►
From YouTube: KubeCon Office Hours: OpenShift 4.6 Features
Description
KubeCon NA 2020 Office Hours about new features in OpenShift 4.6
A
Good
morning
good
afternoon
good
evening
and
welcome
to
another
episode
of
open
shift
tv,
I'm
chris
short
executive
producer
of
openshift
tv
and
we
are
having
a
very
special
kubecon
office
hours.
We're
gonna
be
talking
about
some
features.
Some
new
features
in
openshift46
and
I'm
gonna
hand
it
off
to
josh
burkus,
to
kind
of
give
the
lay
of
the
land
and
introduce.
A
B
Hey
welcome
everybody.
This
is
our
last
office
hours
accompanying
the
kubecon
event.
As
many
of
you
know,
we
had
openshift
4.6
come
out
just
before
kubecon
and
there's
a
ton
of
features
in
in
4.6
integrating
cloud
native
technology
from
all
over
the
ecosystem,
the
and,
as
such,
I,
I
invited
the
openshift
developer
advocacy
team
here
to
talk
to
you
about
what's
in
4.6
and
answer
your
questions
about
4.6
and
about
openshift
in
general
right.
These.
B
C
Yeah
sure
so,
hey
everybody,
I've
been
on
the
developer
office
hours
a
few
times.
You
might
recognize
me
if
not
hey
nice,
to
meet
you,
I'm
brian
tannis.
I
am
part
of
the
openshift
developer
advocates
team.
I
focus
on
developer
advocacy
in
openshift.
Basically,
what
I'm
doing
is
trying
to
figure
out
ways
to
make
openshift
easier
for
developers
and
share
that
experience
with
you
guys
if
you
have
any
questions
or
whatever,
please
feel
free
to
put
them
in
the
chat.
C
I
definitely
want
to
stress
that
this
conversation
can
be
dictated
by
how
the
chat
goes
and
we
would
like
it
to
be
its
office
hours.
We
have
you
know,
general
ideas
that
we
could
talk
about
of
the
openshift
4.6
release
which
we'll
get
to,
but
please
do
type
those
questions
out
so
that
we
could
go
ahead
and
you
know
get
sidetracked.
That's
the
whole
point
it's
office
hours
with
that.
I
will
pass
it
off
to
to
jason
jason's
my
the
manager-
I
guess-
maybe
I
don't
know
we'll
see-
I
guess
yeah.
D
D
If
you
do
have
questions
in
chat,
there's
things
you
want
us
to
focus
on.
This
is
a
very
loose
script.
We
have
things.
We
certainly
are
happy
and
willing
to
talk
about,
but
this
is
one
of
the
few
avenues
in
this
awful
awful
virtual
world,
where
we
can
be
back
and
forth
with
people,
so
by
all
means,
take
advantage
of
that
and
then
last
cedric
can
I
introduce
yourself.
F
Yeah
absolutely
so
my
name
is
cedric
laverne,
I'm
the
developer
advocate
intern
and
I've
been
fortunate
enough
to
be
on
here
before
talking
about.
I
think
the
thing
was
podman
or
something
like
that,
but
I
have
a
series
on
red
hat
developer
on
like
getting
started
with
a
bunch
of
different
things.
Pod
of
build
man.
F
Odo
just
did
some
stuff
with
tekton
or
pipelines,
and
it's
it's
weird
being
here
because
normally
I'm
I'm
the
student.
You
know
in
office
hours
doing
my
exam
right
now,
so
I'm
happy
to
be
here
on
the
other
side
of
the
spectrum
and
yeah,
as
jay
said,
you
know,
this
is
a
whole
interactive
thing.
So
we're
happy
you're
here.
C
G
Yeah,
I
did
have
a
talk
on
the
breakout
track.
Colt.
Thank
you
very
much.
Thank
you.
Yeah.
It's
called
admission
control.
We
have
a
problem.
It
was
kind
of
a
really
an
introduction,
not
a
very
advanced
course,
but
feel
free
to
check
it
out.
Just
trying
to
highlight
the
purpose
of
the
admission
control
feature
on
the
kubernetes
api.
This
is
kind
of
the
front
piece,
one
of
the
major
pieces
of
security
for
validating
what
what's
allowed
to
go
into
your
etcd
data
store
at
the
at
the
core
of
kubernetes.
G
It's
a
feature
that
we
at
red
hat
use
pretty
extensively
to
try
to
make
multi-tenant
safe
clusters.
I
didn't
go
too
much
into
detail
on
on
how
red
hat
uses
it,
particularly.
I
try
to
keep
it
kind
of
generic,
but
really
interesting
feature
if
you're
looking
at
standardizing
clusters
or
trying
to
lock
down
certain
operational
aspects
of
the
cluster,
so
yeah
definitely
hit
me
up.
If
you
have
follow-up
questions
on
that
topic,
I'm
ryan
j
most
places
online
and
I
think
the
breakout
session
is
still
available.
G
You
can
also
put
questions.
I
think
I
think,
still
into
the
q.
A
for
the
session,
possibly
until
the
the
end
of
the
kubecon
week
and
then
there's
also
the
kubecon
slack,
I
think,
there's
a
specific
track
on
extensibility,
where
there's
a
lot
of
people,
keeping
the
conversation
about
that
topic
going
so
yeah
a
pleasure
to
be
part
of
an
honor
to
be
part
of
the
kubecon
lineup
this
week,
really
cool
I've.
I
had
the
opportunity
once
before,
but
yeah
this
was.
This
was
great.
E
D
B
Yeah
so
well,
yeah
well,
wren,
wants
to
know
well,
wren,
just
upgraded
to
ocp
463
built
a
mirror
catalog
and
a
restricted
network,
and
the
cluster
logging
operator,
which
I
guess
they're
trying
to
install,
does
not
have
a
4.6
channel
only
shows
4.5,
as
the
latest
is.
That
is
that,
where
we
are
right
now
with
that
operator,.
D
A
F
D
Building
them,
but
yes,
cheap
plug,
but
there's
a
funny
transition,
because
one
of
the
things
we
don't
typically
talk
about
is
one
of
the
benefits
of
how
it
decouples
all
of
these
features
from
the
platform
itself.
So
as
compared
to
I
mean
the
monoliths
use
the
general
terms.
We
don't
have
to
ship
openshift
with
all
of
these
extra
pieces
at
the
same
time
and
then
coordinate
dev
schedules.
Then,
oh
sorry,
you
missed
4.5
that
sucks
you're
going
to
be
waiting
until
4.6.
D
All
of
these
streams
are
now
independent.
So
these
extra
things
you
add
on
your
serverless,
your
service
mesh,
your
logging
operator,
things
like
that.
All
of
that
can
have
its
own
independent
release
cycles.
So
I
mention
all
of
that,
because
the
naming
on
the
channels
is
completely
up
to
the
operator
maintainer,
so
they
could
do
as
many
have
done.
4.3
4.4
4.5
they
could
have
used
their
own
versioning
scheme.
D
I
don't
for
a
fact
know
that
serverless
does
this,
but
as
an
example,
they
can
use
1.10,
1.11
and
so
on,
so
is
it
expected
possibly
is
the
best.
I
can
give
you-
and
I
know
brian
unmuted,
so
I'm
going
to
give
it
to
him
in
a
second
because
he
may
have
a
better
answer,
but
the
benefit
is
what
I
want
to
point
out
is.
This
is
not
necessarily
a
bug
out
the
gate.
This
does
not
say,
oh,
since
this
is
locked
at
4.5.
Is
this
broken
or
is
the
operator
not
updated?
D
It's
possible
because
of
this
intention
of
having
different
streams
come
out
from
them
that
they
either
didn't
need
to
make
changes
or
they
knew
it
would
work
so
they
weren't
in
a
rush
to
make
changes
and
based
on
their
numbering
scheme.
It
looks
a
little
wonky
that
you're
saying
give
me
the
4.5
stream
onto
4.6,
but
at
the
end
of
the
day,
it's
fine
and
it's
it's.
C
The
operator
life
cycle
manager,
you
know,
will
make
sure
that
in
some
extent
that
you
know
hey
if
the
operator
can
be
installed
and
we'll
work
here
like
it,
you
know,
might
handle
some
of
that
stuff.
Just
like
what
you
said,
jay
a
lot
of
the
times
like
it's
very
up
to
the
specific
operator
and
the
team,
that's
building
that
operator
serverless
has
the
same
way.
I
guess
that
the
the
monitoring
or
the
logging
operator
is
ocp-4.5,
dash,
4.6,
etc.
Whenever
a
new
release
comes,
you
know,
generally,
you
know
some
are
updating
immediately
right
away.
C
Some
might
do
a
little
more
testing
and
then
release
I've
seen
that
happen
with
maybe
like
the
openshift
pipelines,
one.
It's
also
one
where
the
channel
you
match,
with
the
specific
version
of
openshift
that
you
have,
but,
on
the
other
hand,
there's
other
operators
that
are
doing
it
like
openshift
service
mesh
and
they
have
a
stable
channel.
Whenever
updates
come
out,
service
mesh
2.0
came
out
recently,
which
we'll
talk
about
some
but
service
mesh
2.0.
C
The
operator
got
updated
right,
but
that
doesn't
mean
that
your
service
mesh
version
and
what's
actually
being
deployed
and
stuff
got
updated.
So
I
know
that
there
were
a
couple
like
really
concerned
people
that
were
out
there
saying
hey.
You
know
I
I
want
to
use
service
mesh
2.0,
but
not
yet
not
yet,
but
the
operator
got
updated
because
there's
only
a
stable
channel
there,
but
everything
was
okay.
C
The
operator
version
of
that
just
got
updated,
but
the
specific
the
specifics
are.
You
know
up
to
this.
You
know
the
operator
and
the
team.
That's
building
that
whenever
the
pipelines
one
came
out,
save
for
openshift
4.5
when
that
was
released,
there
was
only
four
dot
open
sure
the
ocp-4.4
version
of
the
operator
for
pipelines
available
for
maybe
a
week
or
two,
and
that
generally
was
okay
things.
B
B
I
actually
wanted
to
put
it
to
sort
of
a
panel
thing
and
say:
hey
so
openshift
is
a
distribution
of
kubernetes
uh-huh,
there's
a
clarification
for
martinko
right.
We
we
are
distribution
of
kubernetes,
so
I'd
like
actually
each
of
you
to
give
like
one
major
feature
that
or
integration
that
openshift
supplies.
B
F
Never
it's
it's
completely
on
the
cloud,
and
so
I'm
never
having
to
to
do
a
lot
of
manual
things
that
I
had
to
do
previously,
because
a
pipeline's
when
you
install
it
because
it's
an
operator
from
the
operator
hub
it's
it's
super
great
super
easy
to
install.
It
does
everything
for
me.
So
that's
something
that
that
openshift
only
provides
and
that
I've
been
having
a
lot
of
fun
with.
D
G
But
we
also
contribute
to
and
help
maintain
these
upstream
products
and
projects
and
and
kind
of
keep
the
whole
thing
going,
and
I
think
that's
really
important
important
with
this
cloud
native
ecosystem
is
having
a
lot
of
vendors,
be
able
to
collaborate
together
and
and
red
hat
has
really
with
operator
hub
in
some
ways
really
kind
of
been
leading.
The
way
of
getting
other
vendors
that
are
not
os
providers
or
or
kubernetes
maintainers
to
jump
into
the
game
with
operators
and
other
features
to
kind
of
extend
the
experience.
G
C
I'll
I'll
add
real
quick
to
that.
This
isn't
my
point,
but
I'll
add
real.
Quick
to
that
is
with
openshift.
It
doesn't
matter
if
we're
using
aws
gcp
on
premise:
it's
it's.
It
runs
everywhere
and
it
runs
the
same
and
that's
a
really
big
deal,
and
it's
not
just
it
runs
on
aws
or
it
you
know,
runs
on
gcp.
C
There
are
operators
that
are
out
there
that
work
with
openshift
and
are
supported
with
openshift.
That
allows
us
to
take
aws
resources
and
use
them
insanely
easily
within
openshift,
and
that
really
goes
to
like
my
point
of
developer.
First
right,
we
we
really
really
strongly
have
been
pushing
the
developer
first
centric
things
with
more
recent
versions
of
openshift.
The
developer
console
the
way
that,
like
me,
I'll
talk
to
friends
that
that
don't
work
on
kubernetes
they
they
develop
applications.
They
do
front-end
development,
they
do
back
in
development.
C
The
number
one
question
that
I
get
is
you
know
if
they're
not
familiar
kubernetes
seems
really
difficult.
There's
a
lot
of
things
going
on.
How
do
I
just
deploy
an
application?
How
do
I
get
something
going
and
then,
where
do
I
go
from
there?
Openshift
now
with
openshift
4.6
guides
you
along
the
way.
It
really
does
you
log
in
the
first
time
and
there's
a
guided
tour
that
walks
you
through.
C
This
is
what
the
developer
console
is.
This
is
how
you
see
all
the
things
deployed.
This
is
how
you
deploy
an
app.
It
makes
it
really
simple,
and
because
of
the
ability
to
have
operators
and
extend
that
and
that's
very
solidly
built
into
openshift,
with
the
way
that
you
know
we
we've
been
gathering,
you
know
how
the
operators
work
and
all
this
the
way
that
the
developer
console
and
the
administrator
console
works
right.
It
makes
it
easy
so
that
we
can
extend
that
functionality
and
use
additional
services
right.
C
Maybe
I
don't
necessarily
care
about
manually
pulling
triggers
right.
I
want
my
application
to
scale.
All
I
got
to
do
is
install
the
openshift
serverless
operator
like
seriously
and
then
deploy
it
as
that
application
is
serverless
that
application
instantly
scales.
I
don't
need
to
know
yaml
right.
I
could.
I
could
you
know,
deploy
my
application
via
the
web
console
in
openshift,
which
is
award
winning
as
well
as
within
a
cli
tool.
There's
there's
kn
right,
it's
k
native,
but
you
know
we
do
things
to
make
it
more
open
shift.
C
Centric,
just
like
ryan
mentioned
openshift
is
multi-tenant
right.
I
don't
have
to
have
separate
clusters
for
separate
things.
We
can
have
multiple
users
use
the
same
thing
in
an
easy
to
use
manner,
and
it
really
is
easy
to
use.
It
really
is
that's
really
a
really
important
aspect
of
what
we've
been
trying
to
do
with
this
developer
console
and
how
things
work
with
openshift.
C
You
know,
if
I'm
going
changing
code,
I
could
use
odo
to
watch
the
changes
and
make
sure
that
they're
synchronized
up
on
openshift
and
things
work.
You
know
the
way
that
they
should
and
it's
instantly
refreshed
right.
That's
pretty
cool
and
openshift
provides
those
sets
of
features
to
really
not
get.
In
my
way,
and
that's
the
reason
why
I
look
like
openshift.
D
Okay,
so
I'm
safe,
you
didn't
take
my
answer.
I
have
seen
other
questions
coming
in.
I
think
so.
This
started
out
as
open
shift
is
enterprise
kubernetes
cool
glad
we
got
that
messaging
across.
So
what
does
openshift
had
on
top
of
it?
When
4.0
came
out
the
first
time
I
saw
the
installer
and
the
idea
of
hey
here's
some
credentials
to
a
cloud.
Here's
the
installer
go
and
all
of
a
sudden,
I
I
had
a
full
openshift
install
six.
I'm
sorry,
three
masters,
three
worker
nodes,
everything
configured
everything
ran.
D
You
know
I
I
tend
to
and
my
talks
and
stuff
reference
at
the
2
a.m,
text
message
and
that
kind
of
stuff
that
is
very
cliche.
It's
certainly
not
anything
I
came
up
with,
but
the
idea
of
oh
service
is
down.
We
need
to
get
an
ops
person
in
here.
Coffee
them
up,
so
they
understand.
What's
going
on,
have
them
fix
it.
D
Auto
healing
or
machine
healing
in
general,
has
always
been
something
particularly
interesting
to
me
and
seeing
it
in
action
on
top
of
everything,
coupe
gets
and
just
being
like
yeah
go
scale
yourself
out.
If
you
need
to
have
more
capacity
and
just
having
it
all
magically
happen,
is
really
really
cool
so,
above
and
beyond
what
kubernetes
provides?
That's
one
of
the
cooler
features,
in
my
opinion,
in
open
shift.
C
And
that
auto
scaling
ability
is
literally
a
right.
Click
on
the
developer
console
now
create
a
horizontal
pod,
auto
scaler,
and
I
set
what
my
you
know
limits
just
like
what
you
said
in
a
form
I
don't
deal
with
the
ammo.
I
don't
deal
with
any
of
that
complexity.
I
right
clicked
from
the
web.
I
can
right
if
we,
if
we
want
to
do
that,
we
can
do
that,
but
it
is
very
simple.
G
Of
the
extra
what
jay
pointed
a
little
bit
about
multi-tenancy
earlier
and
one
of
the
other
kind
of
aspects
of
openshift
that
I
really
appreciate
is
yeah
that
the
auto
scaler
is
potentially
right.
Click.
One
button,
you
know
very
easy
to
set
up,
but
each
of
these
control
interfaces
can
be
exposed
to
the
appropriate
members
of
your
team
or
hidden.
G
If
it's
not
your
scope
of
responsibility,
right
and-
and
I
think
for
other
folks,
kubernetes
is
kind
of
like
a
generic
operating
system
or
or
middleware
for
stacking
a
lot
of
other
things
on.
But
it's
not
really
intended
to
be
a
full
complete,
like
distributed
platform
with
multiple
users
and
openshift,
really
kind
of
hits
that
that
full
vision
of
a
multi-user
distributed
platform
that
you
can
run.
G
If
you
need
to
also
for
developers
almost
all
of
that
can
be
hidden,
and
so
you
have
the
production
grade,
control
interface
into
getting
the
right
quality
of
feedback.
So
you
don't
have
any
surprises
when
you
go
to
promote
your
code,
but
without
the
added
complication,
if
it's
outside
of
your
scope
of
responsibility.
So,
there's
a
lot
of
opportunity
for
people
to
get
involved
in
just
the
right
layer
from
a
security
and
administrative
control
perspective,
and
that's
not
always
very
clearly
outlined
upstream
in
in
the
rest
of
the
kubernetes
space.
G
That's
a
lot
of
diy
work
to
secure
your
own
clusters,
and
so
a
lot
of
folks
are
like
well,
everyone
gets
their
own
dev
cluster.
That's
our
security
model
is
one
cluster
per
user
and
that
that's
one
way
of
kind
of
firewalling
off
my
development
code.
But
when
my
development
code
moves
up
a
stage
and
my
team
member's
code
moves
up
a
stage
and
all
of
a
sudden,
we
have
a
mixed
code
environment
in
production.
G
If
you
haven't
been
planning
that,
through
from
a
security
perspective
from
the
start,
you're
going
to
experience
some
pain
somewhere
right
and
I'd
like
to
have
everything
properly
tied
up
and
configured
right
during
my
development
phase
whenever
possible.
So
that's
a
huge
benefit
from
the
openshift
perspective,
to
have
everything
buttoned
up
and
ready
to
go.
D
B
D
One
of
the
things
that
josh
and
I
were
asking
in
chat-
was-
was
curl
running
inside
a
container
or
is
curl
running
on
the
open
shift
node
now
the
answer
was
container.
So
I
will
answer
that.
I
promise,
but
what?
If
it
was
running
on
the
node?
And
this
ties
into
my
previous
point
about
the
machine
scaling
if
you're,
let's
just
say,
if
you're
new
to
openshift
4
coming
from
a
three
background,
but
even
if
you're
not
this
is
going
to
sound
a
little
bizarre
red
openshift
runs
on
top
of
red
hat.
D
On
top
of
red
hat
core
os,
it
is
an
immutable
operating
system
geared
specifically
toward
hosting
containers.
Immutable
is
the
important
part
there
we're
trying
to
to
be
perfectly
blunt,
get
away
from
people
going
into
the
node
with
ssh
and
then
changing
a
whole
bunch
of
stuff.
So
all
of
that
configuration
is
handled
through
openshift
and
kubernetes
apis
and
custom
resources
that
let
us
use
those
apis.
D
I
mentioned
that
in
the
context
of
the
machine
stuff
that
I
talked
about
earlier,
because
that's
how
we
allow
that
to
happen.
The
fact
that
we
have
the
configuration
sitting
inside
of
openshift,
so
it
knows
how
to
provision
a
new
instance
and
then
the
fact
that
we
have
this
immutable
operating
system,
so
we're
not
worried
about
a
whole
bunch
of
churn
or
variability
in
those,
and
it's
easy
for
us
to
manage
the
nodes.
D
If
they're,
that
I
hate
using
this
at
this
analogy,
but
if
they're
that
much
cattle
instead
of
pets
there
yeah,
we
can
just
toss
them
away
and
grab
a
new
one.
Now
that
whole
diatribe
must
mention
that,
if
your
curl
update
was
on
the
nodes,
it
would
come
down
to
a
new
release
of
centos
that
would
be
pushed
out
with
this
new
fix
we've
got
to.
B
D
D
The
new
version
of
core
os
that
comes
out
that
we
replace
in
the
worker
node
it
wouldn't
be
something
where
you're
running
dnf
update
on
the
node.
Now,
that's
not
what
you
asked
me
you
asked
for
inside
of
a
container
because,
okay,
sorry,
I
have
vulnerabilities
and
I
want
to
fix
them
without
updating
the
openshift
version.
So.
G
D
We're
more
and
more
we're
dialing
in
this
question
too,
if
I
have
particular
container
vulnerabilities
and
that's
one
of
the
huge
benefits
of
using
containers
is
that
you
can
just
update
those
specific
workloads
instead
of
updating
the
entire
cluster.
That's
running
it,
but
again,
containers
are
themselves
immutable,
so
you're,
not
sure
you
technically
have
the
option
of
going
into
the
container
and
deploying
the
new
fix.
The
convention,
though,
is
again
back
to
the
pets
and
cattle
thing.
Your
containers
are
typically
meant
to
be
lightweight
and
volatile.
D
So
you
patch
this,
this
version
of
curl,
you
push
out
a
new
image,
and
then
you
replace
your
containers
with
that
new
image.
Now
that
is
a
spectacularly
dumbed
down
view
of
all
of
the
different
variables
that
go
into
that,
but
at
the
end
of
the
day,
that's
essentially
what
happens
now?
How
do
you
do
that
with
zero
downtime?
D
C
D
Traffic
over
the
new
ones
and
then
update
the
other
five,
so
we
don't
get
any
kind
of
downtime
and
this
all
the
way
pays
back
to.
If
he's
still
here
the
question
earlier
about:
what's
openshift
used
for
that's
another,
one
of
those
benefits
is
that
you
can
manage
this.
This
uptime
with
really
high
reliability,
because
container
changes
and
container
removals
and
everything
like
that
don't
affect
each
other,
don't
affect
the
underlying
platform.
D
C
There's
the
container
security
operator
that
gives
you
some
of
the
details
of
hey.
That
coral
version
is
there's
a
there's,
a
problem,
and
you
could
see
that
issue
in
the
openshift
console
with
the
container
security
operator
that
gets
that
data
from
a
bunch
of
different
sources.
Quake
could
be
or
coin
claire
could
be
one.
It
gets
from
a
different,
a
bunch
of
different
things,
I'll
paste
that
in
the
chat,
but
you
could
automate
and
act
upon
that
using
some
kind
of
pipeline.
Just
like
jay
said
right,
so
you
could
build
this
stuff
together.
C
D
One
last
thing
I
want
to
drop
in
this
sorry
right,
I
feel
like
I
cut
you
off,
but
there
was
something
mentioned
in
chat
about
the
image
change
trigger
and
that's
a
very
solid
way
of
doing
it
that
you
define
your
deployment,
your
internal
openshift
resource.
That
says
this
is
the
application
I
want
and
you
define
it
to
keep
an
eye
on
the
image
stream
and
once
a
new
one
becomes
available,
the
next
version
it's
going
to
automatically
deploy
that
for
you
and
again
taking
that
2am
text
message
out
of
it.
D
If
we
can
automate
that
process
of
we've
detected
a
security
vulnerability
in
our
scanning
software,
we've
tied
that
into
our
pipeline
to
build
a
new
image
and
then
openshift
picks
up
and
says:
hey,
there's
a
new
image
in
the
new
image
in
this
stream.
My
deployment
is
focused
on
this
particular
stream.
Let
me
redeploy
it
ryan.
Did
I
step
on
your
answer?
Are
you
going
to
say
the
same
thing.
G
G
Anytime,
a
new
image
arrives
in
our
registry
and
that
trigger
is
going
to
file
fire
off
anytime
a
build
finishes,
so
I
can
essentially
have
a
continuous
build
tied
to
a
continuous
deploy
if
I
hook
up
web
hooks
from
from
github
or
really
any
other
source
into
this
type
of
system.
You
might
not
want
continuous
delivery
into
your
production
cluster
depending,
but
this
is
all
you
know,
configurable
per
instance,
and
and
for
development
purposes.
Yeah
I'd
probably
continuously
deliver
so
yeah
lots
of
lots
of
options
all
customizable
to
what
suits
your
purposes.
B
G
G
Great
great
yeah,
that's
that's.
Definitely
one
place
to
check
for
for
a
recent
gathering
of
experts
from
that
specific
community.
You
can
also
keep
an
eye
on
the
openshift
blog.
We
usually
make
a
decent
a
number
of
announcements
through
there
one
last
source.
I
can
give
you
for
information
like
where
to
go
for
information
on
this
topic.
I
know
diane
mueller,
who
runs
the
open
shift
commons
organization.
G
G
So
those
would
be
kind
of
your
your
two
choices,
they're,
basically
identical.
I
think
okd
is
probably
not
going
to
have
10
years
of
product
support
behind
you
know
or
ability
for
us
to
do
kind
of
back
patching
you're,
hopefully
always
riding
on
that
edge
version,
because
that's
where
the
development
is
happening
and
if,
if
you
have
an
okd
cluster,
that's
a
year
or
two
old,
it
might
might
not
be
functioning
nearly
as
well
as
an
as
an
ocp
production
cluster
that
has
red
hat,
trying
to
do
back
ports
and
pitching
for.
B
D
C
C
Yeah
yeah,
so
with
code
ready
containers,
what
that
is
is
a
way
to
run
a
single
node
virtual
machine
version
of
full
openshift,
open
shift.
You
get
all
the
operators,
all
the
things
that
are
there.
You
know
you
have
all
that
on
your
laptop
workstation.
What
have
you
it
runs
in
linux?
It
runs
on
mac,
os
x.
It
runs
in
windows,
there's
some
requirements
with
windows.
You
need
to
make
sure
that
you're
updated
to
it
needs
to
be
first
on
windows.
10.
It
needs
to
be
on
a
specific,
build
or
hire.
C
Definitely,
I
will
make
sure
that
we
link
the
the
documentation
in
the
chat
so
that
you
could
get
these
those
details
of
which
specific
one,
but
I
think
it's
been
like.
Maybe
the
past
year
update
of
windows
10.,
so
you
probably
are
updated
there.
One
of
the
things
that
you
have
to
do
is
be
able
to
have
hyper-v,
so
you
have
to
have
windows
10,
you
know
professional,
or
what
have
you?
You
need
to
run
the
so
assuming
all
those
requirements
are
met
and
things
are
okay.
C
You
then
need
to
make
sure
that
you
run
the
crc
setup
command
and
you
have
to
have
administrator
rights
on
your
machine.
So
if
you're
running
this
on
a
workstation
where
you
don't
necessarily
have
those
admin
rights
on
that
machine,
you're
going
to
have
to
you
probably
work
with
your
it
team
to
be
able
to
you
know,
get
that
maybe
put
into
the
image
or
get
some
kind
of
exception
so
that
they
could
do
that.
C
C
So
I
would
imagine
most
of
the
problems
that
I've
seen
on
windows
are
generally
going
to
be
because
of
updated
for,
for
you
know,
a
specific
version,
but,
like
I
said
that
yeah
that's
been
like
a
year
from
now
or
a
year
ago.
I
don't
think
that's
a
problem
anymore,
but
it
might
be
and
then
the
second
one
is
generally
administrator
rights.
Maybe
it
isn't
if
it
isn't,
we
definitely,
you
know,
want
to
help
you
and
make
sure
that
you
you
get
this
up
and
running.
So
please
make
sure
that
you,
let
us
know.
B
G
G
D
D
I
don't
know
I
don't
even
have
visibility
into
the
details
of
do
all
six
start
standing
up
at
the
same
time,
do
we
do
some
workers
and
some
matches
at
the
same
time?
Apologies.
I
I
think
I
kind
of
speak
for
all
of
us
when
I
say
that,
as
advocates
we've
been
much
more
on
the
developer
experience
once
it's
upside
of
things
without
ever
having
to
really
look
that
closely
at
the
installer.
So
what.
C
C
G
The
installer
is
a
quite
a
bit
more
streamlined.
We
kind
of
set
up
a
minimal
cluster
and
then
within
the
cluster,
depending
on
which
cloud
you're
on
there's
control
interfaces
that
you
can
use
within
the
admin
dashboard
to
add
more
nodes
and
then
that'll
call
out
to
aws
provision
more
nodes
set
them
all
up.
So
so
you,
you
might
be
able
to
shift
some
of
the
work
to
kind
of
post
post
install
depending
on
or
at
least
compared
to
our
older
versions
of
openshift.
B
Cool
thanks
ask
follow-up
questions
if
it
wasn't
what
you
were
actually
looking
for
in
the
meantime,
a
deal
is
concerned
about
a
different
installation
issue,
which
is
they
want
to
try
out
4.6
at
a
minimum
cost.
What
deal
was
asking
is
whether
it's
possible
to
set
up
just
a
two
node
cluster
in
aws.
I
think
there
might
be
some
other
options
for
trying
it
out
at
minimal
cost,
though.
D
My
understanding
and
I
may
set
myself
up
to
be
told,
you're
wrong
on
this,
but
there's
no
way
to
configure
that
at
install,
but
you
can
scale
it
down
after
the
fact
you
know
actually,
as
I
say,
that
I
might
actually
be
lying
already,
because
pds
by
default
will
generate
us
smaller
nodes,
where
they've
got
set
up
with
a
single
master,
single
worker,
or
at
least
three
masters.
One
worker
is
that
all
for
I
know
I
can
say
with
confidence
that
originally
out
the
gate,
with
the
4.1
release.
D
C
We
call
it
compact
clusters.
I
have
a
bare
metal
version
of
that
running.
It's
three
master,
slash
workers.
Basically,
when
you
create
open
shift,
look
in
the
documentation
for
compact
clusters,
that's
been
a
tech
preview
feature
in
openshift
4
series
for
a
while.
I
it
might
be
ga
now.
If
not
it's
going
to
be
ga
really
soon,
I'm
not
specifically
sure,
but
it
it's
available.
It
runs.
But
that
is
one
way
if
you're
just
trying
to
try
openshift
4.6
to
see
the
new
features.
C
Learn.Openshift.Com
is
a
nice
website
that
you
could
get
a
one
hour,
ephemeral
cluster.
You
could
refresh
the
browser
and
keep
resetting
that
and
try
it
out.
4.6
is
not
in
that
yet,
but
we're
working
really
hard.
It
should
be
there
next
week-
hopefully
maybe
thanksgiving,
maybe
the
week
after
we'll
see
how
that
works
out.
But
that's
one
way,
hopefully.
G
Okay,
so
this
may
be
specific
to
vsphere,
but
generally
there's
there's
controls
within
the
dashboard,
where
you
just
pick
which
number
you
want
and
it'll
start
trying
to
roll
to
the
next
version,
at
least
for
point
releases.
So
within
the
4.5
release
you
should
keep
being
able
to
roll
forward.
At
some
point,
there
may
be
a
point
where
4.5
dot,
something
some
minor
release
may
include
information
about
how
to
roll
forward
to
4.6.
B
B
What
4.6
features
that
were
not
part
of
the
press
releases,
like
you,
know,
sort
of
minor
features
or
offbeat
features
that
came
out
in
4.6.
Do
you
personally
really
like,
like
they're,
super
useful
to
you,
you
think
they're
really
interesting
or
really
creative
solutions
to
something.
You
know
that
weren't,
like
the
big
banner
features,
but
are
nevertheless
going
to
be
super
useful.
G
G
I
don't
know
if
it's
the
largest
issue
from
4
6,
probably
not,
but
there's
something
brian
mentioned
earlier-
the
the
quick
start,
functionality
being
able
that
for
developers,
I
think
or
for
onboarding
a
team
being
able
to
put
in
specific
information
about
next
steps
in
order
to
you
know,
standardize
your
your
onboarding
experience
if
they
want.
You
want
folks
to
use
a
certain
database
or
you
know
certain
way
of
interfacing
with
kubernetes.
G
You
can
make
that
exceedingly
clear
and
have
that
available
right
in
the
experience
rather
than
out
in
a
separate
set
of
documentation,
that's
getting
out
of
date-
and
you
know
so
that's
nice
to
have
it
built
right
into
the
platform.
I
know
we
have
a
blog
post
about
that
specific
feature
for
folks
that
want
more
information,
I'll
paste
it
into
chat.
F
E
C
B
F
Yeah
well,
could
I
pick
something
from
odo
sure
awesome,
okay,
so
odo
2.0
is
is
absolutely
crazy,
phenomenal
because
we
finally
have
support
for
kubernetes
now,
instead
of
just
openshift,
and
actually
it's
crazy,
I
had
a
presentation
fell
on
me
live,
so
I
had
a
great
experience
from
that.
But
there's
this
thing
called
dev
files
which
you're
starting
to
see
in
odo,
but
also
in
crc,
which
is
super
interesting
kind
of
like
a
portable
development
environment.
You
can
essentially
use
anywhere
you
want
on
whatever
cluster
you
want,
kubernetes
or
openshift.
F
C
For
me,
I
probably
would
say
that
the
biggest
thing
is
the
general
usability
updates.
Like
I
said
I
like
to
focus
on
things
that
make
my
life
easier
and
the
general
usability
updates
in
the
topology
view
and
the
developer
console
are
really
a
big
deal
to
me.
If
I'm
working
on
something
I
you
know
before
you
couldn't
change
which
columns
you
see
in
an
administrator
view
right.
If
I
want
to
see
something
specific,
I
might
have
to
go
drill
in
and
look
at
the
pod
and
look
at
what's
actually
going
on
with
openshift
4.6.
C
We
now
have
the
ability
to
go
and
customize
the
columns
that
are
there
and
it
keeps
a
cookie.
So
I
could
see
you
know
if
I
go
back
to
it
later.
It
didn't
forget
where
I
was
right.
So
if
I
go
drill
into
something
I
could
go
back,
it'll
it'll
still
be
there.
So
that
was
a
really
big
deal
for
me.
The
other
thing
is
really
around
all
the
services
that
are
that.
Are
there
right,
so
serverless
eventing.
That's
a
really
big
thing
that
we
talked
about,
but
really
that's
one
of
my.
C
You
know
the
biggest
things
that
I
think
is
really
important.
It's
basically
the
way
that
we
could
integrate
containers
and
things
with
other
sets
of
software
right,
maybe
slack
messages
or
you
know,
or
maybe
a
cron
task.
If
we
want
to
do
something
simple,
we
could
pass
messages
using
cloud
events.
We
talked
about
it
at
kubecon.
I
heard
that
at
the
keynote
cloud
events
are
awesome
and
serverless
eventing
allows
us
to
consume
some
of
those
things
so
that
what
I'm
working
on
could
be
able
to
react
to
stuff.
C
I
came
from
mobile
app
development
before
you
know,
working
on
kubernetes
and
cloud
to
device
messaging
is
one
of
the
biggest
things
that
I
think
is
ultra
interesting,
because
it
allows
you
to
connect
your
app
on
users
devices
to
a
message
that
came
in
right.
We
could
do
those
types
of
things
now
with
serverless
eventing
and
it's
generally
available
because
it's
you
know
it's
not
tech
preview.
It's
there
right,
it's
stable
we're
supporting
and
that's
a
really
big
aspect.
D
C
So
like
in
the
in
so
there's
a
couple
of
things
so
in
the
administrator
view,
if
I
go
and
look
at
say
like
the
list
of
pods
or
a
list
of
really
anything,
I
think
that
list
view
in
the
administrator
thing
I
could
go
and
modify
the
columns
that
I
see
so
that,
like
I
could
drill
into
the
details
that
I
want
to
see
for
everything.
I
have
a
nice
good
overview,
but
there's
also
application
monitoring
that
I
get
by
you
know
default
right.
C
I
could
go
and
also
see
that
you
know
the
cpu
usage
for
this
particular
set
of
deployments,
or
this
particular
pod
you
know
has
been
ramping
up
or
there's.
I
could
have
visibility
into
what's
going
on
and
that's
all
available
now
in
the
developer
console
as
of
openshift
4.6,
so
it
makes
my
life
easier
to
monitor
and
keep
track
of
what's
going
on
right,
but
I
also
have
the
ability
to
to
drill
into
things
a
lot
easier.
So
I
don't
have
to
you
know
waste
cycles
to
see
the
details
of
each
specific
thing.
D
D
Right
cool,
it's
I
don't
I
I
don't
look
at
our
docs
regularly,
so
I
don't
know
how
well
is
this
covered?
But
if
you're
not
aware
on
the
left
side
menu,
the
bottom
section
of
them
is
customizable.
So
as
you
go
through
different
pages,
you
can
click.
Add
this
to
my
favorites
and
it's
pretty
populated
with
a
few
and
it's
very
easy
to
overlook
and
think.
Oh
they
added
some
new
stuff
to
that
part
of
the
menu,
but
that's
customizable.
D
C
G
I
just
posted
a
link
yeah
that
that's
another
really
kind
of
like
we
haven't
talked
about
that
as
much,
but
you
can
theoretically
yeah
you
don't
like
the
openshift
logo.
Put
your
own
put
your
corporate
logo.
There
you
know,
dress
it
up.
You
can
kind
of
get
openshift
as
kind
of
a
white
label
experience
and
re-offer
it
to
your
engineering
team
and
have
it
look,
however,
your
corporate
overlords
expect
you
to
have
it
look.
G
G
G
You
know
not
all
of
them
are
built
around
those
type
of
concepts,
so
the
operator
hub
within
openshift,
not
just
it,
doesn't
just
allow
you
to
have
a
nice
interface
for
administrators
to
enable
selected
features,
but
also
the
users
of
those
features
can
interact
with
a
specialized
control
interface,
that's
kind
of
offered
by
the
the
author
of
the
operator.
So
so,
if
there's
something
that
is,
I
don't
know
you
don't
have
to
deal
with
yaml
you
can.