►
From YouTube: Community Stream #16 - Kicking it with Russ
Description
Broadcasted live on Twitch -- Watch live at https://www.twitch.tv/nrelabs
A
A
B
B
B
A
B
So
other
than
that
I
don't
know,
you
know,
I
just
I've
been
working
on
a
new
next
top
draft
for
bgp
link-local
next
autograph
for
BGP
and
Tony
P
and
I
are
working
on
some
stuff
and
i4i
Esaias
drafts
and
I've
got
to
I
gotta
ride
it.
I
gotta
update
the
draft
on
this
top
just
just
up
flood
distributed
optimum
flooding,
so
yeah
it's
a
lot
going
on.
I've
got
a
webinar
in
a
couple
of
weeks
with
Pearson
as
well,
and
I
haven't
even
done
the
slides,
yet
I
hope
nobody
from
Pearson's
lissa.
A
Actually,
I've
had
that
problem
many
many
times
and
I
I'm
past
the
point
of
being
ashamed
of
admitting
that
that
I
do
those
kind
of
things
last-minute,
because
a
there
is
a
documented
personality,
type
that
excels
and
those
in
those
situations
and
I
feel
like
I'm,
one
of
them
not
to
mention
like
if
it
ain't
broke,
don't
fix
it
so
like.
If,
if,
if
I've
been,
you
know
producing
satisfactory
content
at
the
last
minute,
why
would
I
try
anything
else?
A
B
A
B
B
A
B
A
A
It
conferences,
especially
man,
you
know
I
used
to,
especially
when
I
was
like
you
know
little
earlier
in
my
career
and,
like
you
know,
doing
all
the
doing
all
the
drinking
and
staying
up
all
night
and
all
that
kind
of
stuff
like
that
was
fun
for
a
little
bit,
but
man
even
then,
even
just
even
even
just
when
I
was
trying
to
like
make
connections
in
the
industry.
It
was
still
so
draining
and
I
felt
I
felt
like
there
was
no.
There
was
no
escaping
it,
and
so
you
could
actually
get
home
cuz.
A
Even
if
you
went
to
your
hotel
room,
it's
like
you're
still
out
yeah.
You
know
it's
not
even
that
much
of
an
escape
to
just
go
to
your
hotel
room
and
and
even
then
people
are
like
yeah
weirdo
you're,
not
staying
you're,
not
staying
up
until
2:00
a.m.
how
dare
you
not
be
part
of
me?
You
don't
I
mean
so
it's
just
yeah
I,
don't
know,
I,
don't
know
one
of
the
reasons
I
like
the
whole
meetup
craze
is
because,
like
you
know,
it's
like
three
hours
and
then
you
go
home.
B
B
A
B
A
B
A
A
We
did
a
bunch
of
shows
there,
but
that
kind
of
fizzled
out
I'm
involved,
obviously
with
like
the
tech,
village
podcast,
but
I'm
not
running
that,
and
so
you
know,
I'm
not
responsible
for
that
continuing
so
generally,
I'm,
not
very
good
at
keeping
up,
but
like
I
kind
of
want
to
try
like
experimental,
different
kinds
of
recordings,
like
I've
thought
about.
Maybe
like
strapping
a
GoPro
to
my
head.
You
know
the
kind
of
like
hooks
are
on
front.
It
shows
your
face
and
then
just
like
riding
my
bike,
you
know
and.
A
B
A
A
B
A
B
A
B
B
A
A
B
B
A
B
Yeah,
well
not
really
to
get
rid
of
it,
but
you
know
I
think
we
spent
far
too
much
time.
Just
configuring
devices
regardless
and
the
network
should
just
run
I
mean
you
should
just
rack
and
stack
stuff
and
plug
it
in,
and
this
should
just
work.
It's
crazy!
How
much
time
I
mean
for
things
like
policy
and
stuff
like
that
yeah
you
need
to
do
configuration
you
need
to
overlay,
you
need
control
or
anything
else,
but
why
really
are
we
spending
time,
configuring
stuff
all
the
time
so.
A
It
sounds
like
it
sounds,
like
your
opinion
is
like
do
we
just
have
too
many
options
like
we've
got,
we've
got
like
yeah
a
smorgasbord
of
ways
to
do
things,
but
in
reality,
especially
lately,
there's
there's
really
not
a
need
for
that.
Much
flexibility.
We
just
think
that
we
need
that
flexibility
right.
A
A
A
A
B
A
A
B
A
A
B
Really
good
on
a
datacenter
fabric,
the
fastest
I
could
get
bgp
to
converging
on
a
large
scale.
Data
center
fabric
is
about
a
minute
and
a
minute's
a
little
bit
too
long
in
large-scale
changes
and
stuff
like
that,
and
the
second
reason
is
BGP
has
become
the
can
of
the
internet,
and
we've
just
got
to
stop
throwing
stuff
in
the
trash.
Can
it's
overflowing
people?
Stop?
B
No
I
can't
remember.
When
Cisco
IOS
broke
a
million
lines
of
code,
I
would
bet
there
are
bgp
implementations
that
are
two
or
three
hundred
thousand
lines
of
code.
Today,
it's
just
crazy.
It's
like
its
own
operating
system,
it's
just
insane
and
it's
too
complex
and
it's
not
a
knock
on
bgp.
It's
just
that
we
keep
throwing
junk
at
it
and
we
just
keep
trying
to.
B
Oh,
let's
make
it
where
it'll
out
of
configure
when
you
configure,
when
you
bring
up
two
bgp
peers,
we'll
make
it
do
auto
discovery
like
is,
is
doesn't
know,
SPF
does
yeah.
That's
not
really
the
point,
that's
not
what
it
was
just
time
to
do.
Oh
well,
we'll
convert
it
into
rip
for
a
data
center
fabric.
No,
that's
not
really
what.
A
It's
not
a
knock
against
danceable
I
think
I
think
the
nature
of
these
tools.
If
you
have
a
tool
that
is
flexible
and
allows
you
to
do
all
kinds
of
different
things
with
it,
everybody
tries
to
make
it
do
everything.
I,
don't
know
what
it
is
about.
The
networking
industry
that
is
like
that,
it's
kind
of
unique.
If
you
look
at
like
the
sysadmin
industry
they're,
not
nearly
that
bad
I
know,
there's
some
corners
of
it.
But
I,
don't
know
something
about
something
about
the
networking
industry
wants
to
try
to
make
a
single
tool.
B
We
build
another
way
to
make
it.
Yes,
it's
been
a
pattern
for
a
long
time.
Yeah
so
I
mean
I,
guess
I'm,
not
gonna,
be
GP.
I,
just
think
it's
not.
We
just
need
to
stop
trying
to
make
it
do
everything
as
far
as
eius
eius
versus
OSPF.
The
main
reason
I
like
is,
is
better
for
this
particular
situation
is
just
that
it's
easier
to
work
with
its
tio
V's
and
it's
so
so
easy
to
just
throw
stuff
in
there
and
make
things
work
and
do
stuff
and
there's
some
efficiency.
Things
too.
B
B
A
A
I
was
like
honestly,
that
was
the
biggest
struggle
that
I
had
coming,
not
from
a
service
provider
background,
like
my
routing
experience,
was
with
BGP
was
just
very
like
you
know.
User
centric
like
I
know
how
to
peer
with
stuff
like
I'd,
never
built
anything
advanced,
and
so
this
idea
of
like
building
meshes
that
like
laid
on
top
of
each
other,
was
just
very
foreign
to
me.
It
was
very
confusing
because
I.
A
B
A
A
B
B
A
Really
can't
beat
it,
I
mean
just
the
the
like.
The
programmability
of
like
being
able
to
spin
things
up
in
the
cloud
generally
is
useful,
then,
but
then
of
course,
also
the
billing,
as
you
go
so
like.
If
you
don't
need
it
anymore,
you
could
shut
it
down.
If
you
combine
that
with
the
fact
that
you're
not
actually
running
in
a
VM,
that's
like
that's
a
huge
deal
and
pack
it.
You
may
be
a.
B
A
B
B
A
B
A
A
B
Saw
that
right,
Pierson
just
went
digital
first,
so
they're
gonna
only
print
when
you
ask
for
print
or
something
like
that:
oh
cool
yeah,
so
it
changed
their
entire
billing
system
and
everything
to
be
digital
first,
which
is
was
interesting
to
me
because
that's
a
major
shift
in
the
publishing
industry.
When
somebody
like
Pearson,
does
it
yeah.
A
B
A
B
Right
so
what
I
would
like
to
see,
particularly
for
like
a
data
center
fabric,
but
everywhere,
if
possible,
you
shouldn't
have
to
configure
IP
routing
like
you
should
just
work.
You
know
you
should,
if
you
need
to,
if
you
need
to
configure
a
policy
to
do
whatever
else,
that's
fine,
and
that
goes
on
top.
That
goes
as
an
automatical
thing.
Well,
I
shouldn't
be
bringing
up
a
device
and
configuring
ice
ice
on
it
right
now.
A
It
and
I
think
this
is
I
think
this
is
one
of
the
biggest
shortcomings
with
the
network.
Automation
conversation
is
that
whenever
you
hear
the
words
Network
automation,
everybody
assumes
what
you
mean
is
that
you're
automating,
the
configuration
of
the
network
and
like
that
can
be
true
and
often
is,
but
it's
definitely
not
the
the
whole
picture
like
if
you
look
at
any
reasonably
reasonably
automated,
you
know
operations,
network
operations
or
other
other.
You
know
just
generally
infrastructure
operations,
team
you'll
see
pockets
of
automation
that
tie
disciplines
together
or
tie
sort
of
technology.
A
You
know
areas
together
like
there
will
be.
You
know
a
lot
of
specially
larger
scale,
companies
they'll,
release
versions
of
their
own
infrastructure
in
like
pods
and
they'll,
and
there's
a
whole
team
dedicated
to
like
versioning.
You
know
pods
of
infrastructure.
It's
like
v3
of
the
pod
that
runs
Oracle
or
whatever
it
is
and
like.
A
Automation
within
that,
but
then
there's
also
automation
that
ties
different
like
applications
together
and
you
say
like.
Oh,
we
have
this
automation
that
says:
okay.
Whenever
a
ticket
is
open
in
ServiceNow,
then
we
touch
this
other
script,
that's
specific
to
version
2
of
this
pod
and
oh
by
the
way
it
also
reaches
into
our
wireless
system
and
our
voice
system
and
all
kinds
of
stuff,
like
that
and
I.
Think
I
think
that's
the
biggest
piece
that
people
are
missing
when
they
talk
about
network
automation.
A
They
limit
themselves
mentally
and
and
often
verbally
to
this,
like
very,
very,
very
small
scope
of
Network
automation,
equals
automated
network
configuration
changes
and
like
right,
hey,
we
shouldn't
be
making
that
many
configuration
changes
to
the
network
and
B.
There
is
so
so
so
much
more
on
the
average
network
engineers
plate
than
making
configuration,
changes
on
a
network
device
right.
The.
A
B
A
B
Should
did
you
know
that
should
just
happen
like
what?
Why
am
I
doing
that?
That's
crazy
or
interface
addresses?
Why
am
I
doing
that
stuff?
I
should
just
happen.
It
should
just
be
there.
Why
am
I
can'
configuring,
our
sis
now
again,
bgp,
because
you
have
policy
is
the
other
way
around
you
do
want
to
configure.
In
fact,
you
want
to
be
very
intentional
about
what
you're
configuring
but
dhcp.
Why
can't?
B
I
just
tell
this
network
where
there's
a
dhcp
server
and
it
just
figures
out
that
everybody
gets
an
address
out
of
that
pool
sure
why?
Why
am
I
doing
this
stuff?
That's
just
crazy
and
then,
where
that
would
allow
me
to
spend
my
time
thinking
about
things
like
policy
right,
like
QoS
other
things
in
troubleshooting.
B
Why
do
I
need
to
manually
go
out
and
figure
out
that
if
this
interface
is
having
trouble
that
I
need
to
collect
different
kinds
of
information,
why
isn't
the
network
just
do
that
yeah
or
why
is
it
there
automation
just
to
do
that
right
by
the
time
I
get
to
looking
at
it
and
go
hey
yeah,
there's
a
problem
with
this
interface.
All
that
information
should
already,
it
should
already
know
how
to
dig
into
what
additional
things
I'm
gonna
want
to
collect
figure
out.
What's
going
on
yeah.
A
And,
and
and
the
thing
that
you're
still
on
the
hook,
for
which
is
policy
is,
is
not
always
a
very
easy
thing
to
come
across
right
a
lot
of
times.
You
don't
necessarily
know
how
you
want
the
policies
to
be
constructed.
You
need
to
go
actually
go
figure
that
out
and
you
need
to
reach
into
other
systems,
often
at
the
application
layer
to
in
order
to
figure
out
exactly
how
to
construct
your
policies
and
yeah,
and
a
lot
of
those
things
can
and
should
be
done,
programmatically
and
aren't
currently
being
done.
A
That
way,
because,
again,
typically
they're
done
via
like
different
silos,
a
store
like
storage
is
a
great
example
I
used
to,
and
and
this
is
still
happening
today
in
fiber
channel
storage
networks
and
I
I'm,
sorry
I
know
Firebird
channels
like
old
hat,
but
it
is
still
dominant
in
the
storage
space,
like
the
fact
that
we
still
have
to
say,
okay,
I
have
my
server
and
I
need
to
look
up
the
storage
equivalent
of
a
MAC
address,
which
is
a
double
double
UPN
I
gotta.
Take
that
I
gotta
copy
it
into
my
switch
config
manually.
A
The
fact
that
that
still
exists
is
horrendous.
That
is
a
heinous
crime
that
we
have
that
we
are
on
the
hook
for
like
like
looking
at
this
one
system
and
this
other
system
be
like
copy
paste.
There's
you
know
it's
like
stupid.
It
should
just
be
one
of
those
things,
and
this
is
one
of
the
earliest
things
that
I
ever
automated
was
okay.
A
B
A
B
A
You
know
you
could
take
that
to
an
extreme
and
honestly,
this
sounds
cool
cuz,
it's
super
nerdy
but
like
if
you
could
like
print
out
a
packet.
It's
like
you
know
just
like
the
hex
of
a
payload
and
then
like
take
that
via
paper,
walk
it
over
to
another
terminal
and
type
it
in,
and
that
goes
into
another
network
court
and
then
you're
a
routing
protocol.
Oh
oh.
A
B
A
B
B
A
B
A
The
that
was
one
of
the
coolest
things
I
did
with
my
home
infrastructure
was
buying
an
ass
cuz.
You
can
run
all
kinds
of
cool
stuff
on
him.
Now,
like
there's
some
naz's
that
come
with
docker
installed,
you
can
run
like
your
own,
like
internal
version
control.
That's
that's
pretty
well
and
then,
of
course,
you
know,
storage
being
as
cheap
as
it
is
like
there's
not
really
a
reason
not
to
do
it.
Oh
yeah.
B
B
But
I
do
I
love
like
2
or
3
terabytes
of
just
videos
that
have
been
ripped
off
DVDs
and
it's
just
so
much
easier
than
figuring
out
where
the
DVD
is
stored
and
flipping
through
some
book
or
whatever.
And
you
know
it's
just
hey
you
just
go
to
the
NASA
interfacing
it
yeah
right
there.
That's
what
we're
gonna
watch
nice,
you
know,
and
so
all.
A
B
Sure
sure
sure
you
can
always
find
me
at
rule
11,
not
tech.
That's
that's!
That's
kind
of
the
hub
for
everything
that
I
work
on
and
do
most
the
time
I
cross
post
everything
there.
You
know
he's
finally
at
juniper
doing
whatever
I'm
doing
in
Jena
burg.
I
don't
know.
Maybe
somebody
juniper
should
tell
me
what
I'm
doing
well.