►
Description
How Red Hat Enterprise Linux (RHEL) users and admins can benefit their organizations and improve their careers by learning how to use containers, Kubernetes, and Red Hat OpenShift.
Learn more at https://red.ht/leveluphour
A
A
B
So,
let's
see
I
will
share
this
slide,
but
first
I'm
lagging
white
and
we'll
do.
Let's
see
how
do
I
want
to
do
this
today,
rink?
How
about
just
that
share
all
right,
and
so
this
is
the
level
up
hour
in
case
you
are
lost,
and
here
is
where
we
talk
about.
Why
containers
are
super
useful
for
kind
of
everybody?
You
don't
have
to
be
just
a
developer.
You
don't
have
to
be
just
a
programmer.
B
You
can
join
us
on
our
discord.
If
you
want
to
chat
with
us
and
you
don't
use
twitter
or
even
if
you
do
use
twitter
and
remember,
there's
always
sweet
sweet
internet
points
for
lots
of
the
activities
you
do
but
we'll
get
to
those
later
in
the
show.
So
episode,
16
eric
jacobs
and
as
eric
mentioned,
we
don't
have
show
notes
from
last
time
this
week.
Yet
basically,
I
just
ran
out
of
time.
Yeah
hard
drive
failures,
even
with
most
backups
you
know,
is,
is
still
a
nasty
recovery
activity.
B
Unless
you
have
nice
raid
arrays,
which
this
particular
machine
did
not,
so
let's
get
right
into
it,
because
that
is
the
last
slide
before
the
sweet
sweet
internet
points.
So,
let's
see
so,
we
were
first
talking
about
the
turkey
hat,
which
of
course
is
important,
and
if
anybody
has
any
questions
about
the
turkey
hat,
please
feel
free
to
put
them
in
the
chat.
A
A
You
know
for
those
who
don't
know
what
american
thanksgiving
is,
so
we
were
having
the
conversation
internally
about
how
I
was
going
to
cover
the
the
live
stream
and
somehow
turkey
hat
came
up
or
silly
hat
or
whatever,
and
I
said
well,
if
a
turkey
hat
shows
up
in
my
house,
it
will
get
worn
during
the
stream
and
so
sure
enough.
A
turkey
hat
has
shown
up
and
so
living
up
to
my
promises,
I'm
wearing
it.
B
If
you're,
if
you're
of
the
christian
bent
there's
a
saints
day
every
day
of
the
year
too,
so
let's
see
what
was
I
gonna
say.
Oh
so
two
comments
I
wanted
to
make.
One
is
when
you
mention
sides,
I
think
it's
remiss
to
not
mention
sweet
potatoes
and
marshmallows
as
one
of
the
de
facto
thanksgiving
standards.
A
B
A
A
B
A
B
The
other
thing
is
casseroles
are
always
like
your
mom's
is
the
right
one.
You
know
it's
like
you
know,
they're
they're,
very
different.
You
know
every
recipe
is
gonna,
be
slightly
different
and,
like
the
best
one
is
the
one
that
your
mom
makes.
You
know
I
used
to
say
this
a
lot
about
like
achar
and
salsa,
and
you
know,
there's
a
bunch
of
different
things
that
if
your
mother
makes
it
it's
kind
of
like
for
lack
of
a
better
term,
it's
kind
of
like
the
salsa
of
india,
except.
B
A
A
C
B
Jp
dad
says
chicken,
no,
it's
turkey
week,
turkey
week.
So
what
was
I
going
to
say?
Oh,
the
other
comment
I
was
gonna
make
about
thanksgiving
too,
is
my
daughter
has
a
very
funny
remark
that
basically,
first
of
november
through
christmas-
oh
no
he's
all
celebratory
around
christmas
like
if
you
go
to
the
stores,
there's
you
know
signs
and
everything
else
do.
A
C
A
B
B
A
B
Make
sure
you're
good
mushrooms
are
interesting,
though,
because
they're
also
very
different
in
the
us
versus
europe
in
europe.
Most
of
the
time,
if
you
pick
a
mushroom
you,
it
won't
kill
you,
whereas
in
the
u.s
most
of
the
time,
if
you
pick
a
mushroom,
it
will
kill
you.
So
it's
basically
the
different
breeds
breeds
no
varieties
species.
The
species-
that's
the
word
I
was
looking
for
are
much.
B
Sure
that
they're
clean
all
right,
so,
okay,
so
moving
on
so
you
know
part
of
I
took
advantage
of
the
fact
that
eric
was
gonna
host
the
show
and
say
hey.
Why
don't
we
talk
a
little
bit
about
stuff
you
like
to
talk
about
and
one
of
the
things
that
you
brought
up
before
the
show
was
you
know,
kind
of
the
size
difference
between
like
and
one
of
the
advantages
of
containers
right
is
basically
huge,
their
portability
right.
What's
huge
sorry,
I.
B
A
C
B
The
reason
this
was
particularly
interesting
was,
I
part
of
my
hard
drive
failure.
Was
I
lost
the
machine
that
I
used
for
this
show,
and
so
I
was
recreating
it
there's
not
a
lot
on
it,
so
it
wasn't
too
big
a
deal,
but
you
know
it's
still:
it's
a
rail
install
you
know
and
all
that
stuff.
So
I
made
it
on
this
machine
here.
This
fish
jump
mist,
and
I
created
this.
You
know
tluh
vm,
but
I
actually
want
it
over
on
this
machine,
which
is
hemdall
in
case
you're
interested.
B
No
mist
and
hemdall
are
fist.
Jump
is
the
name
of
the
freelance
company
that
I
have
had
for
many
many
years.
So,
let's
see
so
so
I
was
trying
to
so.
I
was
like
all
right,
hey
I'll,
just
go.
You
know
to
this
like
migrate
button,
which
is
normally
at
least
not
grayed
out.
I
don't
know
why.
Oh
maybe
because
it's
off
oh.
B
B
Yeah,
but
if
that
doesn't
exist
then
we
can't,
I
can't
be
running
kvm
at
all,
so
that
doesn't
seem
logical.
So
then,
what
do
I
do?
Well
then?
I
have
to
so
then
I
have
to
kind
of
go.
Oh
boy
go
over
to
you
know
one
of
the
machines
right
and
then
I
have
to
you
know
scp
it
around
or
are
sync
it
around,
but
then
what
you
were
kind
of
pointing
out
was,
you
know
how
good
is
your
storage
and
I
will
say
so
just
to
make
it
a
little
easier
read.
B
B
B
A
B
B
A
B
Well,
the
letters
are,
there
is
no
actual
yum
installed
anywhere.
I
can't
remember
if
it's
just
sim
linked
or
if
it's
actually,
if
it's
actually
the
file,
is
actually
called
yum,
but
it
yeah.
I
don't
know
I
should
do
that,
but
I'm
waiting
for
the
install.
So
the
thing
that's
and
so
part
of
the
reason
I
try
to
use
yum
the
the
three
letters
all
the
time
on
rel
to
look
for
bugs
honestly
to
see
if
there's
any
mismatches
between
how
you
use
yum
and
how
you
use
dnf.
B
A
C
B
Know
what
you
get
anymore,
if
you
install
like
yum
yutils,
because
repo
query
and
stuff
like
that
all
works
differently
now,
so
the
but
supposedly
at
least
with
core
yum.
It's
like
99.9
command,
math,
both
inputs,
but
also
outputs.
So
if
you
have
scripts
that
rely
on
the
output
of
yum,
which
is
terrible,
mind
you,
they
should
still
work
so
so
podman
is
a
rapper.
Much
like
yum
is
a
rapper
or
something
so,
let's
see
I
was
looking
at
the
where's.
The.
B
Well,
because
there's
a
there's
a
map,
so
if
you
do
podman
docker
this
guy,
if
you
install
this
one,
it
will
actually
and
then,
if
you
type
docker,
because
your
muscle
memory
tells
you
to
so
much
like
yum
and
dnf,
it
will
actually
run
podman
underneath
which
I
find
kind
of
amusing,
and
that
also
actually
has
nearly
one
for
one
mapping.
So
to
I
don't
know,
ultra
four
half,
maybe
vert
manager
yeah.
I
actually
use
cockpit
a
ton
now
to
do
virtual
management.
B
You
know
of
of
machines,
but
part
of
it
is
I'm
so
used
to
vertman
that
I
and
I
wanted
to
try
the
migration
thing
and
I
know-
or
at
least
I'm
pretty
sure
cockpit
doesn't
do
the
migration.
So
that's
why
I
was
kind
of
fooling
around
with
vert
manager,
but
mostly
I
use
cockpit
now
for
virtual
machines,
which
has
got
really
good
and
which
is,
I
think,
super
weird.
So
but
the
point
we
were
trying
to
make.
B
A
B
C
A
B
Oh
so
somebody
confirms
that
cockpit
is
on
centos
yeah,
so
oh,
but
you're
not
fond
of
it.
I
would
like
to
hear
your
criticisms
of
it
and
they
would
too
so
if
you
have
remarks
or
ideas
or
whatever
feel
free
to
file
them,
they
are
very,
very
responsive.
Oh,
this
is
cockpit.
Look
at
that
yeah.
So
what
I
wanted
to
show
is
like
yeah,
you
get
all
these
nice
little
things
and
then
you
can
kind
of
say,
network
interfaces.
B
The
only
thing
that
I
dislike
about
it
but
vert
manager
has
the
same
problem
is
that
you
can't
get
an
ip
on
anything.
That's
a
mac,
vtap
or
basically
a
a
loopback
which
this
is
something
I
still
don't
really
understand.
But
so,
if
you
you
know,
if
you
want
a
public
interface
on
this
machine
and
you
don't
go
through,
you
know
high
quality
networking
skills
which
I
do
not
have
yeah
you.
You
can't
actually
see
that
ip
from
the
machine
it's
on
because
of
mac
vtap.
A
B
Right
right,
exactly
I'm
not
even
sure
it's
high
enough
quality
called
baloney,
maybe
would
go
with
spam
sandwich
meat
yeah
exactly
so
there's
a
few.
Let
me
point
out
a
few
really
cool
things
in
cockpit,
which
are
not
necessarily
there
by
default
imagebuilder,
which
lets
you
create
blueprints
of
the
isos
or
kvm
images.
You'd
want
to
make
the
only
thing
that
I
dislike
so
far
about
it
is
that
it
there
aren't
any-
or
at
least
I
haven't
found
any
pre-built
blueprints
that
you
can
kind
of
base.
A
B
B
So
I'm
pretty
sure
the
upstream
name
of
it
was
leap
l-e-a-p,
maybe
two
ps,
but
I'm
not
sure
what
the
product
shipped
as
it
might
have
shipped
as
leap,
but
it
might
have
shipped
to
something
else,
but
it
might
even
go
six
to
eight
as
well,
but
that
the
the
project
leap
was
trying
to
cover
like
all
of
the
upgrades
world.
You
know.
So
you
know
somebody
who
I
used
to
work
with
a
lot.
Was
the
team
lead
for
that
project?
B
So
I
knew
a
little
bit
about
it,
but
not
enough
to
like
say
anything.
Intelligent
got
it,
but
so
yeah.
So
overt
now
has
a
plug-in
directly
inside
cockpit.
If
you
want
to
play
with
that,
you
can
muck
with
storage,
you
can
actually
muck
with
your
podman
containers.
Obviously
this
the
machine
I'm
pointing
at
is
probably
not
the
best
choice.
Just
because
hey.
B
A
B
B
Yeah,
so
what
was
I
saying,
a
commitment
to
have
much
more
regular
rail
releases.
You
know
in
the
approximately
three
year
time
frame
so
we're
coming
up
on
a
three
year
time
frame
since
the
release
of
rail
eight.
So
has
it
been
three
years
since
l.a
came
out,
let's
find
out
for
sure
we
can
ask
wikipedia
but
yeah.
I
think
so.
A
B
B
Way,
well,
no!
So
if
there
was
but
alphas
tend
to
come
out
like
a
year
before
the
release,
I
wouldn't
know
yeah,
I
don't
know
so
I
guess
it's
not
that
long.
No,
that
can't
be
right.
18
months
ago,
wow
time
flies
when
you're
having
pandemic
all
right.
Moving
on,
let's
see
so
what
we
wanted
to
show
was
red
hat
dot
io.
B
A
A
C
B
B
B
A
A
C
A
B
That's
the
other
thing
I
would
make
the
point
of
too
is
like
distribution
in
general
is
so
much
easier
because
there's
like
the
whole
concept
of
having
a
repository
of
images
and
all
that
stuff
is
just
built
in
to
the
concept
of
our
current
container
world.
Right
I
mean
you
know.
Containers
have
been
around
for
a
long
time,
but
you
know
we
we
do
have
to
give
props
to
docker
for
really
making
them
consumable
and
one
of
the
things
that
they
did.
A
C
B
Well
and
put
together
a
distribution
network
that
until
very
recently
yeah
I
mean
we
don't
know
how
what
the
fallout's
going
to
be
of
basically
them
having
all
of
their
containers
start
to
expire
essentially,
but
you
know
that
that's
huge
yeah,
there's
a
lot
of
rel4
and
cell
towers
right
is
is
the
way
I
hear
it.
So,
let's
see
what
was
I
gonna
say?
Oh
so
I
should
have
done.
B
B
C
B
Right,
yeah
and-
and
the
nice
thing
is
right
so
now
that
I
have
this
image
right,
I
can
just
go:
throw
that
in
a
repository
or
whatever,
and
then
jump
to
this
other
machine
and
just
pull
it
down
right
like
there's
nothing
to
it,
which
is
what
I
think
is
super
like
additionally
super
useful.
Not
only
is
it
small
or
smaller,
but
it's
also
super
easy
to
distribute.
Normally,
I
complain
about
people
getting
excited
or
too
excited
about
things
like
alpine,
because
you
know
for
me
the
difference
between
100
megs
and
50
megs.
B
Of
but
the
point
being
basically
is
like
does,
you
know,
does
50
megs
matter?
You
know
it's
like
100
megs
versus
two
gigs.
That
makes
a
difference,
but
really
what
you're
looking
for
is
minimal,
basically
attack
surface
area,
and
that,
I
think,
has
a
lot
more
to
do
with
the
distribution
and
the
content
inside
it
like
you
can
have.
I
could
have
a
25
meg
image
that
is
massively
insecure,
like
the
physical
size
of
the
image
doesn't
make
any
difference
whatsoever.
B
A
B
A
B
One
of
the
things
we
should
you
know
one
of
the
things
we
should
talk
about
right
is,
you
know
at
some
point
is
there's
what's
his
name
gross,
there's
a
really
cool
demo
by
a
security
person
named.
I
think
his
name
is
david
gross
security
and
what
he
does
is
he
has
a
demo
of
two
different
vms
inside
aws,
which
he
has
set
up
under
different
accounts,
so
they're
like
both
not
owned
by
him.
If
you
know
what
I
mean
you.
C
B
B
Basically,
I
think
it's
some
of
the
meltdown
bugs,
but
he's
also,
if
you
ever
heard
of
rowhammer.js,
oh
so
rowhammer
is
basically
a
javascript
library
or
app
or
whatever
you
might
want
to
call
it
that
can
crack
a
machine
based
on
somebody
coming
to
your
website,
and
so
he
did
that
as
another
demo,
but
yeah
he's
super
famous
for
roe
hammer,
but
then
he
like
a
year
or
two
ago.
He
did
this
really
crazy.
Video
about
video,
which
I
thought
was
funny
gross.
B
I
think
that's
what
it
was.
I
can't
remember
what
technique
he
used
so
this
the
thing
I
put
in
the
chat
is
just
his
about
row.
Hammer
and
it's
just
he's,
he's
really
interesting.
You
know
the
stuff
he's
been
doing
so
if
you
want
to
see
crazy
security
stuff
he's
a
good
one
to
to
go
check
out
all
right.
B
So
what
I
wanted
to
move
on
to
before
we
run
out
of
time
is
you've
been
recently
getting
into
ai
and
openshift,
and
I'm
curious
coming
from
primarily
an
admin
background
right
like
what
is
interesting
about
that.
For
you
like.
What
about
that
aspect?
Is
it
you
know?
Is
it
the
actual
ai
stuff
or
is
it
how
ai
interacts
with
openshift.
C
A
So
I
have
like
personal
interests
in
ai
and
machine
learning
for
side
project
stuff,
and
so
it's
like
well,
we
do
a
lot
of
that
at
red
hat.
So
I
guess
that
makes
sense
for
me
to
spend
time
there
because
I'll
have
to
learn
it
and
then
I'll
end
up
knowing
more
about
it,
and
so
it
sort
of
just
bleeds
into
you
know.
Human
beings
are
inherently
selfish,
and
so
that's
that's
really
what
it's
all
about.
A
A
C
A
Loosely
understand
a
little
bit
about
how
that
stuff
works.
So
it's
interesting
in
the
context
of
like
openshift
and
containers
make
it
pretty
easy
to
provide
access
to
gpus
to
data
scientists
so
that
they
can
do
their.
You
know,
model
training
and
fitting
and
predictions
and
other
things
I
just
haven't
done
any
of
it
directly
myself.
B
It's
really
just
fancy
fancy
statistics,
but
the
gp
like
one
of
the
one
of
the
challenges
I
think
in
this
kind
of
like
highly
distributed
world,
is
how
do
you
you
know
kind
of
leverage,
so
the
thing
about
machine
learning,
generally
speaking,
is
it
uses
much
more
like
set
theory
to
do
its
numerics
or
to
do
its
calculations
right?
A
lot
like
statistics
does,
which
is
why
gpus
are
better
at
it
right.
I
I
didn't
say.
A
That
does
things
that
look
like
humans,
so
any
any
task
that
a
human
can
do
that
is
performed
by
software
is
what
we
would
consider
to
be
some
form
of
artificial
intelligence.
Yes,
machine
learning
is
a
specific
subset
of
that,
where
you're
leveraging
fancy
statistics
to
do
to
make
predictions,
but
making
predictions
is
something
that
a
human
could
and
would
do
given
the
set
of
data
and
a
knowledge
of
statistics
right
like
so
spam
detection
right,
I
can
look
at
an
email
as
a
human
and
be
like
okay,
this
this
looks
like
junk
mail.
A
You
know
it's
spam,
it's
not
spam!
That's
a
that's
a
label
classification
task
that
a
human
can
easily
perform.
So
while
it
falls
under
the
specific
genre
called
machine
learning,
because
it
uses
those
algorithms
and
it's
a
label
classification
task
which
is
considered
machine
learning
and
blah
blah
blah,
it
still
falls
into
this,
like
bigger
umbrella
of
artificial
intelligence.
Again,
as
far
as
the
way
that
we
message
and
discuss
it
because
you're
using
software
to
do
something
that
a
human
would
would
normally
do
you're
replacing
a
human
task
with
an
algorithm.
Basically.
B
So
so
in
that
definition,
I
my
pet
peeve,
starts
to
fall
down.
I
guess
the
the
distinction
I
make
is
that
or
the
way
I
tend
to
think
about
it
is
that
ai
is
functions.
A
computer
can
do
that.
We
don't
know
how
they
do
it,
and
so,
if
you
look
at
things
like
deep
learning
or
like
neural
networks
or
other
kinds
of
like
traditional
ai.
B
A
B
A
B
B
Right
so
that
was
kind
of
my
point
is
like
the
difference.
It's
like
saying:
programming
is
ml
or
ai.
Like
you,
you
can
use
it
in
either
scenario.
It
doesn't
necessarily
indicate
that
this
is
a
machine,
learning
activity
or
deep
learning
activity
or
an
ai
activity,
or
you
know
or
anything
else,
it's
just
this.
The
kind
of
like
the
approach
you're
taking
does
that
make
sense
like
I'm
trying
to
classify
it.
That
way,
it's
not
it's
like
it's
not
yeah.
B
It's
like
it's
like
a
data
structure
like
just
because
you're
using
a
hash
table
doesn't
mean
you're
necessarily
using
a
particular
programming
language
or
you
know,
ai
or
ml,
or
anything
else.
It's
just
it's
a
it's
a
technique
and
that's
what
supervised
versus
unsupervised
learning
is
they're,
just
two
different
techniques.
There
are
other
kinds
of
techniques.
A
C
A
B
A
B
B
A
Based
on
how
we're
defining
ai
here
right,
if
you
had
a
complicated
nested
set
of
if-thens
and
case
statements,
that's
artificial
intelligence,
because
it's
in
imitating
intelligent
human
behavior
right,
even
dr
spatzo,
under
this
definition
is
artificial
intelligence
because
it
imitates
intelligent
human
behavior.
It
is
not
using
machine
learning,
because
it's
not
learning
anything.
It's
just
faking.
A
You
know,
like
you
explicitly
coded
the
behavior.
That's
what
made
it
sort
of
appear
to
be
artificially
intelligent
machine
learning.
Is
I'm
not
explicitly
coding
the
behavior,
I'm
feeding
a
data
set
into
something
and
then
the
modeling
technique
produces
a
desired
behavior
sort
of?
If
you
will,
but
it's
it's
all
sort
of
predictive
in
a
way,
and
then
you
know
again
deep
learning,
even
even
deeper
than
that
so
yeah
it's!
This
is
anyway.
This
is
the
way
that
we
we
have
been
looking
at
it
and
describing
it.
A
B
This
is
kind
of
one
of
those
like
you
know.
Those
tipping
points
or
change
points
that
you
have
in
your
life,
where
I'm
curious
how
my
life
kind
of
would
have
been
different
because
I
was
doing
neural
networks
in
college
and
then
I
went
and
worked
for
an
r
d
center
like
immediately
out
of
college,
and
then
I
bailed
out
and
went
to
software
consulting
and
I'm
curious.
B
B
A
B
B
A
A
B
Now
I
feel
like
I
should
go,
watch
it
again,
all
right
so,
but
if
you
want
to
get
more
points
for
this,
show,
go
and
fill
in
the
the
level
up
form
and
file
for
points
you
can
you
win
points
for
joining
our
discord?
You
win
points
for
doing
a
pull
request
or
doing
issues
against
the
episodes
repo,
and
all
of
that
is
explained
in
the
activities
link,
which
I
linked
a
little
bit
earlier.
B
A
B
B
A
B
A
B
C
B
B
A
B
B
A
B
I
think
it
was
9
600.
yeah,
I'm
trying
to
remember
so.
My
first
job
in
tech
really
was
actually
working.
Tech
support
for
an
isp
in
way
back
in
the
day
and
and
that's
what
I
vaguely
remember
from
that
was
that
was
one
of
the
things
we
would
have
to
check
with
on
our
customers
is
like.
Did
you
buy
a
fast
enough
modem
because
it
ain't
gonna
work?
If
it's
not
fast
enough,
so
yeah
crazy,
crazy?
Oh
wait!
No
2400!
Oh
yeah
yeah!
Wait!
No,
I'm
no!
Definitely
2400!