►
From YouTube: The Level Up Hour (E64) | Prepare for a RHEL 9 future
Description
Red Hat Enterprise Linux users are living in exciting times right now since RHEL 9 is almost here, and we need to look at everything you need to get started right now. Systems engineer Jim Nauer from Case Western Reserve University joins us as we explore RHEL improvements, benefits and experiences and insights gained during the beta program. Join us as we dive into what's new with RHEL 9 and demo how you can use system roles to build a standard operating environment. We'll also go over some hints and tips that you should be familiar with, lessons learned from upgrading from RHEL 8 to RHEL 9, how to set things up and configure them, common errors and how to avoid them, as well as many other topics and answer any questions you may have.
Want to learn more?
https://www.redhat.com/en/about/press-releases/red-hat-defines-new-epicenter-innovation-red-hat-enterprise-linux-9?sc_cid=7013a000002qG1GAAU
Try Linux: https://www.redhat.com/en/technologies/linux-platforms/enterprise-linux?sc_cid=7013a000002qG1GAAU
Try Red Hat OpenShift: https://red.ht/3uYcuhU
Follow us:
Randy Russell https://red.ht/rrr-linkedin
Scott McBrien https://twitter.com/stabby_mc
Jaafar Chraibi https://twitter.com/ChraibiJc
OpenShift https://twitter.com/openshift
Subscribe to Red Hat's YouTube channel: https://www.youtube.com/redhat/?sub_confirmation=1
What is the Level Up Hour?
Every other Wednesday at 9am ET we stream live demos on 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
Level Up Hour on GitHub https://github.com/level-up-hour/
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/3uXIbaQ
OpenShift on github https://github.com/openshift/
#RedHat #OpenShift #RHEL #tluh
A
C
Good
morning,
good
afternoon,
good
evening,
wherever
you
might
be
welcome
to
the
level
up
hour
where
we
usually
talk
about
containers,
kubernetes
and
open
shift,
however,
and
please
do
like
subscribe
and
share
if
those
are
topics
that
you
are
interested
in,
but
today
today
we
actually
have
a
very
special
episode,
I'm
very
excited
about
it.
I'm
joined
by
my
co-host,
mr
scott
mcbrian,
the
legendary
stabby
mick
and
we'll
introduce
our
guest
in
a
moment,
but
scott.
C
C
Years,
yeah
yeah
20
years,
and
so
I
I
reached
deep
into
the
the
wardrobe
vault
to
get
a
shirt
from
from
yesteryear
that
I
think,
is
probably
about
that
vintage.
So
you
know
a
look
backwards
there
at
20
years
of
what
you
know:
red
hat
enterprise
linux
has
contributed
to
the
world,
but
today
is
actually
interesting
in
another
respect,
and
why
is
that.
A
Because
today
should
be
the
general
availability
announcement
of
red
hat
enterprise:
linux
9..
We
started
uploading
bits
to
the
customer
portal
and
content
distribution
networks
yesterday.
So
I
see,
for
example,
luna
g
gernberg
has
said
that
they
were
installing
real
nine
already
because
the
content
was
available
in
some
places
yesterday.
A
C
C
A
B
B
Well,
so,
as
a
customer,
I
love
the
shorter
release.
Cadence,
you
know
five
five
years.
It
was
too
long.
You
know
to
be
really
really
honest
and
in
particular,
because
I'm
in
academia,
I've
got
a
kind
of
schizophrenic
use
case
where
I
am
responsible
for
servers,
running
a
billion
and
a
half
dollar
a
year
business
on
one
side
and
on
the
other
side
I've
got
researchers
who
want
bleeding
edge.
B
So
we
absolutely
want
the
enterprise
in
enterprise
linux.
We,
you
know,
we
we
depend
on
it
to
run
our
business,
but
at
the
same
time
we
really
do
need
an
up-to-date
user
land
and
update
kernel
and
c
library
so
that
the
latest
and
greatest
software
can
build
against
it.
Because
that's
what
our
researchers
want
to
run,
they
you
know
they
do
also.
B
Sometimes
you
know,
buy
a
machine
and
and
then
never
upgrade
it
for
like
15
years,
and
you
know
you
can't
take
it
away
from
them
until
you
pry
it
out
of
their
cold
dead
hands
or
until
the
magic
black
smoke
comes
out
of
it
right,
but
when
they,
when
they're
first
building
stuff,
it's
got
to
be
brand
new
and
that
has
a
real
challenge
in
the
enterprise
linux
world.
So
well.
C
You
know
it's
really
interesting
to
look
at
the
history
of
that,
as
both
of
you
will
recall
is
that
if
I,
if
we
think
about
the
predecessor
to
red
hat
enterprise
linux,
we
had
red
hat
linux,
which
was
on
a
major
release
cadence
of
a
year
and
a
half
approximately
and
customers
said.
Could
you
please
stop
and
partners
were
like?
No?
C
But,
on
the
other
hand,
when
we
think
about
being
able
to
innovate,
move
forward,
use
new
functionality
and
things
like
that
that
we're
also
providing
in
you
know
an
enterprise
ready
version
of
that
on
that
sort
of
time
frame
that
people
need.
So
I
think
this
is
a
very
exciting
change,
so
scott,
why
don't
we
actually
switch
things
up
a
little
bit
today
and
go
straight
to
the
demo.
I
mean
seriously.
What
could
possibly
go
wrong.
A
What
could
possibly
go
wrong?
Oh
so,
the
demo
that
we're
going
to
do
today
is
on
rail
system
rolls
and
the
reason
I
chose
this.
One
is
because
I
know
that
jim
has
been
working
on
building
his
standard
operating
environment
for
route
nine,
and
I
think
that's
one
of
the
things
that
we're
probably
going
to
touch
on
during
our
discussion
later
and
one
of
the
tools
that
we
started:
shipping
with
rel
nine,
I'm
sorry
roll
up,
seven
continuing
to
roll
eight
and
we'll
even
more
continue
in
row.
A
A
A
Right
so
I'm
just
going
to
use
yum
or
dnf
since
they're,
really
the
same
to
install
these
two
packages
and
again
for
r9
that
would
be
ansible
dash
core
is
the
name
of
the
package.
You
want
and
run
system
rules
and
what
we're
getting
from
this
is
a
whole
selection
of
ansible
roles
that
allow
us
to
execute
certain
changes
across
our
population,
so
just
like
normal
ansible
can
be
distributed
for
the
purposes
of
our
demo.
A
A
The
t-log
system
role
so
t-log
is
the
session
recording
that
we
ship
with
route
8
and
9,
and
it
allows
me
to
record
people's
terminal
sessions
so
that
I
can
play
back
that
session
later
and
see
kind
of
what
happened
during
their
their
shell
session,
all
right
so
just
to
show
that
things
are
working.
If
I
take
a
look
at
just
one
of
the
values,
that's
in
the
list
for
these
kernel
settings.
A
A
and
you
know
what
I'll
go
ahead
and
show
you
the
other
one
to
3
ratio
right.
So
this
one
says
that
when
my
dirty
pages
are
30
of
memory,
I
should
like
wake
up
and
do
things
and
we're
going
to
change
that
to
40
percent
of
memory
and
then
there's
a
couple.
Others
here
as
well
like
semaphore
values
all
right.
So
the
whole
goal
here
is
to
create
a
playbook
that
you
can
execute
across
your
environment
to
set
up
kind
of
your
standard
operating
environment.
So
if
I
do
an
ansible
playbook.
A
Oh
yeah
so
notice
that
for
this
one
for
system
roles
I
actually
didn't
have
the
system,
roll
rpms
installed
or
sorry,
not
system
session,
recording
so
take
care
of
installing
the
software.
For
me,
in
addition
to
making
the
changes
to
the
configuration
to
enable
it
all
right.
So
here
we
are
we're
done
and
remember
we
looked
at
this
last
time.
Swappiness
was
just
a
30,
but
in
my
playbook
I
had
asked
it
to
be
sub
20.
A
A
And
again,
if
you
added
in
more
hosts,
it
would
then
apply
this
change
across
all
the
hosts
in
the
host
list
and
then,
if
any
of
the
hosts
had
gotten
out
of
whack,
because
somebody
had
like
manually
adjusted
something
on
them
running
this
playbook
again
would
reset
them
all
to
your
standard
operating
environment
settings
all
right.
So
it
was
20.
A
Right
right
here,
the
session
recording
was
installed
and
enabled
and
activated-
and
now
it
just
happens
on
all
the
boxes
that
I
applied
this
playbook
to.
We
can
also
add
in
more
stuff.
So
if
we
wanted
to,
we
could
do
things
like
add
a
crony
time
configuration
to
the
box
by
adding
a
new
system
role
and
then
adding
a
new
section
to
our
playbook
to
indicate
what
parameters
we
wanted
there.
A
So
you
can
apply
postfix
changes
throughout
your
organization,
there's
one
for
vpn
configuration,
so
you
can
actually
do
vpn
connection
configuration
throughout
your
population.
So
there's
a
bunch
and
we're
going
to
continue
to
increase
that
the
list
of
things
that
you
can
apply
a
crash
population
using
system
rules.
B
So-
and
I
just
saw
there's
a
an
interesting
question-
came
up
in
the
chat
about
using
ansible
for
config
files
under
etsy,
and
that's
that's
been
my
favorite
thing
to
automate
for
several
years
now.
Even
before
I
was
using
ansible,
we
were
a
puppet
shop
and
I
just
want
to
point
out
one
of
the
new
features
that
landed
within
the
last
week
and
both
were
all
8.6
and
9.0.
B
If
you
notice,
over
the
years
red
hat,
has
been
adding
or
changing
configuration
files
to
use
an
include
directive
and
a
you
know,
filename.d
directory.
So
you
know
we've
kind
of
been
doing
this
for
years
with
the
chrom.d
directory
and
and
some
other
stuff,
the
sshd
config
just
landed.
Sshd
config
can
now
be
concatenated
out
of
blocks
in
a
d
directory.
You've
been
able
to
do
that
for
years
with
sudoers,
for
example,
sudoers.d.
Instead
of
trying
to
massage
a
single
large
file,
you
can
drop
well,
then,
potentially.
B
C
A
In
november
we
announced
the
beta
and
every
subscriber,
even
those
with
a
free
developer
subscription
are
eligible
to
access
the
beta
now.
So
anyone
who
has
a
red
hat
subscription
can
access
the
beta
if
they
desire,
and
then
we
also
did
what
we
call
a
high
touch
beta
program.
So
some
folks,
like
the
esteem
jim
nauer
here,
were
nominated
by
their
by
internal
red
hatters
jim's,
a
part
of
a
program
called
accelerators
yeah.
A
Yeah
so
we
selected
some
people
to
have
a
more
curated
beta
experience
where
we
like
meet
with
them
periodically
through
town
halls,
to
give
updates
about
what's
going
on
in
different
aspects
of
the
product
and
that's
a
little
bit
more
more
exclusive.
I
guess
I
don't
know
how
to
describe
it
in
a
nice
way.
A
C
B
B
So
if
you
were
a
customer
and
you
weren't
in
the
high
touch
beta,
you
could
get
a
little
sneak
preview
kind
of,
but
you
got
no
bug
fixes
for
six
months
and
no
security
fixes
no
nothing.
It
was
just
kind
of
okay,
here's
a
hypothetical
peek
at
maybe
what
it
might
be
like,
but
you
can't
really
do
anything
because
none
of
the
bugs
are
gonna
get
fixed
for
six
months,
whereas
now
the
whole
world
has
been
getting
every
four
weeks
has
been
getting
a
drop
of
new.
B
You
know
new
rpm
content
for
rel
nine,
so
you
know
whether
you're
a
third-party
software
vendor
or
you
know
any
customer
anywhere
in
the
world.
You've
had
the
same
access
to
software
that
I
have
for
the
last
six
months,
and
I
I
really
like
that,
as
a
change
you
know
means
it
means
even
internally
in
my
organization.
I
don't
have
to
worry
quite
as
much
about
nda,
stuff
and
say
here.
This
is
the
same
software.
Everybody
else
in
the
world
has
access
to
go
ahead
and
install
it
in
a
test.
C
A
C
This
approach
is
just
making
so
many
more
eyes,
and
it's
also
being
cyclical
in
the
sense
that
you
get
to
see
the
fix
to
maybe
see
through
the
use
case
that
might
have
hit
a
brick
wall,
otherwise
with
something
that
was
encountered
in
the
beta.
So
I
think
it's
a.
I
think
it's
a
great
change,
scott,
any
you
know
additional
thoughts
on
that
or.
A
A
B
I'm
probably
not
going
to
jump
on
9-1
super
fast,
just
because
I've
already
got
my
hands
full
of
stuff
to
do
with
9-0
and
86.,
and
there's
there's
there's
a
lot
of
there's.
There's
a
lot
of
pieces
still
still
that
need
to
fall
into
place.
So
I
think
scott
mentioned
earlier.
I'm
working
on
it
on
on
building
my
standard
operating
environment
via
ansible,
of
course,
a
little
little
longer
more
complex,
but
I
love
the
example.
You
picked
scott
because
that
looks
some
of
those
kernel
tuning
settings.
B
Look
like
you
copied
and
pasted
them
right
out
of
the
oracle,
install
guide
so
very
familiar
with
those
settings,
but
we
also
do
things
like
you
know.
We
we
have
a
you
know
an
ldap
and
kerberos
authentication
setup
to
go
to
our
enterprise
authentication
system.
We
set
that
all
up
through
you
know
three
ansible
playbooks,
all
of
our
other
local
customization,
plus
the
tuning
that
we
need
forever
for
the
applications.
A
B
To
manage,
or
you
know,
to
keep
our
versions
consistent
between
test
and
production
environments.
We
also
are.
We
also
use
red
hat.
You
know
said
we
use
ldap
for
authentication.
The
ldap
server
is
red
hat
directory
server.
You
know
the
the
the
paid
product
version
of
389,
which
again
our
business,
is
completely
dependent
on
so
and
those
are
both.
B
B
I
was
very
happy
that
there
were
no
show
stoppers
this
time.
For
me
at
least,
it
was
all
just
little
annoying
paper
cut
kind
of
things.
You
know
glitches
and
anaconda.
I
found
an
infinite
loop
in
anaconda
if
you,
if
you
set
up,
set
it
up
to
install
from
an
nfs
source
and
you
stupidly
screw
up
and
point
it
to
an
empty
directory
instead
of
the
one
with
the
actual
iso
files
in
it,
anaconda
will
fall
into
an
infinite
loop
instead
of
saying
file
on
found.
B
Well,
it
keeps
looking
right
yeah,
it's
it's
we'll
try
until
nfs
gives
up,
which
is
never
yeah.
So
little
buglets
like
that
yeah,
so
I've
got
you
know
a
couple
of
been
resolved.
One
of
them.
I
was
having
problems
downloading
packages
because
of
a
wonky
hypervisor
turns
out.
You
know
that
was
that
was
actually
testing
on
my
on
my
my
m1
based
mac,
the
which
is
not
supported
by
red
hat,
but
it
does
work
at
least
for
real
nine.
You
heard.
C
C
B
Yeah
there's
so
yeah,
and
we
had
we
had
talked
about
this
before
so
my
personal
preference
is
first
of
all,
even
for
upgrades
in
general,
my
preference
has
always
been
lift
and
shift
ins
instead
of
upgrading
place
and
and
the
betas
are
no
different.
I
actually
ran
into
problems
during
the
beta
program
just
doing
yum
updates
from
one
version
to
the
next,
with
a
bare
minimal
install.
B
It
was
not
it's
not
an
issue,
but
I
had
been
playing
around
with
the
os
build
composer
packages
and
with
those
installed
I
was
running
into
problems
where
yum
update
from
like
beta
3
to
beta
4
would
fail
if
you
had
se
linux
turned
on
and
enforcing
which
we
all
should
as
right.
Now
it's
been
encouraging
for
many
years.
I
think
in
my
organization
we've
got
to
turn
down
on,
maybe
six
machines
a
little
to
do
work
there,
but
I
you
know
sorry.
B
I
was
running
the
test
config
with
I
seen
things
enforcing
and
it
was
failing
upgrades
because
of
that
and
the
situation
is
weird.
If
you
had,
if
you
install
beta
4
fresh
works
perfectly,
you
know,
yum
install
os,
build
composer
works.
Fine,
it's
just
the
there
was
something
wonky
with
the
se
linux
settings
between
going
from
one
beta
release
to
the
next
beta
release,
where
they
didn't
have
everything
exactly
right
and
it
just
didn't
work.
B
Unless
you
do
things
like
disable
se,
linux
and
then
you've
gotta,
you
know,
go
back
and
you
know
turn
it
back
on
and
restore
con
and
all
kinds
of
other
fun
junk,
or
you
know
just
reinstall,
which
you
know
it's
cases
like
that
red
hat.
Doesn't
you
know
they?
They
give
us
a
lot
of
support
on
the
ga
releases.
They
support
all
kinds
of
crazy
scenarios
that
they
don't
during
the
beta.
I
think
another
one
people
griped
about
during
the
beta
was
secure
boot.
B
B
A
C
You
know,
I
think,
it's
incumbent
on
us
to
make
sure
that
people
are
participating
in
the
beta
program.
Understand
that
that
it
is,
you
know
it
is
different
in
kind
and
and
therefore
you
have
to
kind
of
expect
that
it's
it's
going
to
be
a
rebuild
for
anything
that
you're
using
there,
but
that
doesn't
mean
it's
not
worth
doing.
It
just
means
you
have
to
go
into
it
with
that
expectation.
At
the
outset,
right.
B
B
B
A
We
have
one
yeah,
so
we
published
beta
ubi
images
for
the
first
time
with
real
line,
and
that
was
interesting,
but
also
unique
in
that,
like
the
signing
keys
that
we
used
for
them
were
the
beta
ones.
So
you
had
to
do
a
little
bit
of
work
on
your
container
hosts
in
order
to
not
have
it
throw
up
about
the
content,
not
matching
your
known
signatures
as
a
side.
Note,
the
ubi
9
ga
versions
are
all
available
in
the
container
catalog
today.
A
So
so
you
can
go
out
and
find
them
and
not
have
to
worry
about
updating
to
the
red
hat
beta
key
for
for
signatures
and
whatnot.
But
that
was
interesting
because
we
had
never
done
that
before
we
had
always
well.
Ubi
was
introduced
after
relate
was
released
by
a
couple
of
months,
but
now
we've
gotten
really
good
at
building
and
dropping
them
pretty
much
the
same
day
as
not
only
majors
but
also
dot
releases.
C
Well
and
I'll
just
you
know,
do
a
shameless
plug
here
that
you
actually
did
a
a
really
good
episode.
Actually,
we
did
an
episode,
I
believe,
with
scott
mccarty
on
ubi
and
I
think
you've
subsequently
done
some
additional
episodes,
and
so,
if
you're
interested
in
ubi
and
that's
not
something
you're
familiar
with
well
take
a
look
through
some
of
our
episodes,
and
you
know
find
that
find
that
episode,
and
I
don't
know
I'm.
C
B
C
B
B
B
C
We
have
we
have
customers
who
have
who
have
needs
for
very
extended
life
cycles,
because
very
often
it
might
involve
that
they're.
You
know
if
we
think
about
something
like
a
railway
system,
for
example,
they
work
on
time
frames
that
are
very,
very
different
from
a
lot
of
other
sectors
because
you're
buying
trains.
You
want
to
use
that
train
for
a
long
time
and
you
know
consequently,
a
lot
of
it.
C
Infrastructure
is
also
a
very
long
time
horizon,
so
yeah
2038
seems
like
forever
away,
but
in
a
way
for
for
certain
sectors,
it's
really
not
that
far
away
at
all,
or
even
it,
to
look
at
your
particular
case,
where
you've
got
the
cutting
edge
researchers,
you
know
and
wanting
the
bleeding
edge
it
still
seems
like.
You
also
have
some
use
cases
where
you
know
16
years.
B
Is
just
not
that
far
off
yeah
as
they
it
can,
it
can
get
it
can
get
a
little
ugly.
Sometimes
as
far
as
those
upgrades
and-
and
you
know,
it's
weird
thinking
that
ten,
you
know,
the
10-year
support
life
cycle
of
rel
is
long
but
yeah
you
mentioned,
you
know
certain
industries,
a
single
decade
doesn't
cut
the
doesn't
cut
the
mustard
and
you
know
they
they
think
on
multi-decade
timelines,
so
software
that
will
last
is
kind
of
important
and
and
the
fact
that
the
file
system
changed.
B
You
know
that
came
up
in
our
beta
testing
one.
You
know
one
of
the
things
I've
been
doing
aside
from
building
our
standard
operating
environment
is.
We
also
have
a
qualification
checklist
that
we
go
through.
As
you
know,
does
this
work?
Does
this
work?
Does
this
work?
Does
this
work
and
one
of
them
is:
does
our
backup
software
work?
B
B
So
we,
you
know,
we
have
a
lot
of
virtual
machines
and
we've
shifted
in
recent
years
to
mostly
using
vm
image-based
backups.
We
still
have
some
databases
in
particular.
You
want
to
have
an
old-school
agent
on
the
machine,
so
it
can
do
the
right
thing
with
backup
mode
and
get
the
right
files
and
all
that
jazz,
but
by
and
large
we're
using
image,
backup
and
and
restore
the
catch
there
is
to
we.
So
we
tested
against
the
rel
nine
beta.
The
image
image.
Backup,
of
course
works.
B
Fine
vmware
doesn't
care
what
the
bits
look
like
on
the
disk.
It's
just
an
image
back
it
up,
restore
it
done,
but
the
file
level
restored
doesn't
work
because
the
backup
software
depends
on
a
proxy
vm
to
grab
the
restored
image
mounted
and
let
you
browse
the
files
and
pick
the
one
that
you
want
to
restore
and,
of
course,
their
their
proxy
image
is
not
running.
Rel.
Nine!
Ga,
because
you
know
it's
been
what
40.
A
B
C
A
C
Right
and
we
demon
doctor
mentioned
having
a
ralph
3
server
hobbling
around
for
years
the
nightmare.
Well,
I
mean
there
does
come
to
be
the
point
where
you
know,
there's
enough
differences
that
people
might
not
remember
how
to
actually
get
it
to
do
what
you
want,
not
to
mention
the
absence
of
any
sort
of
updates
upgrades
anything
you
know,
that's
so
something
to
think
about
real
three.
B
Later,
if
you
did
a
command
line
like
this
long,
when
I'm
doing
my
kickstarter
installs,
that
sets
it
up
right
kills
all
the
stuff.
I
don't
want,
including
you
know,
netf
names,
bios,
dev
names,
yeah,
and
it's
been
a
long
time
since
you
know
I
just
because
we
only
have
a
handful
of
bare
metal
machines.
It's
been
a
while
since,
but
I
haven't,
you
know,
since
you
know
mike
jones
brought
it
up,
I'm
remembering,
oh
god,
yeah
those.
B
If
you've
got
a
machine
with
like
eight
nicks
in
it
or
eight
nick
ports,
you
gotta
it
configures
the
first
part
and
you
guys
that
they're
twiddling
your
thumbs.
While
you
wait
for
a
dhcp,
timeout
and
seven
of
them,
which
they're
you
know,
they're,
not
even
cabled
up.
Why
is
that
somebody
skipping
all
these?
A
Yeah
and
one
of
the
other
gotchas
for
nine
is
we've
obsoleted
the
network,
scripts
directory
and
files.
So
if
you
look
in
etsy
scripts,
it
is
empty.
Instead,
that's
all
managed
through
network
manager.
Now
you
can
make
changes
to
the
file
configuration,
but
you
have
to
do
it
through
the
new
location
for
network
manager
to
pick
it
up
so
that
that
was
a
for
people
who
are
used
to
like
operating
at
the
file
level
and
in
their
postscripts
they're
doing
seds
to
like
change
things
yeah.
You
probably
don't
want
to
do
that
anymore.
C
B
A
Is
the
default
for
eight
wow?
We
shipped
something
called
networking
that
was
like
just
scripts,
so
you
could
disable
network
manager
and
use
networking,
but
you
don't
you
know
you
didn't
want
to
do
that
innate.
It
was
basically
like
people
who
were
really
really
fighting.
The
change
would
do
that
to
themselves,
I'm
spite
and
in
nine,
like.
B
Don't
take
you're
you're,
making
me
think.
I
need
to
log
into
our
our
hpc
cluster
later
today
and
take
a
peek
because
we,
we
are
transitioning
that
in
our
our
high
performance,
compute
cluster
into
you,
know
firmware
7
or
l8,
and
I
I
I
am
quite
positive.
They
had
network
manager
not
even
installed
on
the
raw
seven
systems
and
I'm
not
and
now
that
you
bring
it
up,
I'm
not
sure
how
they're
handling
eight,
because
they
do
like
a
lot
of
control,
because
these
are
machines
that
have
complex
network
topologies.
B
A
Right
into
six
network
manager
was
not
the
best
in
row:
seven,
it
was
pretty
good,
but
because
of
the
experience
in
six,
we
like
didn't
make
it
the
default
and
in
eight
it
is
the
default
and
works
pretty
much
interchangeably
so
like
it
would
work
with
the
ifcfg
scripts
in
eight.
So
you
likely
made
the
change
and
never
even
noticed
the
change.
It's
not
until
nine.
That.
B
B
It's
been
pretty
good
in
the
rally
for
for
my
stuff,
the
the
high
performance
compute
world,
like
I
said
they
have
a
whole
separate
provisioning
mechanism
and
build
scripts
and
whatnot
that
are
somewhat
built
off
of
kickstart
again
and
their
their
automation
system
was
violently
allergic
to
network
manager,
at
least
in
the
real
seven
days.
I
don't
know
if
the
relay
version
has
completely
gone
in,
you
know,
gone
got
what
the
program
or
not
I'll,
find
out.
C
So
you
know
hearing
about
the
change
to
network
manager,
in
particular
in
the
and
the
end
of
of
you
know
the
the
network
scripts
directory
that
has
been
around
for
a
very
very
long
time
puts
me
in
the
mood
to
do
another
shameless
plug.
I
don't
do
too
many
of
them,
but
I'm
going
to
occasionally
do
do
one,
and
in
this
case
I'm
going
to
say
that,
with
the
release
of
red
hat
enterprise,
linux
9
coming
right
on.
A
C
It
is
official
we
can
actually
talk
about
it
now
and
actually
the
the
very
evil
toaster
pointed
out
that
just
got
their
or
hcsa
for
relate.
Well,
you
know
what
here's
the
good
news
is,
even
though
we
mentioned
this.
The
thing
with
you
know
with
network
manager
and
some
other
things
that
are
probably
a
bit
different,
that
real
eight
knowledge
is
going
to
apply
almost
entirely
to
what
you
do
in
rel
nine,
you
know
jim,
do
you
have
some
thoughts
on
on
sort
of
the
tran?
C
B
B
About
the
tests,
but
there's
a
lot
of
technologies
that
are
in
those
courses
in
courses
that
come
up
on
the
test
and
it's
a
lot
of
the
newer
features
in
the
os
that
maybe
you
don't
use
in
production
in
in
your
environment,
but
having
been
through
a
couple
of
classes
like
that,
it
definitely
leaves
you
a
better,
rounded
assisted
means
to
know
both
the
old
way
and
the
new
way
or
all
three
ways.
In
some
cases,
because
yeah
it's
been
20
years,
things
have
changed
twice
three
times
absolutely.
C
A
A
Yeah,
that's
my
go-to.
Now
I
mean
so
e-links.
Usually
I
would
use
it
to
like
grab
a
page
and
dump
it
somewhere.
So
I
could
grab
some
data
out
of
it
or
whatnot
or
just
like
grab
a
file,
and
w
get
does
that
now
and
then
I
was
just
thinking
about
like
looking
at
a
web
page
in
text
in
my
browser,
and
it
was
giving
me
like
flashbacks.
A
C
C
B
On
I
still,
I
still
install
and
occasionally
use
links,
the
other
one,
the
lynx
version,
mostly
when
I'm
mad
at
web
developers
and
want
to
show
them
the
error
of
their
ways
as
far
as
accessibility,
because
because
you
know
that's
the
classic
trick
that
we've
been
using
for
decades
and
I'm
not
in
the
web
development
world.
But
you
know
to
make
sure
our
page
is-
is
accessible
to
screen.
Readers.
C
B
And
I
just
I
just
did
that
I
switched
screens
and
and
did
did
a
you
know,
yum
list
e-links
to
see
if
it
was
in
the
e-bell
repository
on
my
on
my
on
my
machine,
but
I
want
to
point
out.
This
is
another
big
difference
between
relate
and
roll
nine
and
when
rel
eight
came
out,
you
know-
and
I
know
in
some
environments-
epel
is
like
radioactive
and
they
can't
touch
it
if
you're
a
bank
or
a
three-letter
agency
yeah.
B
C
A
A
C
Well,
that
actually
takes
us
to
a
couple
of
things
that
we
can
kind
of
wrap
up
with
here,
and
that
is
is
first
of
all,
we
touched
upon
it,
but
you
know
how
does
one
prepare
for
the
rel
9.1
dot
release?
What's
the
best
way
to
get
ready
for
that,
because
there
are
some
capabilities
that
are
are
sought
after
and
interesting
in
9.1
that
didn't
make
it
into
the
into
to
nine.
B
B
C
B
B
Yeah
was
specifically
not
tested
as
part
of
the
beta
because
they
want
to
do
it
just
for
you
know
the
the
86
to
90,
not
for
eight
five
to
nine
beta,
because
it's
so
I'm
I'm
guessing,
because
it's
a
very
meta-data-intensive
piece
of
software
they
didn't
want
to
have
to
constantly
be
redoing.
The
metadata.
B
B
So-
and
I
did
test
it
during
the
during
the
relay
beta
and
and
found
all
kinds
of
interesting
issues
and
eventually
got
it
working.
So
life
was
good.
I
don't
you
know,
like
I
said
earlier
in
the
podcast,
I'm
very
much
into
the
lift
and
shift
approach.
So
we
don't
do
a
whole
lot
of
in-place
upgrades,
but
every
now
and
then
you
got
a
really
nasty
use
case
where,
for
whatever
reason,
application
issues,
it's
just
better
to
do
it
in
place.
C
B
The
case
right
for
whatever
reason
that
doesn't
work
for
you
there's
another
option,
and,
and
so
and
so
you
know
like
I
said
I'd
like
to
see
if
redcat
can
find
a
way
to
engineer
it,
so
that
it
is
practical
to
beta
test
in
in
future
betas.
That
would
that
would
be.
That
would
be
an
awesome
that
would
be
my
my
rfe
for
for
all
10..
I
don't
know
if
I
can
pull
it
off.
B
But
it
it'd
be
it'd,
be,
I
think,
it'd
be
a
good
use
case
and
there
are
other
customers
who
use
it
a
lot.
Apparently
you
know
we've
seen
this
discussion
in
open
forums
and
in
webinars
you
know
big
companies
with
tens
of
thousands
of
machines
that
they,
just
it's
not
practical
for
one
reason
or
another
to
do,
lift
and
shift
on
that
many
systems
they
want
to
want
to
in
place
option.
So
it's
good
it's
there,
even
though
I
personally
don't
use
it
a
lot.
B
A
B
Yeah
well
right,
because
there
were
you
know
and
back
in
the
8
00
days
when
I
for
the
8-0
beta,
when
I
was
testing
leap,
there
were
issues
and
that
had
not
that
were
not
really
related
to
the
specific
release.
The
eight
owners
of
it,
it
was
related
to
you
know,
leap
had
some
limitations
that
just
made
it
not
work
in
certain
under
certain
circumstances.
C
B
You
know,
like
I
said
earlier,
boring
in
the
best
possible
way
it
worked
really
well
for
me,
I'm
really
glad
to
see
you
know
the
increased
openness
from
red
hat,
giving
you
know
all
of
their
customers
the
same
access
that
I've
had
to.
You
know
to
the
updated
software
and
I'm
looking
forward
to
doing
this
dance
again
in
three
years.