
it's
something
people
are
are
definitely
looking
at
and
you
know
at
first
you
kind
of
look
at
it
and
think
like
well.
Why
you
know,
are
you
just
doing
it
in
order
to
say
that
you
can
do
it
and
then
move
on?
E
A
Yeah
I
still
remember
going
to
it
was
kubecon
in
seattle
and
listening
to
the
ludsa
guys,
which
is
a
german
service
provider.
I
think
you
they
were
using
kubevert
to
create
virtual
worker
nodes
to
then
join
back
into
the
parent
cluster
so
that
they
could
basically
increase
pod
density
and
stuff
like
that,
as
well
as
provide
isolation
for
their
for
their
customers
for
their
tenants.
I
thought
that
was
really
cool.
A
A
I
only
remember
that.
Well,
I
remember
that
session
only
remember
was
in
seattle
because
my
wife
came
out
at
the
end
of
the
conference
and
we
got
to
tourists
around
seattle,
which
I
don't
usually
do
when
I
travel
see.
Mario,
I
see
that
stephanie
has
put
up
your
question
here
is
an
alternative
to
move
workloads
from
vmware
while
preparing
the
apps
to
be
cloud
native
and
use
openshift.
Is
that
a
good
path
to
follow?
I
would
say
yes,
potentially
you
know.
A
One
of
the
things
that's
interesting
to
me
is
you
can
use
with
openshift
virtualization.
Remember
the
as
john
said
the
vms
are
in
pods,
so
you
can
use
them
like
pods,
and
I
I
think
that's
one
of
the
things
that
we're
going
to
talk
about
here
in
just
a
moment.
So
things
like
you,
can
you
know
if
it's
connected
to
the
sdn,
you
can
use
a
service
to
connect
other
pods
to
that
virtual
machine.
A
You
can
expose
that
virtual
machine
via
a
route
or
so
it's
you
have
that
ability
to
interact
with
them
and
then
in
theory,
when
you
take
the
application
inside
of
that
virtual
machine
and
move
it
out
into
a
container,
you
know
it's.
It's
no
longer
virtualized,
it's
just
containerized
and
put
it
into
a
non-vm
pod,
a
regular
pod,
there's
no
change
from
the
way
that
the
other
things
interacting
with
it
or
talking
to
it
right.
It's
just
one
day,
it's
in
a
vm
the
next
day
and
it's
pot.
A
E
Yeah,
that's
one
of
the
things
that
I've
been
working
on.
So
you
know
some
some
of
our
workload
kind
of
goes
at
the
previous
release,
so
so
410
is
out
obviously,
and
so
our
our
kind
of,
like
you
know,
do
by
411
release
workflow
for
our
team
is
all
410,
stuff
and,
and
so
part
of
that
is
service
mesh.
E
And
so
you
know
we
we
had
the
the
service
mesh
team
on
like
a
number
of
weeks
ago
that,
and
so
I
started
looking
into
how
to
add
a
vm
into
a
service
mesh,
and
so
I'm
kind
of
in
the
middle
of
that
blog
post.
You
know
getting
that
all
done,
but
the
default
kind
of
toy
app
that
they
use
for
you
know
demonstrating
service
mesh
stuff.
Is
this
a
book
info
thing
where
it
has
like
three
or
four
micro
services?
E
And
you
know
you
you
go
into
this
page
and
it
says
there
are
a
couple
of
books
and
then
these
books
have
ratings
and
these
books
have
reviews
and
each
one
of
those
things
is
a
microservice,
and
so
I
was
able
to
set
up
one
of
their
toy
microservices
on
a
vm
and
kind
of
set
it
up
so
that
the
whole
service
mesh
should
work
with.
You
know
pointing
at
the
vm
as
opposed
to
microservice,
and
so
you
can
play
with
that.
So
you
know
the
general
idea
is
okay.
E
You
just
forklift
a
vm
straight
into
your
your
micro
shift,
phase
and
or
sorry
into
your
your
openshift
kind
of
microservices
architecture
application
and
you
can
concentrate
on
whatever
piece
of
it.
You
know
makes
the
most
sense
for
you,
while
letting
those
other
pieces
of
the
application
run
in
vms.
Until
you
get
around
to
you
know
my
you
know,
kind
of
upgrading
them
as
well.
A
Be
on
cloud.redhat.com
right,
slash,
blog,
okay,
we'll
it
we'll
include
a
link
to
that,
we'll
we'll
bring
it
up
in
the
top
of
mind
once
I
see
that
one
published
so
feel
free
to
ping
me
internally,
once
once
you
get
that
one
out
because
to
me
that's
that's
really
cool
right.
We
had
ort
went
on
a
few
weeks
ago
to
talk
about
service
mesh
and
all
the
stuff.
That's
there,
and
you
know
again
going
back
to
that.
You
know.
Yes,
it's
a
vm,
but
I
can
treat
it.
I
can
interact
with
it.
A
E
Right,
another
interesting
thing
is:
if
you
go
and
like
edit
a
deployment
right-
and
you
say-
okay-
well,
I
want
to
change
the
the
image
in
the
deployment.
Well,
it'll,
happily
go
on
the
back,
end
kill
the
pods,
and
then
you
know
fire
up
new
pods
or
you
know
start
to
fire
up.
New
pods
then
kill
the
pod
and
switch
over
vm
it'll,
just
kind
of
sit
there
and
say:
well,
you
know
it's
a
vm,
I'm
not
going
to
touch
that
for
you,
I'm
not
going
to
delete.
E
A
A
D
And
one
thing
to
note
is:
if
you
are
running
vms
and
you
go
into
your
virtual
machine
and
you
decide
to
shut
it
down,
you
know
it's
going
to
get
restarted
automatically
because
it's
tied
to
a
pod,
so
it
will
actually
restart
the
virtual
machine
so
all
right,
very
important
to
have
your
guest
fs
tools
installed
and
you
shut
it
down
from
within.
You
know,
openshift
and
then
it'll
do
a
graceful
shutdown
and
then
power
it
off
for
you.
A
E
There
is,
there
is
tooling
to
make
sure
that
the
behavior
works
as
you
intend.
It
is
just
the
default.
Behavior
is
always
run,
and
so,
if
you
kind
of
clone
one
of
our
template
vms
and
and
set
it
up,
and
then
you
think,
okay
well
I'll,
just
shut
down
the
vm
here
and
we'll
we'll
be
fine,
it'll,
just
pop
it
back
up
and
and
won't
even
kill.
The
pod
like
it'll,
just
come
back
up
so.
B
B
C
A
It's
just
different
behavior
that
as
an
administrator,
if
you're
not
used
to
it,
if
you're
not
prepared
for
that,
it
can
be
unexpected
behavior
that
that
may
not
get
you
what
you
want.
So
thank
you
stephanie
for
popping
up
another
question
from
kmx:
can
we
integrate
openshift
virtualization
with
a
fibre
channel
san
and
rev?
I
have
a
big
storage
domain,
so
I
think
the
answer
here
is
yes,
but
it's
going
to
be
different
than
you
would
expect.
A
So,
there's
no
storage
domains
right,
the
traditional
virtualization.
We
have
this.
You
know
massive
1,
2,
5,
10,
fifty
terabytes
storage
domain
and
we
dump.
You
know
dozens
hundreds,
thousands
of
virtual
machine
disks
onto
that
and
then
that
single
storage
device
is
mounted
across
all
of
the
hypervisor
nodes
with
openshift
virtualization.
Instead,
we
put
one
vm
disk
per
pvc,
which
would
be
a
lon
coming
from
each
one
of
those
mapped
to
a
single
one.
Coming
off
of
the
storage
device
fibre
channel.
Is
this
supported
storage
protocol
for
openshift?
D
E
A
Machine,
maybe
I
don't
know,
kmx
how
to
migrate.
My
600
vms
300,
terabytes
of
storage
from
rev
to
openshift,
so
yeah.
That
will
be
a
good
one
for
next
week.
We
can
talk
with
the
migration
toolkit
folks
about
that,
so
they
have
via
the
migration
toolkit
for
virtualization,
which
is
an
operator
included
with
openshift.
If
you
go
to
the
operator
hub,
you
can
search
for
it.
It's
in
there.
You
can
point
it
at
your
rev
environment
and
basically
map.
A
You
know
this
storage
domain
maps
to
this
storage
class,
this
source
network
maps
to
this
destination
network
and
select
the
vms
that
you
want
to
go
and
tell
it,
and
I
think
it'll
even
do
a
warm
type
of
cut
over
thing.
So
it'll
leave
the
vm
running
as
long
as
possible
in
the
rev
environments
and
copy
all
of
the
data
from
you
know,
storage
domain
to
pvc
and
then
turn
off
the
vm
do
the
last
copy
and
then
turn
back
on
the
vm.
I
think
right.
E
I
believe
you're
right
about
that.
You
know
kind
of
looking
at
the
questions
I
was
thinking
well.
The
one
piece
that
doesn't
really
that's
not
obvious
to
me
is
how
you
could
get
out
of
actually
having
to
copy
from
from
sand
volume
to
sand
volume
in
a
different
format,
and
I
think
you
know
maybe
you're
just
going
to
have
to
bite
the
bullet
on
on
some
of
that
300
terabytes
kind
of
thing,
yeah.
A
A
A
As
far
as
I
know,
things
like
network
policy
also
works.
So
if
you're,
using
things
like
acs
to
manage
communication
between
pods
between
different
things
inside
of
openshift,
that
would
apply
to
vms
just
as
well,
it's
just
a
vm
and
a
pod
right.
So
it
inherits
all
of
those
things
are:
live
migration,
seamless
for
openshift
virtualization.
A
Yeah,
so
when
I,
when
I
dug
into
this,
because
I
had
the
same
question,
which
was
wait,
a
minute
pot,
you
know
pods
terminate
pods,
don't
move
right
and
the
way
that
it
works
is
effectively.
We
schedule
a
second
pod,
so
the
the
vm
in
a
pod
is
running
on
host
a
we
say:
hey,
I
need
to
schedule
a
new
pod,
basically
with
the
same
characteristics
and
the
scheduler
says.
A
Okay,
I
want
the
new
pod
on
on
node
d
and
then
something
I
assume
the
operator
tells
those
two
pods
to
talk
to
each
other
and
it
uses
the
native
qmu.
Kvm
live
migration
mechanism
at
that
point,
so
it
moves
the
state
from
one
to
the
other.
So
I
think
that
the
requirement
there
is
rwx
storage
for
the
pvcs
because
it
has
to
be
mounted
on
both
source
and
destination,
which
some
of
you
might
say
like
the
person
just
a
minute
ago
right.
But
what
about
my
fiber
channel?
What
about
my
iscsi
block?
Storage?
A
Many?
Maybe
most
I
don't
know,
but
many
of
the
block
storage
vendors
do
support
rwx
pvcs
with
block
modes
pvcs.
I
specifically
said
block
mode
there,
not
file
system
mode
because
block
mode,
basically
the
the
the
host
the
hypervisor,
the
openshift
node,
gets
out
of
the
way
and
just
hands
it
over
to
the
pod
or
to
the
container.
And
since
it's
qmu
libert
it
it
manages
access
and
keeps
double
access
out
of
the
way.
I
guess.
A
E
A
D
And
you
might
even
get
a
you
used
years
ago,
older
versions
of
openshift,
when
I
did
that
you
actually
had
a
performance
increase
with
using
multis
on
a
separate
interface.
Now,
whether
you
do
that
now
or
not,
they
may
have.
Actually
things
have
gotten
a
lot
better
now
interesting,
a
lot
different.
So
things
to
if
you're
looking
for
performance
for
your
rdp
sessions,
you
know
it
might
be
something
to
test
in
your
environment
to
see
see
how
it
performs.
E
A
A
Wow,
so
I'm
gonna
have
to
think
about
that
one,
because
in
my
home
here
in
my
house
like
I,
I
use
a
mac
and
it's
an
m1.
So
I
I
keep
linux
vms.
I
keep
windows
vms
that
I
just
rdp
into
or
I
ssh
into
to
do
various
other
testing
and
stuff
like
that.
So
now
I'm
wondering
if
I
could
use
something
like
openshift
fert
to
do
that.
Yeah.
A
E
A
A
Let's
see,
mario
in
case
a
worker
goes
down
and
there
are
containers
and
virtual
machines.
Vms
are
migrated
to
another
worker
or
it
needs
to
be
manual
intervention.
What
kind
of
file
system
is
needed
to
do
that
so,
mario,
is
this
a
high
availability
situation,
or
is
this
a
graceful
maintenance
type
of
situation?
So
I
think
from
my
testing
in
my
lab.
If,
if
I
do
like
a
cluster
update
where
it
does
according
and
drain
of
the
nodes,
it'll
live
migrate,
those,
but
if
the
node
fails,
then
the
vms
are
they're.
A
Okay,
yeah,
I
would
expect
it
to
to
recover
just
like
any
other
pod
right.
It'll
it'll
go
through
the
scheduling
process
and
just
be
put
onto
one
of
the
hosts
right
and
there's
no
file
system
there.
Because
technically
I
mean
there
is
a
file
system,
but
it's
not
relevant
because
it
would
only
be
mounted
by
one
at
a
time.
In
that
case,.
A
B
A
A
C
A
Just
registered
for
summit
thanks
a
reminder,
yep
summit
and
summit
this
year
is,
I
think,
it's
two
parts
last
year
was
three
parts
this
year
it's
two
parts,
there's
the
one
in
may
that's
the
in-person
hybrid
event,
and
then
there
will
be
some
events,
a
a
distributed
group
of
regional
events
that
I
think
will
also
be
in
person
from
what
I
remember
so
in
more
ways
you
might
think
due
to
golang
x,
509
requirements.
Don't
forget
that
jeremy
cluster
is
500
fips
compliant.
When
you
stand
it
up
johnny.
C
C
A
A
A
E
A
C
A
Here
from
australia,
thank
you,
mario,
although
that
is
a
ludicrously
late
hour
to
be
up
listening
to
us,
but
we
do
appreciate
you
joining
us.
Johnny
we've
talked
about
it
before.
Maybe
we
should.
We
should
reinvigorate
that
conversation
of
occasionally
you
know,
maybe
once
a
quarter
or
something
doing
one
of
these
at
an
apac
friendly
time.
E
A
E
A
A
A
A
A
It
hasn't
been
that
long
ago,
like
maybe
six
months,
so
I
I
would
expect
that
with
the
proliferation
of
m1s
internally,
that
will
drive
the
priority
of
that
up.
Pretty
significantly.
I'm
surprised,
we
haven't
done
it
externally,
but
there
is.
There
was
a
thing
and
I
I
can
never.
I
always
lose
it
for
a
while:
the
x86
mac,
os,
cli
tools,
oc
and
and
cube
cuddle,
and
all
that
other
stuff.
A
They
would
fault
out
on
m1
and
it
was
a
golang
thing
and
there
was
like
a
certain
golang
flag
that
you
had
to
set
when
executing
it,
and
I
I
find
the
blog
posts
like
once
every
couple
of
months
and
always
remember:
oh
yeah,
that's
really
cool
that
that
actually
works,
and
then
I
promptly
forget
about
it.
So
I'll.
C
A
Yeah
it's
so
I
was
like
I
was
a
very
early
adopter
for
m1,
and
so
it
just
didn't
work
at
all
for
me,
and
it
was
one
of
those
like
I
I
have
like
an
x86.
You
know
rel8
vm,
that
I
just
use
and
that's
my
kind
of
bastion
or
my
jump
post
for
for
everything.
So
I
just
stopped
thinking
about
it
and
it
turns
out
that
there's
been
massive
progress
johnny.
A
You
were
talking
a
couple
weeks
ago
about
how
podman
works
on
the
m1
now
so
I've
just
I've,
never
taken
the
time
to
circle
back
and-
and
you
know
retry
and
relearn,
and-
and
you
know
actually
to
do
it
correctly.
Instead
of
just
working
around
my
problem,
christian
more
awake,
yep
are
you
christian?
Are
you
running
to
volunteer
volunteering
around
the
10
o'clock
stream?
Well,
christian
is
on
the
west
coast
of
the
u.s.
So
I'm
I'm
okay
to
throw
him
under
that
bus.
B
A
That
was
stephanie
for
anybody
who
hasn't
heard
her
before.
Thank
you,
stephanie
all
right.
So
I
think
we've
caught
up
on
on
chat
here,
so
chandler
one
of
the
things
I
wanted
to
ask
you
about
and
and
seeing
christian's
name
and
talking
about
christian,
reminding
me
this
you've
been
doing
some
work
with
get
ops
and
openshift
virtualization.
A
E
Yeah,
for
the
most
part,
there
was
actually
a
little
bit
of
integration
that
I
did
as
part
of
this
to
make
sure
that
the
argo
cd
gui
actually
understood
what
was
going
on.
So
you
know,
get
offs,
isn't
quite
as
simple
as
just
saying
there
should
be
a
manifest
on
the
cluster,
make
it
so
like.
Why
isn't
that
manifest?
E
And
so
you
know,
if
you
just
kind
of
had
blindly
or
like
I
did,
you
know,
throw
all
these
things
in
and
say
you
know,
go
make
it
so
argo
cd.
It
will
create
the
the
data
volume,
create
the
vm
and
then
expect
them
to
to
work
together,
while
the
vm's,
you
know
not
coming
up,
because
the
data
volume
isn't
there
yet
like
its
root,
disc,
isn't
isn't
available,
and
so
there
is
a
little
bit
of
tooling
that
I
added
just
in
terms
of
lua
scripts
wasn't,
which
was
an
interesting
learning.
E
Programming
language
exercise
basically
say
like.
Is
this
manifest
that
I
get
back
you
know
in
in
the
status
field?
Is
my
data
volume
there
progressing
in
error?
You
know
where
are
we
and
so
feeding
that
back
into
argo
cd
then
gives
it
the
ability
to
say
like
hey
this?
This
thing
failed
because
my
data
volume's
an
error,
or
this
thing
is
progressing
because
the
data
volume
is
progressing
and
that
kind
of
thing-
and
you
know,
wait
to
put
the
second
piece
in
before
you.
C
E
E
You
have
this
concept
of
the
data
volume,
which
is
a
little
bit
more
of
a
friendly
handle,
for
you
know
underlying
pvcs,
and
so
you
can
say
you
know
I
want
to
create
a
data
volume
that
will
get
its
content
once
bound
from
a
url
or
from
another
data
volume
or
another
pvc,
and
so
you
basically
can
declare
that
you
know,
for
instance,
in
in
openshift
virtualization.
We
have
this
concept
of
templates
of
vms
and
you
know
the
the
most
recent
kind
of
development
was.
E
A
So
yeah,
so
so
that
that
whole
data
volume
clone
thing
was
what
triggered
me
to
do.
My
last
like
lab
rehash,
because
I
I
needed
a
csi
provisioner
that
could
support
cloning
right
and
the
way
that
I
sort
of
figured
out
or
the
way
that
I
I
map
it
in
my
head-
is
the
data
or
cdi
the
data
volume.
It
is
what
abstracts
away
the
like
storage
operations
so
when
you
say,
like
yeah,
create
a
new
vm
based
off
of
this
template.
A
If
it's
a
csi
provisioner
that
supports
it,
it
says:
hey,
go
and
clone
this
for
me
and
it's
offloaded
to
the
storage
array
or
if
the
storage
provisioner
doesn't
support
cloning,
then
it'll
do
a
host-based
copy,
so
it
it
abstracts
all
of
that
into
one
single
set
of
operations
from
the
user's
perspective,
but
in
reality
there's
a
lot
of
there's
a
lot
of
intelligence
that
goes
into
that
to
to
have
it.
Do
the
right
thing.
E
Right
and
not
to
mention
the
snapshotting
in
snapshot,
restores
which,
which
is
a
really
interesting
concept
when,
when
you're
dealing
with
you
know
kind
of
virtual
machines
and
pods,
and
you
know
with
their
root
discs
and
pvcs,
you
want
to
take
a
snapshot.
So
there
there
was
an
orchestration
that
they
had
to
invent.
Basically
that
would
go
in
and
tell
the
guest
agent
all
right.
Do
a
an
fs
freeze
right
run
the
snapshot
unfreeze.
E
You
know,
stop
stop
writing,
and
that
includes
everything
from
like
sql
server,
all
the
way
down
to
the
the
bearer
os
components,
and
so
you
can
kind
of
you
know,
make
sure
that
that's
working,
but
that
that
mechanism,
you
know,
besides
giving
you
a
kind
of
like
a
live
snapshot,
backup
of
a
vm
in
case
you
want
to
restore
to
a
previous
point
in
time,
also
is
used
for
doing
those
data
volume
clones.
If
it's
available
in
in
your
underlying
storage
class.
A
E
E
C
A
C
The
cool,
the
cool
parts,
the
heritage
and
keeping
up
with,
like
the
you
know,
like
okay,
this
snap
goes
with
this.
This
virtual
machine's
root
disk-
and
this
goes
with
this
data-
does
consider
because
back
in
the
day
with
openstack
that
was,
that
was
a
big
issue
with
snapshotting.
It's
like
you
would
lose
some
of
that
heritage
and,
like
you
know
the
associations
and
stuff
like
that.
So
that's
that's
super
awesome!
That's
exciting!
Yeah.
E
There's
also
a
like
a
new
custom
resource
that
gets
added
for
openshift
virtualization
to
and
you
know
upstream
I
have
to
keep
remembering
like
okay,
it's
not
just
openshift
virtualization.
It's
also,
you
know
covert,
but
there's
a
vm
snapshot
object
and
a
vm
snapshot
restore,
perhaps
object
that
then
ensures
that
you're
keeping
track
of
all
the
drives
in
the
vm,
because
you
may
not
just
have
one
right.
A
I
see
stephanie
just
found
the
openshift
virtualization
playlist
on
youtube
there.
Thank
you,
stephanie.
Let's
see
a
couple
of
questions
here,
so
lester
ran
into
this
with
odf
in
one
of
our
patterns.
We
needed
to
wait
for
odf
pods
to
be
completely
ready
before
we
could
run
certain
jobs
so
lester's
on
your
team
right
tony.
C
Yeah
yeah,
so
what
he's
talking
about
is
like
before
you
can
actually
do
anything
else.
Right,
odf
had
to
be
completely
deployed
before
you
could
do
like
some
like
self-object
store,
user
integration,
and
so
the
way
we
were
doing
is
we
had
ansible
just
going
out
and
waiting
just
pulling
the
nuba
endpoint
deployment
and
wait
till
it
says:
okay,
I'm
good
to
go
so
the
status
says
that
you
know
it
gives
a
little
fist
bump.
Then
then
we
can
move
on.
E
E
C
A
A
A
I
think,
regardless
of
whether
it's
a
vm
or
a
pod
or
an
application
it
abstracts
away.
You
know
here
I
need
to
back
up
my
application
components
into
one
thing:
I'm
describing
this
really
poorly,
I'm
sure,
if
they're
great
sure,
if
the
oadp
guys
are
listening,
they're,
all
cringing
and
and
like
you
know,
they're
gonna,
be
I
have
everything
silenced
right
now,
they're
sending
me
you
know
rude
messages,
but
it
provides
that
abstraction.
A
So
it's
it
doesn't
matter
which
storage
vendor
you're
using
it
doesn't
matter
which
backup
vendor
you're
using
it
should
work
so
long
as
they
support
support
that
api.
Endpoint.
Don't
mention
that
app!
Our
storage
folks
decided
to
reject
our
trident
backends,
removing
the
media
types
there's.
What's
the
ssd
hdd
right,
it's
just
a
couple
of
letters.
It's
fine!
It's
no.
A
A
I
see
that
come
up
periodically
again
kind
of
what
I
was
talking
about
in
the
top
of
mind
of
hey
we're
decommissioning,
this
storage
array.
We
need
to
move
all
of
these
pvcs
from
one
to
the
other
and
today
there's
no,
the
most
efficient
way
that
I
know
of
to
do
it
is
migration
toolkit
for
containers.
I
I
I
can
assume
that
like
well
I'll,
pick
on
netapp
again
so
netapp.
I
think
they
can
move
volumes
transparently
at
the
storage
array
level.
A
B
A
Is
13
past
the
hour?
I
do
want
to
be
cognizant
of
everyone's
time,
not
just
here
on
the
stream,
but
everybody
who's
watching.
So
we'll
give
it
a
couple
more
minutes.
I
john
I
wanted
to
ask
you.
You
said
before
we
started
that
you're
doing
some
stuff
with
windows.
I
wanted
to
ask
about
some
things
there,
but
we
have
just
a
couple
more
minutes.
Anybody
who
has
any
questions
please
feel
free
to
submit
those
in
clarifications.
A
If,
if
we
accidentally
missed
your
question,
anything
like
that,
please
please
do
resubmit.
Those
if
we
don't
have
a
chance
to
get
to
your
question
today
or
if
we
didn't
answer
it
adequately
or
anything
like
that,
please
don't
hesitate
to
reach
out.
I'm
super
easy
to
get
to
andrew.sullivan
redhat.com
more
than
happy
to
answer
those.
A
I
also
have
had
a
few
people
reach
out
to
me
on
twitter
recently
at
practical
andrew.
So
please
that's!
You
can,
of
course,
send
me
a
dm
there
if
you
would
like
or
johnny,
and
I
are
both
on
reddit
as
well,
these
same
usernames
as
twitter,
so
yeah,
don't
don't
hesitate
to
reach
out
to
us
in
any
of
the
various
ways
and
we're
happy
to
help.
A
Although
I
do
apologize
because
for
for
people
who
reach
out
to
me
on
twitter
there's,
usually
a
delay,
because
I
had
to
turn
off
like
all
of
the
alerts,
because
it'll
send
me
an
email
like
every
time
somebody
likes
a
post
or
does
anything
like
that
which
got
way
too
noisy.
So
sometimes
it
can
take
a
little
bit
for
me
to
see
that
I
have
a
message
and
and
respond
to
you,
but.
A
Yeah
yeah
so
yeah
john.
I
I
wanted
to
ask
about
windows,
and
you
know
I,
I
think
one
it's
interesting
big
picture
to
see
how
much
cooperation
and
togetherness
between
microsoft
and
red
hat
there
is
these
days
right.
You
know,
azure,
red
hat,
open
shift
and
you
know
windows.
Containers,
windows,
nodes
is
a
part
of
openshift
all
that
other
stuff,
so
windows,
vms
in
openshift,
virtualization
one.
It's
very
much
a
thing
and
two.
D
D
So
I
just
wanted
to
make
sure
everybody's
aware
that
it
it
does
support
windows
windows
will
run
inside
openshift
virtualization,
which,
with
some
of
the
questions
you
know,
others
apparently
know
that.
So
that's
that's
great.
So,
but
yeah
you
can
you
can
do
you
can
do
syspreps?
You
can
do
several
other
cool
things.
D
There's
a
live
guest,
fs
utility
that
you
can
use
against
not
only
just
windows,
vms,
but
your
your
your
linux,
vms
to
sys,
prep,
your
linux,
vms
and
such
and
and,
like
I
said,
I'm
not
the
one
working
on
that.
I
just
wanted
to
make
it.
You
know
visible
that
you
know
it
is
supported.
Windows
is
supported,
there's
a
tooling
out
there,
for
it
give
it
a
try-
and
you
know,
as
always
upstream
community
is
a
great
place
to
give
feedback
or
to
red
hat.
C
C
Okay,
no
that's
awesome
and
then
my
my
next
question
is
really
low
level
literally,
and
you
know
technically
but
like
when.
Do
you
know
if
you
can
like
interrupt
the
boot
sequence
on
like
a
windows
machine
to
say,
like
maybe
hook
up
like
a
debugger
or
whatever
and
actually
kind
of
like
get
into
like
a
debug
console
through
openshift
virtualization?
Because
I
know
with
like
openstack,
it
was
really
complex
and
almost
impossible,
but
I'm
just
wondering
if
that's
even
a
possibility
within
openshift.
D
C
A
I'd
have
to
go
and
look
up
the
kcs.
That
has
the
specific
versions,
but
it's
it's
limited
to
windows.
Server
windows
desktop
and
I
think
rel
has
officially
fully
supported,
but
it's
kvm
so
like
basically,
every
operating
system
you
can
imagine
will
run
it's
just
not
necessarily
supported
by
red
hat.
A
E
A
A
A
All
right,
so
I
I
won't.
I
won't
bore
anyone
with
any
more
of
my
drivel,
so
thank
you,
everyone
for
joining
us
today
here
on
the
stream.
Again,
if
you
have
any
questions
any
other
comments,
please
don't
hesitate
to
reach
out
to
us
more
than
happy
to
help.
However,
we
can,
with
that
just
remove
the
tpm
from
the
config.
A
I
guess
that'll
work,
so
please
don't
hesitate
to
reach
out
again
next
week,
we'll
be
talking
with
the
migration
toolkit.
Folks,
we've
got
already
got
a
number
of
questions
coming
from
this
stream
that
we'll
be
addressing
next
week,
so
we
look
forward
to
having
them
join
us,
but
I
don't
want
to
underplay
or
or-
and
I
can't
thank
you
enough,
john
and
chandler
for
joining
us
this
week.
This
has
been
super
informative.
A
I've
learned
quite
a
bit
about
openshift,
virtualization
and
now
there's
a
bunch
of
stuff
that
I
need
to
figure
out
when
I'm
gonna
have
time
to
to
to
test
it
out.
I
I've
already
rejiggered
my
home
lab
once
or
twice,
and
it
sounds
like
it's
time
to
do
it
again.
So
thank
you
so
much
for
joining
us
for
our
audience
again.
Thank
you
for
joining
us
today.
We
will
follow
up
with
a
blog
post.
A
C
Yeah
thanks
again
chandler
and
john.
This
is
yeah
super
exciting,
especially
from
the
nerd.
You
know
like
just
it's
I'm
getting
triggered
wanting
to
do
stuff
now
this
is
pretty
awesome
and
then
you
know
to
all
the
all
the
viewers,
thanks
for
all
the
questions.
This
is
what
you
know
makes
this
exciting.
So
thank
you
very
much.
That's
it
for
me,.