►
Description
What's TLUH & Quick Container Apps
Come join us as we kickoff TLUH! Wait, what's TLUH? Well, you have to come by to find out :). We will also demonstrate using a bunch of system tools within containers so you don't have to install everything locally.
Learn more at https://red.ht/leveluphour
A
Hello,
everybody
good
morning,
good
afternoon,
good
evening,
wherever
you're
hailing
from
welcome
to
openshift.tv,
I
am
chris
short
principal
technical
marketing
manager
on
the
openshift
team
at
red
hats.
I
am
joined
today
for
a
very
special
new
program
that
we're
starting
up
here
on
openshift
tv
called
the
level
up
hour,
and
this
is
actually
led
by
my
coworker
and
co-teammate
or
co-teammate
teammate
langdon
white
langdon.
Please
introduce
yourself
for
everybody.
A
B
B
As
I've
talked
about
in
the
past,
I've
worked
a
little
bit
as
a
sysadmin,
and
the
environment
I
worked
in
was,
as
I
think,
the
technical
term
is
janky.
So
the
organization
was
in
the
middle
of
doing
a
mass
migration
and
you
know
always
take
long
longer
than
everyone
expects,
so
it
was
to
keep
the
lights
on
on
the
existing,
both
bare
metal
and
virtualized
environment.
B
While
we
build
out
a
new
one
and
that
environment
had
by
some
previous
admin
had
nagios
deployed,
but
nothing
else,
and
so
I
wanted
to
maintain
those
machines
in
sort
of
an
automatic
way,
and
so
I
was
investigating
puppet
at
the
time.
This
was
a
bunch
of
years
ago
and
was
like
I'm
not
installing
a
puppet
agent
on
those
machines.
They
frighten
me
and
anything
installed,
may
make
them
never
come
back
again
right.
C
B
I
wrote
a
whole
mess
of
nagios
checks
that
actually
did
config
management.
I
I'm
very
very
sad
that
I
that
I
mia
the
source
code
somewhere.
I
probably
could
convince
them
to
open
source
it,
but
it
didn't
even
cross
my
mind
that
it
might
be
useful
to
anyone
else,
but
that's
kind
of
the
idea
for
this
talk
or
for
this
kind
of
series.
How
can
we
use
containers
to
our
advantage
as
like
an
admin,
especially
where
you
either
don't
want
to
or
can't
touch
kind
of
the
setup?
B
C
B
Program
has
some
support
for
taking
classes
and
of
basically
being
able
to
take
some
of
the
tests,
and
you
know
hopefully
more
stuff
to
come
so
that
you
can
kind
of,
as
it's
called
right
level
up
right
is.
How
do
you
add
containers
to
your
toolkit
as
a
as
a
rel
admin
and
both
take
advantage
of
that
as
a
as
an
admin
and
just
make
your
life
easier,
but
then,
on
top
of
that,
actually
make
you
more.
B
A
A
Joined
today
by
a
few
of
our
friends
narendev-
and
it's
just
excuse
me-
assistance
master-
welcome,
welcome
everyone
joining
again.
So
yeah
like
this
is
an
interesting
like
concept
and
show
for
me
because,
like
I
started
off
as
a
real
admin
right
like
back
in
the
day
before
it
was
rel,
it
was
red
hat.
B
B
A
A
Just
to
give
you
an
idea
of
where
I
learned
yeah
my
rail
chops,
so
yeah,
I
got
an
rh
csa
certification
back
in
like
2011
2012,
I
think,
and
then
what
was
it
two
years
ago,
I
got
my
ansible
certification,
so
this
is
like
this
is
like
helping
me
go
from
that
that
you
know
20
something
30
something
year
old,
rhcsa
up
to
where
I
am
now
right,
like
working
with
containers
on
a
daily
basis,
kind
of
thing,
so
yep
the
level
up
hour.
That's
what
we're
talking
about
so
you
know
I,
I!
A
B
Yeah
and
my
my
vast
experience
with
red
hat
enterprise,
linux
and
red
hat,
linux
was,
as
I
spent
a
lot
of
time,
in
consulting
both
freelance
and
working
for
companies,
and
I
used
to
use
red
hat
as
a
way
to
get
my
clients
off.
My
back
by
you
know,
developing
an
application
or
whatever
and
then
deploying
it
on
rel
and
saying:
hey,
call
red
hat.
B
If
you
have
challenges
with
the
operating
system,
because
they
were
basically
the
only
ones
who
are
really
offering
real
like
material
support-
and
I
knew
the
vast
majority
of
questions
would
then
go
to
their
paid
support
rather
than
calling
me,
as
you
know,
as
you
do
as
a
consultant,
the
part
of
the
reason
for
the
appeal
of
consulting,
is
you
never
want
to
work
on
the
same
project
for
very
long,
so
I
used
to
use
it
a
ton.
As
a
result,
I
actually
switched
over
to
linux
entirely.
B
I
don't
know
in
like
2005,
maybe
wow
my
like
exclusive
operating
system,
because
I
just
actually
the
reason
I
did
it
is
because
I
was
like
I'm
deploying
to
all
these
linux
machines
and
I
really
need
to
be
much
more
comfortable
working
with
them
directly.
So
I
just
switched
all
my
laptops
and
you
know
kind
of
never
looked
back
and
it's.
B
A
There's
no
questions
other
than
you
know,
tluh
the
level
up
hour
right,
that's
the
abbreviation!
It's
a
it's!
A
bot
command
that
mods
and
admins
can
run.
So
if
you
want
to
learn
more
there's
the
link
to
hit
up.
If
you
want
to
you,
know,
listen
to
us
and
read
over
that
page.
While
we
go
that'd
be
great.
B
Yeah,
oh,
and
also
in
the
future,
for
the
show
I
just
kind
of
saw
the
the
guest
comment
or
whatever
we
do
want
to
have
kind
of
guests
on
the
show.
Yes
using
you
know
who
who
might
be
relevant
to
this
particular
topic,
one
I
already
have
hopefully
lined
up
soon,
we'll
be
talking
about
toolbox
which
we'll
get
to
in
a
later
episode.
But
let's
jump
just.
B
B
A
A
B
A
B
And
so,
as
a
result,
there's
a
lot
of
stuff
that
you
might
find
useful,
not
here,
and
let
me
just
see
if
I
can
find
my
notes
window.
Oh
one
thing
I
did
want
to
mention
and
maybe
I'll
try
to
modify
the
slide
towards
the
end
is.
I
did
create
a
github
repo
called
the
level
up
hour
or
something
brilliant
like
that,
where
I'm
trying
to
keep
the
show
notes.
So
if
you
want
to.
B
A
A
B
Yeah
so
I
started
some
show,
notes
there
and
yeah.
That's
basically
there.
So
one
of
the
first
things
that
one
of
the
common
ones
that
I
use
kind
of
this
technique
for
is
nmap
nmap
is
at
least
in
my
experience,
almost
never
installed
on
any
linux
distro
by
default.
However,
is
super
handy
when
under
like
very
unusual
conditions
right
so
right,
you
really
need
it
when
you
need
it,
but
you
don't
need
it
all
that
often
so.
The
last
thing
you
want
to
do
is
like
change
your
golden.
B
You
know
deployment
to
include
nmap,
so
instead
I
use
it
via
a
container.
So
I'm
gonna
here,
because
I
kind
of
you
know,
try
to
prep
a
little
bit,
but
so
what
we
do
is
we
do
podman.
So
I
use
podman
pretty
exclusively,
you
know.
Docker
will
basically
it's
command
interchangeable,
so
you
can
use
docker
here.
I
like
podman,
better
particularly
for
the
build
side,
because
I
think
it's
a
lot
faster
and
for
most
rel
admins,
as
well
as
any
other
kind
of
admin.
B
They
don't
like
demons
running
in
the
background
and
so
having
the
docker
damon
running.
In
the
background,
all
the
time
people
dislike
that
so
podman
doesn't
have
a
daemon
like
that.
The
other
thing
that
podman
can
do
that
docker
generally,
at
least
last
I
looked
last,
I
knew
you
can't
run
anything
as
non-root.
B
B
C
B
I
use
the
rm
I
is
interactive.
T
is
tty
both
of
those
just
make
things
better
and
then
I'm
gonna
run
so
there's
this
okay.
So
here's
where
you
get
into
a
little
bit
of
diceyness
as
an
admin.
In
my
opinion,
okay,
using
random
containers
from
the
internet
is
dangerous
right.
Yes,
so
you
know
for
this
call
or
for
this
show,
I'm
going
to
use
a
couple
of
random
internet
containers.
B
If
it
was
my
own
data
center,
I
would
build
these
containers
myself
and
then
use
them
somewhere.
You
know
from
then
on
either
by
going
and
looking
at
the
person's
github
and
and
forking
it
and
being
able
to
take
advantage
of
their
work
sure.
But,
generally
speaking,
I
don't
recommend
kind
of
running
random
internet
connect
containers.
B
The
risk
is
limited
here
right
because
we
are
running
it's
not
running
as
root.
It's
running.
You
know
in
this
case
it's
running
in
a
trash
virtual
machine,
but
I
am
giving
it
enough
privilege
to
do
something
like
nmap,
so
so,
if
you're
familiar
with
nmap,
this
is
basically
from
the
example.
Docs
is
just
you
know
you
can
scan.
B
You
know
the
the
nmap.org
site
so
that
you
know
you
can
kind
of
test
it
out
and
I
also
cheated
because
I
previously
downloaded
or
theoretically
previously
downloaded
the
nmac
container.
So
it'll
take
a
lot
less
time,
but
what
I
think
is
super
cool
is
that
now
all
you
have
to
do
once
you've
done
this
once
is
wait
for
nmap,
which
is
of
course
not
a
quick
and
dirty
little
application.
It.
A
Yes,
so
we're
talking
about
trusts
and
images.
So
when
we
talk
about
trusted
images
at
red
hat,
what
we
mean
is
images
that
you
have
put
together
and
have
put
in
a
trusted
repository
right
so,
like
you
have
taken
a
base
container,
for
example,
narendev
and
and
and
have
added
things
to
it,
to
make
it
your
unique
container
or
image
for
your
applications
or
your
specific
needs.
So
you
would
take
something
like
ubi
and
then
add
in
you
know,
like
nginx
or
whatever
components
you
need
to
this.
A
B
C
B
It's
a
relatively
limited
set
trusted.
You
know
you
can
probably
trust
docker
to
to
do
the
the
elements
in
their
registry
that
are
theirs.
Those,
I
think,
are
all
indicated
by
the.
If
you
notice
this
is
like
the
way
this
is
named
is
this
is
like
a
namespace
here
oops,
which,
in
it's
in
the
case
of
the
docker
registry,
is
also
the
username
and
then
here's
the
specific
container
we're
talking
about.
I
believe
the
docker
one,
the
trusted
stuff
is
all
underscore
and
that's
the
trusted
registry.
B
If
you
trust
another
organization
within
the
docker
registry,
you
might
also
be
able
to
pull
stuff
from
there
and
be
okay.
Trusting
it
so
bitnami,
for
example,
is
an
example
of
of
some
people
who
have
content
like
that.
The
other
thing
you
can
do
is
you
go
to
somebody,
you
trust
and
use
their
registry.
So,
for
example-
and
I
was
surprised
that
not
a
lot
of
people
know
about
this,
but
if
you
look
in
the
container
registry
list,
I
always
do
that
backwards.
B
You
can
actually
see
here
are
the
ones
that
rel
ships
with
okay,
so
as
common
with
most
com
files.
Right,
if
you
you
know,
the
default
is
commented
out.
So
here
are
two
registries
by
red
hat,
which
I
trust
red
hat.
So
therefore
I
trust
their
registries.
These
are
not
open
registries
and
so,
as
a
result
they
are
trusted,
or
at
least
I
trust
them.
And
then,
if
you
actually
go
to
registry.red
hat
dot
io,
you
can
find
a
particular
container
there
and
you'll
see
that
it
actually
has
a
rating.
B
Running
security
scans
on
them,
so
so
you
have
to
trust.
There's
a
bunch
of
layers
of
trust
in
this
problem.
It's
very
much
like
trusting
in
rpm,
though,
in
that
you
need
to
trust
the
organization.
That's
delivering
you,
this
binary
blob
in
the
case
of
an
rpm,
it's
the
rpm
itself
in
the
case
of
container
it's
the
oci
image,
so
you
got
to
trust
that
the
that
what
is
in
that
binary
blob
is
what
the
registry
says.
It
is
right,
then,
on
top
of
that,
because
containers
are
it's
kind
of
like
downloading
a
vm.
B
So
if
you
ever
use
vagrant,
for
example,
you
download
boxes
or
you
bitnami
used
to
have
a
big
business
in
downloading
vms.
Amis
are
a
great
example,
so
you
have
to
trust
the
content
inside
the
the
container
itself
yeah.
So
we
have
well-known
ways
of
doing
that
because
it's
all
linux,
so
therefore
they
have
trusted
package
repositories
behind
them,
so
you
take
fedora
or
you
take
ubi,
which
is
the
relever.
You
know
rel
one,
and
you
can
trust
that
the
rpms
that
you're
installing
using
their
install
installation
tools
are
the
correct
ones.
B
A
And
I
would
like
to
point
out
that
the
the
open
source
quay.io,
like
the
quay.io,
that
you
can
use
for
free,
will
do
security
scanning.
You
know
it's
with
clear.
You
know
some
light
security
scanning
of
your
containers
for
you
like.
Even
then,
though,
you
still
have
to
trust
the
source
right.
Like
I'm.
Looking
at
one
of
my
containers
here
and
you
know-
yes,
it
passed
all
its
security
scans
17
days
ago,
but
do
you
really
trust?
A
A
B
A
B
A
C
B
But
what
you
can
do
is
you
can
cheat
you
can
well,
you
can
do
the
obvious
cheat,
which
is
give
the
ip
address
of
the
machine
you're
actually
on,
but
another
way
you
could
do
it
in
theory.
I
don't
know
if
this
will
actually
work,
but
I
think
it
should,
but
you
can
actually
give
the
container-
and
this
is
a
security
hole
right.
So.
B
Kinda
but
yeah
so
now,
what's
interesting
is
as
you,
as
you
probably
know,
right
nmap
is
actually
a
privileged
operation.
Yes,
so
I'm
actually
getting
an
error
here,
because
I'm
trying
to
end
map
directly
from
inside
the
machine
without
using
sudo
right
so
not
not
to
make
everything
super
complicated.
But
the
point
is
for
most
use
cases
except
for
trying
to
hit
yourself.
You
know
nmap.
Actually
I
wonder
if
it'll
work
without
actually.
A
B
All
right,
so,
let's
move
on
to
the
next
thing
and
we
can
see
another
one
that
I
use
a
fair
amount,
especially
if
I
want
to
know
if
something
I'm
reading
on
the
internet
as
a
trusted
source
is
who
is
so?
What
I
did
was
I
found
kind
of
another
random
container
on
the
internet.
That
does
a
who
is,
and
so
I
can
run
a
whois
against
redhat.com
and
look
at
that
now
I
have
who
is
my
other
problem
with
who
is
is
at
least
on
fedora.
B
There's
it's
been
replaced
by
this
weird
java
thing.
That's
called
j
who
is-
and
I
don't
know
what
that
is,
and
I
find
it
very
confusing
so
yeah.
So
I
like
regular
old
who
is
so
this
is,
as
you
expect,
the
you
know
who
is
of
redhat.com,
although
I
don't
think
I've
ever
seen.
This
come
lauded.com
thing
before
I
hate.
A
B
Is
that's
just
who
we're
querying
so
again?
Random
container
on
the
internet
quality
may
vary,
but
it
does
show
you
it
does
work
right.
In
the
vast
majority
of
cases,
it'll
be
okay,
you
can
often
you
know
you
can
at
least
get
some
level
of
trust
if
you
actually
go
to
whatever
registry
is
in
has
mostly
has
some
sort
of
landing
page
for
the
particular
container.
B
C
B
A
C
B
A
B
B
A
B
Like
half
of
it
is
in
is
actually
in
podman
directly
like
podman
is
like
a
runtime,
it's
actually
it's
just
a
wrapper
around
build
a
scopio
and
like
cryo,
I
want
to
say
and
so
podman
I
don't
know
that
it
actually
does
anything
on
its
own.
So
I
think
I
have
scorpio.
I
just
can't
use
it
directly
like
I
have
the
library
or
something
right.
A
B
A
C
A
B
B
A
B
Like
that
I've
been
following
her
trying
to
mimic
her,
I
want
to
do
all
of
my
development
inside
containers
mechanism
and
it's
actually
in
some
ways
very
similar
to
the
flat
pack
model.
It's
kind
of
interesting
it's
worth
checking
out.
I
don't
know
if
I
can
share,
because
the
real
way
I
would
do
it
is
with
a.
Let
me
see
if
I
can
share
like
this
whole
window
or
this
whole
desktop.
B
B
A
B
C
A
B
So
I
say
we
abandoned
the
telnet
a
little
bit
about
a
couple
of
others.
So
what
happens
when
you
either?
Don't
trust
whatever
but
or
alternately
want
it
can't
find
a
container
for
the
thing
you
want.
So
I
made
a
couple
because
I
wanted
to
solve
this
problem
so
chris.
This
is
funny
because
I
think
we
were
just
talking
about
ss,
which
I
don't
know
how
to
use
right,
but
I
know
how
to
use
netstat.
So
I
want
to
use
oh
by
the
way.
B
B
B
Yeah,
so
it's
it's
just
kind
of
funny.
If
you
really
want
to,
you,
can
always
do
a
sim
link
across
to
it.
I
actually
one
of
the
common
ones
I
use,
because
I
often
want
to
make
multi-base
containers
so
like
I'll,
want
to
have
a
fedora
version
of
it,
but
I
might
want
to
have
a
ubi
version
of
it,
and
so
what
I'll
actually
do
is
have
like
oci
file,
dot,
fedora,
oci
file,
dot,
ubi
whatever
and
then
sim
link
to
the
correct
one
that
make
sense.
B
C
B
We
are
doing
containers
that
we
are
going
to
use
all
the
time
and
we
actually
in
a
sense,
want
upgrades
right.
We
want
the
changes
that
might
land,
because
we
want
security
fixes
and
all
that
stuff,
and
we
know
that
our
netstat
is
not
going
to
break
something
in
production.
If
you
know
the
fedora
version
upgrades
and
it
gives
us
a
new
version
in
that
stat,
for
example,
this
could
easily
be
ubi
or
anything
else
that
you
like,
but
then
the
next
line
here
is
just
how
do
we
install
netstat?
B
B
A
C
B
A
B
Case
what
I'm
implying
with
this
line
here
is
colon
latest
okay,
that
is,
that
is
literally
what
I'm
actually
writing,
even
though
I
don't
have
to,
and
so
because
of
this
use
case,
where
what
I
really
want.
Every
time
I
build,
this
container
image
is
to
get
the
latest
version
of
fedora
with
the
latest
version
in
netstat.
B
Want
to
do
is
something
like
this,
because
then
I'll
know
that
I'm
getting
a
guaranteed
version
of
whatever
linux
and
depending
on
the
linux,
you
know,
I
would
say,
95
99
of
the
time.
If
you
declare
their
major
version,
you
will
only
get
backwards,
compatible,
fixes
right,
that's
not
bug
for
bug
compatible,
which,
if
you
ever
go
and
see
any
of
my
rants
about
modularity
and
app
stream.
B
That
is
actually
something
I
would
like
to
be
able
to
do.
I
would
like
to
be
able
to
be
guaranteed
that
things
are
bug
for
bug
compatible
or
know
when
they're
bugged
for
bug
compatible,
because
I
may
be
taking
advantage
of
that
bug.
One
of
my
great
stories
about
this
is
when
windows
95,
I
think,
came
out.
B
C
B
B
B
In
short,
I
am
I
try
to
be
as
declarative
as
possible
when
I'm
building
the
container
image
or
when
I'm
describing
the
container
image
in
the
oci
file
or
the
docker
file,
and
then
I
change
it
just
so
that
it's
a
known
quantity
when
I
want
to
upgrade-
or
god
forbid,
actually
use
ti.
That's
another
option
too
right.
B
B
B
B
While
I
try
to
figure
out
what
I'm
trying
to
do
and
preserve
the
cache
so
that
build
time
is
faster
and
then
I
go
back
and
clean
up
the
oci
file
at
the
end,
once
I
know
I
have
all
my
dependencies
correct
and
all
that
jazz.
But
if
you
do
dnf
install
you
know,
net
tools
and
then
psu
tills
and
then
whatever
as
separate
lines,
you
can
take
advantage
of
the
caching
for
the
rebuild
times.
A
B
B
A
So
that's
interesting
in
the
case
where
you're
building
that
software
in
a
container,
you
would
have
to
say,
dnf
install,
which
you
know
and
then
install
the
software
itself
in
that
container.
That's
cool
right
and
you
could.
You
can
still
build
it
as
a
container,
and
you
know
you
can
add
the
dependencies
in
one
by
one
as
opposed
to
having
this.
C
B
B
Unlike
netstat,
for
example,
which
I
was
like
wait,
it's
not
called
netsat
so
now
we
just
build
this
guy,
and
so
we
just
do
podman
build
and
we
give
it
a
pretty
name
like
netstat,
and
then
I
actually
don't
know
if
it'll
work
with
dot.
So
if
you're
doing
docker
build
with
a
docker
file,
you
can
put
just
dot
and
it
will
discover
what
it
needs,
but
that
does
not
work
so
you
do
need
that.
B
The
way
this
container
works
is
it's
actually
going
to
put
this
as
the
entry
point
into
the
container,
so
you
can
pass
it
things
directly
with
flags,
so
user
bin
netstat.
The
thing
to
remember
about
this,
though,
is
that
this
doesn't
work
with
cmd
because
of
the
difference
and
like
the
difference
between
entry
point
and
cmd
is
so
subtle.
It.
B
B
So
this
will
not
work
if
right
here,
you
had
typed
cmd
so
getting
out
of
here
all
right,
so
we
built
it
and
now
we
can
run
it,
but
here's
where
it
gets
wonky,
because
if
we
just
run
it
netstat
has
just
a
default
output,
so
I
don't
actually
have
to
pass
any
flags,
but
if
I
do
this
again,
we
run
this
problem
of
wait.
That's
the
inside
of
the
container's
net
stat
that
isn't
very
useful
to
me.
So
for
this
guy
we
have
to
run
net
host.
B
B
So
these
are
the
these
are
what
we
expect
to
come
out
of
netstat.
If
it's
run
as
if
it
was
on
this
machine,
you
know,
but
then
we
can
also
pass
flags
to
it.
So
we
can
say
ln,
which
is
what
I
usually
do,
because
I
can
never
remember
the
labels
and
stuff,
and
I
just
want
port
numbers,
we
can
pass
out
their
command.
Oh,
that
didn't
come
out
pretty,
but
we
can
pipe
it
to
the
things
and
stuff
like
that.
You
know.
B
A
B
B
A
A
A
B
Maybe
there
we
go,
you
know
I
can
also
you
know
I
can
drop,
and
so
as
soon
as
I
add
something
new,
let's
say,
volume
opt
and
it'll
recognize
that
there's
a
new
line
in
there
and
I
happen
to
choose
a
super
fast
line
just
for
the
sake
of
argument,
but
if
you
notice
it
built
a
new
layer
to
to
do
that,
one.
So
hopefully
that
answers
the
question
all
right.
So
I
have
one
more
example
I
was
going
to
give,
which
is
one
of
my
favorite
tools
called.
B
A
B
B
So
the
way
we
get
it,
let
me
do
this
and
it
also
has
a
bug
so
we'll
talk
about
that
in
a
second
but
same
dealio
with
the
dnf
install
right
so
just
to
get
python
and
pip,
but
then
I
actually
do
the
pip
install
now
remember
this
is
running
as
root
right,
so,
but
it's
inside
a
container,
so
I
don't
care
if
it
does
a
global
install.
Actually,
I
would
prefer
a
global
install
because
it's
easier
to
address
my
first
guess
as
to
where
it
would
have
put.
A
A
B
A
A
B
C
B
C
B
B
C
B
A
B
Image,
it
was
just
in
that
one
instance.
You
know
one
of
the
things
I
try
to
explain
thing.
The
thing
is
is
if
you
ever
took
a
arbitrary
design
class.
You
know
the
container
image
is
the
object
right
and
the
container
you
know
that
you
see
under
ps
right
is
the
instance.
Yes,
so,
oh
boy,
that's
not
gonna
help
all
right.
So
now
we
just
got
to
build
it
again.
B
B
B
A
B
B
Yeah
right
right
so
like
before
it's
like,
I
just
need
to
remember
where
the
semicolon
do.
I
need
to
semicolon
before
the
do
or
after
the
due.
I
can
never
remember
right
like
that
kind
of
thing,
yeah,
so
so
yeah.
So
there's
how
do
I
I
always
like
to
pitch
it
just
because
I
think
it's
a
great
little
piece
of
software
and
I
don't
know
we're
almost
out
of
time.
Do
we
have
any
more
questions.
A
Hang
on
someone
is
not
realizing.
I'm
on
here
right
now,
hang
on.
No
there's,
not
really
many
more
questions
or
some
general
comments
right,
like
narendo,
is
very
happy
with
rubo
inc
for
getting
some
some
free
courses
for
spinning
up
some
some
linux
knowledge
for
him.
Narendev
says
like
php,
is
kind
of
a
foreign
thing
to
him
and
he
needs
to
learn
the
linux.
You
know
linux
file
systems
and
linux
systems
in
general,
a
little
bit
better,
so
yeah
like
that.
That
works.
B
B
As
opposed
to
right
user
bin
or
whatever
right
non-native
install
also
used
a
local,
so
yeah,
so
it
like,
if
I
thought
about
it
more,
I
probably
would
have
guessed
it
myself.
I
just
you
know
also
want
to
show
off
how
you
get
which
so
I
also
had
a
little
demo
of
mtr
we're
almost
out
of
time,
but
basically
I
was
using
mtr
kind
of
the
same
way.
B
A
B
B
So
lots
of
stuff
was
using
fedora.
So
that's
why
it
didn't
clear
that
one,
but
nothing
else
was
using
python
and
pip.
So
that's
why
it's
building
that
again,
or
at
least
that
exact
command
right.
So
if
I
had
python
pip
and
which
it
would
still
build
a
new
container
layer,
this
is
going
to
take
a
minute.
A
A
A
Yeah,
well,
you
know
I
mean
I
try
not
to
it
was
back
in
the
day
right,
like
back
in
the
day
back
in
the
90s,
when
I
was
building
and
rebuilding
and
building
and
rebuilding
linux
systems,
all
the
time
is
because
I
couldn't
clean
the
cruft
off
of
them
right.
So
I
wanted
a
clean
system
all
the
time,
so
I'd
end
up
installing
a
bunch
of
stuff
and
being
like
crap.
A
B
You
want
them
and
in
fact,
what
we're
going
to
talk
about
in
one
of
the
future
episodes
is
actually
how
to
distribute
these
containers.
So
let's
say
you
do
go
to
the
effort
of
building
your
your
whip
container
right
and
your
netstat
container
and
all
those
things
we're
going
to
talk
in
a
later
episode
about
how
you
distribute
those
inside
your
data
center
in
a
way
that
you
can
kind
of
access
them
from
anywhere.
B
A
B
B
So
somebody
commented
on
using
links
for
browser
stuff
in
the
cli.
I
will
often
do
that,
but
I
usually
use
e-links
just
because
I
think
it's
faster
and
better
these
days,
but
one
thing-
and
this
is
a
knock
a
little
bit
on
the
open
shift.
Folks,
we
gotta
we're
gonna
talk
to
them.
One
of
these
days,
e-links
doesn't
support
javascript,
at
least
on
fedora
it's
compiled
out
and
the
openshift
console
will
not
open
without
javascript.
B
So
if
I
want
to
install
openshift
on
a
particular
like
a
vm
or
something,
and
then
I
want
to
go
and
make
sure
it's
working
and
making
sure
that
it
like,
I
have
a
networking
problem
like
a
routing
problem
rather
than
a
running
problem,
I'll
use
e-links
to
go
and
load
it
whatever.
It
is
to
make
sure
I
can
actually
see
it
from
localhost
right,
even
though
it's
headless
and
then
come
to
the
other
machine
and
realize
whether
it's
a
networking
issue,
or
did
it
not
come
up
at
all.
A
B
A
Yeah,
if
you
need
something
a
little
bit
more
like
trusted,
enterprise
hardened
tested,
secure,
ubi,
universal
base.
Images
from
red
hat
will
give
you
rel
like
images,
it's
basically
rel
under
the
hood.
Essentially,
no
licensing
needed
kind
of
thing
right,
like
you
just
need
to
have
access
to
the
registry
grab
the
ubi
images
themselves
and
you
can
build
from
those
like
that's
kind
of
my
default
starting
point.
Langdon
years
might
be
different.
B
B
If,
however,
I'm
building
something
that
is
real
right,
something
that
might
actually
get
deployed
somewhere
I'll
use
ubi,
because
I
want
the
enterprise
environment
from
the
get-go.
So
one
of
the
things
right
so
I
want
whatever
apache
is
deployed
in
rel
eight.
Today,
not
what
will
be
deployed
in
rail
nine.
You
know
two
years
from
now
or
I'm
just
making
updates.
Just
you
know
for
the
sec
is
listening.
So
that's
that's
kind
of
my
choice.
A
lot
of
the
time.
I
also
use
fedora
a
lot
so
I'm
familiar
with
where
stuffy
is
so.
B
B
And
this
should
just
work,
and
I
don't
know
if
you
noticed,
but
I
actually
used
dnf,
not
yum
right,
but
it
should
still
work.
However,
I
have
to
go
get
the
image,
because
this
is
my
my
little
demo
vm.
So
I
didn't
have
it
installed,
but
yeah.
So
dnf
and
yum
are
actually
pretty
much
interchangeable
on
rail8,
so
they
they
both
work.
B
B
B
B
A
B
Yeah,
so
on
developers.redhat.com,
if
you
search
my
name,
it's
pretty
unique.
So
if
you
just
search
for
langdon,
it
usually
works
there.
I
wrote
an
article
explaining
the
different
repos
in
rel
eat
and
the
short
version
of
that
article
is
actually
let
me
see.
I
can
find
it,
I'm
looking
for
it
yeah.
No.
I
was
just
gonna
list
the
repos,
so
I
can
explain
them
yeah,
but
I'm
not
sure
if
this
will
be
too
big.
B
The
let's,
actually,
let's
not
start
with
code
ready,
let's
start
with
base
os
and
app
stream
rally,
base
os
this
guy
or
or
really
whatever
the
right
one
is.
But
you
know
what
I
mean
the
base.
One
is
where
all
the
stuff
that
makes
the
operating
system
run
lives.
Okay,
so
it's
like
the
bare
minimum
as
as
trim
as
we
could
make
it
to
give
you.
You
know
basically
the
rpms
that
are
there.
B
It
has
a
very
long
life
cycle,
generally
speaking,
very
little
variance,
etc,
etc,
and
so
the
idea
of
that
is
that
this
is
the
core
of
the
operating
system
where,
because
linux
right
so
and
this
again,
if
you
see
my
rants
about
modularity
in
app
stream
linux
distributions,
the
problem
is
both
the
fact
that
it's
a
distribution,
but
it's
also
an
advantage.
So
a
distribution
goes
lock
step,
as
it
goes
version
of
version
version,
so
it
means
that
all
the
os
stuff,
plus
all
the
applications,
all
version
together.
B
This
can
be
a
huge
advantage
in
that
you
know
everything
works
together,
except
you
know,
of
course,
when
it
doesn't,
but
the
downside
is-
and
my
typical
example
for
this
was
in
fedora
land.
For
example,
there
was
a
ruby
version,
change
that
puppet
didn't
support,
so
puppet
was
out
of
the
u.s
bang
done
so
that's
a
pain
right
so
with
rel
and
now
with
fedora
as
well.
C
A
B
That
sounds
right,
yeah
all
right,
throw
me
the
link
or
put
the
link
there
and
I'll
check
and
make
sure,
but
so
the
app
stream
is
where
the
applications
live.
The
thing
that's
cool
about
app
stream
is
because
it's
based
on
modularity
or
what
rail
calls
app
streams
is.
You
can
have
multiple
versions
of
the
same
thing
available,
so
you
can
have
multiple
versions
of
ruby,
multiple
versus
apache,
et
cetera,
et
cetera.
B
A
B
B
So
base
os
between
base
os
and
upstream
they're,
the
equivalent
of
all
of
those
repos.
Then
we
also
have
develo
repos
so
basically
stuff
that
you
need
to
build
other
stuff,
but
you
wouldn't
actually
run
in
production
right,
so
header
files,
all
that
kind
of
stuff.
That's
what's
in
code,
ready
builder,
and
so
that's
what's
in
that
repo
and
then
there's
one
more
repo
which
is
probably
relevant,
which
is
supplementary,
which
is
an
analog
to
the
real
seven
supplementary
and
supplementary,
is
basically
licensed
encumbered
stuff.
B
B
B
A
B
B
B
The
problem
is,
is
that
when
you
run
an
application,
you
have
to
declare
what
version
you
want
it
to
be
right
so
because,
obviously,
there's
no
way
for
the
operating
system.
Well,
that's
not
true,
but
in
the
in
the
tool
chain
that
we
have,
there
was
no
way
for
the
operating
system
to
know
which
version
of
the
runtime
you
wanted
to
use
for
any
given
application
right
right.
So
you
know
the
drupal
example
run,
so
you
could
have
php
five,
two
and
five
three
both
installed
at
once.
B
In
theory,
I
don't
know
if
they're
actually
there,
but
in
theory
and
five
two
could
be
used
for
drupal
and
five
three
could
be
used
for
wordpress
on
the
same
machine.
What
app
streams,
let
you
do
is
also
declare
what
version
you
want,
but
it
doesn't
support
the
multiple
versions
installed,
but
it
also
doesn't
require
you
to
declare
what
version
you
want
to
use
and
what
we
found
from
most
users
is:
that's
actually
what
they
want.
They
don't
actually
want
to
be
able
to
run
both
drupal
and
wordpress
on
the
same
machine.
C
C
B
A
A
B
B
B
A
A
B
B
B
A
B
Oh
good
call,
I
could
have
done
that.
Oh
somebody
reports
that
it
works
cool
all
right,
yep,
it
worked
cool
and
if
you
check
out
the
github
repo,
there
may
also
be
points
for
submitting
issues
or
prs.
If
you,
if
you
thought
anything
good
that
you
wanted
to
add,
there
will
be
more
internet
points,
but
those
will
be
manually
added
because
of
our
high
quality
technology
system
for
collecting
points.
A
A
Yay
points
all
right.
Thank
you
very
much
langnon.
This
was
very
educational.
I
appreciate
everybody
joining
us
today
on
the
level
up
hour.
Please
go
to
red.ht
level
up
hour
to
learn
more
and
look
forward
to
us
every
week,
every
wednesday,
this
time
slots,
langdon
and
chris
and
maybe
other
guests
as
well.
Oh.