►
Description
Topic Summary:
Kubernetes represents one of the most mature container orchestration systems available today.In this session we will demonstrate how Rancher, an open source container management platform, automates the complex task of deployment and management of Kubernetes clusters. The presentation will cover Ranchers cloud agnostic approach to infrastructure management as well as it's production-grade Kubernetes distribution.
Speaker:
Jan Bruder is a Linux containers and open source enthusiast. He joined Rancher Labs in 2016 and is working with Rancher users all over Europe on next-gen containerized application environments.
A
B
A
A
All
right,
so,
let's,
let's
try
it
again
so
today,
I
want
to
talk
with
you
about
how
Rancher
how
to
deploy
cumulative
clusters
and
start
of
a
Dyson,
Germany
I've
joint
venture.
What
eight
months
ago
and
I'm,
mainly
working
with
people
in
Europe,
alanciar,
doing
trainings
two
deployments
and
spoke
at
that
and
rancher
has
basically
two
products.
So
the
first
one
is
our
hospitals,
potato
management
platform.
A
A
That's
why
we
support
a
number
of
august
ducker,
orchestrated
in
rancher,
one
of
them,
cuba,
ladies
and
that
is
also
the
most
popular
amongst
our
users
or
another
orchestration
engine
with
which
is
developed
path
and
that
evolved
quite
popular,
is
more
more
simpler,
orchestrated,
not
such
a
steep
learning
curve,
Cuba
naughty
and
it
supports
taco
composed
so
I
mean
a
lot
of
people
that
guide
with
containers.
How
can
you
live
with
dr.
Polk?
So
it's
very
easy
for
them
to
get
started
with
the
cattle
of
the
situation
engine.
A
Implement
a
number
of
cellulite
controls,
such
as
user
authentication,
audit
logging
and
role-based
access
control
measure
come
ship
shipped
with
its
own
load
balancer,
and
we
also
implement
implemented
dns
service
discovery
for
this,
for
the
services
that
are
running
in
Rancher
and,
most
importantly,
we
also
manager
on
overlay
networking,
and
so
that
is
sending.
There
is
always
a
problem
if
you
doing
container
deployments
with
multiple
hull,
maybe
even
look
at
it
in
different
zones
or
region.
A
Rancho
manages
a
single
overlay
Network
for
you
at
the
moment.
We
support
ip6
networks
and
VX
lamb,
but
since
we
are
implementing
the
CNI
specification
in
the
future,
we
will
be
will
be
able
to
plug
in
other
FBI
providers,
such
as
we,
for
example,
and
last
but
not
least,
Rancho
also
provides
crystal
storage
services.
A
I
mean
this
is
one
of
the
biggest
challenges.
I
would
say.
The
pocket
containers
is
that
containers
by
itself
at
the
marrow,
but
the
number
of
situations
where
they
really
storage.
With
one
brief,
you
deploy
database
as
a
container
or
some
kind
of
key
value
stores,
and
that
is
where
Rancho
helps
you
on
deploying
persistent
storage
volumes,
backed
by,
for
example,
and
ranch
AWS
ebf
or
you
can,
and
to
get
an
impression
wherever
you
sit
in
your
own
infrastructure.
A
A
So
now,
tick.
You've
already
created
is
currently
the
meeting
open
source
coach.
An
Orchestrator
is
quite
popular
and
it's
used
by
a
number
of
big
enterprises
and
it's
a
lot
of
users
come
to
rancho
Wanda
G,
looking
for
a
method
to
manage
their
kinetic
clusters
and
what
Q
is
basically
does
is
fitted
ultimate
application
deployment
and
make
sure
that
the
applications
which
you're
deploying
on
top
of
communities
are
healthy,
and
if
there
are
some
issues
it
recovers
containers.
It
also
handles
scaling
of
your
application
and
seamless
upgrade
and
so
I'm
going
to
skip
this.
A
And
if
you
look
at
humanities,
the
Cuban
it
is
deployment
excel.
If
you
can
see
that
it's
not
a
lot
of
simple
deployment,
it
has
many
solutions,
many
parts
and
it
can
be
quite
challenging
not
only
to
deploy
a
Cuban
Eddie's
cluster,
but
also
to
manage
it
and
to
maintain
Ock's,
upgrade
it
and
make
sure
it's
healthy.
So
our
challenge
is
so
that
human
I
discussed
imaginal
should
be
handled
in
the
same
way
that
cuban
is
itself
and
with
the
application
on
top
of
it.
A
A
The
second
challenges,
if
I
mean
a
lot
of
people,
are
just
using
a
double
yet
and
make
a
using
maybe
a
single
region
for
their
deployment.
But
a
lot
of
people
have
some
requirements,
but
they
need
to
put
it
pulling
on
computers
for
different
region
or
them
they
might
want
to
have
some
multi-region
kind
of
cluster,
and
that
is
quite
a
challenging
from
task
as
well
and
a
lot
of
challenges
managing
user
access.
A
They
are
pushing
out
a
lot
of
you
features
every
month
so
or
if
you
want
to
be
on
top
of
the
queue
latest
game
up
trading
is
also
a
challenge,
and
then
last
but
not
least,
for
the
case
that
there's
a
a
big
problem
with
your
state,
which
is
a
TD
in
Cuba
lady.
You
want
to
have
some
backup
mechanism
as
well,
so
here's
how
rancher
solve
this
problem.
A
Some
of
our
customers
include
Disney
and
UK
national
health
services
and
other
big
enterprises
and
government
agency,
and
this
is
how
we
approach
the
Cuba
native
integration.
Eventual
so
important
is
that
we
test
and
certify
all
every
Cuban,
a
distribution
that
we
push
out
to
the
ranch
and
you
look
and
we
simplify
the
configuration
I
mean
Cuba
natives
have
hundreds
of
configuration
flex,
and
this
is
really
quite
challenging.
So
what
we
want
to
do
is
we
want
to
simplify
configuring,
the
Phoenix
cluster
and
then
we're
managing
are
also
the
Lopez
natural.
A
So
this
is
what
I
address.
What
I
mentioned
before
is
a
simplified
configuration
so
that
it's
really
a
smaller,
more
manageable
set,
and
we
also
I
mean
cumulative-
includes
both
stained
table,
but
also
a
lot
of
fantastic
features
will
be
making
sure
that
our
distribution
only
includes
tested,
stable
production,
ready
feature.
And,
of
course
all
that
applies
to
wherever
you
deploy
on
your
cluster.
So
could
be
your
private
cloud
to
be
a
public
cloud
like
AWS
digitalocean
on
google
cloud
or
edger,
and
this
is
a
demo.
Will
you
over
the
infrastructure
services
that
we
manage?
A
So,
as
I
mentioned
again
already,
venture
has
a
built-in
load
balancer,
which
functions
as
an
increased
controller
with
cuban
edit
department.
Currently,
it
is
based
on
h8
proxy,
but
it
is
built
in
a
very
modular
way,
so
you
can
actually
plug
in
your
own
load.
Burlington
solution
like
traffic
or
engine
acuras
proxy,
or
even
you
could
even
interact
with
some
external
or
financing
solution,
and
the
current
process
network
is
using
IPSec
or
the
explan
the
tanks
on
your
environment.
A
We
also
mention
credentials
for
private
Tucker
registries,
which
most
of
the
deployment
will
help
you
most
enterprises.
Loans
will
not
use
the
top,
they
will
have
their
private
registry
and
on
top
of
that
also
certificate,
because
we
ever
you
have
a
load
balancer.
So
we
also
need
to
manage
certificate
for
the
node
connection.
A
One
very
important
aspect
of
range
is
that
we
support
multiple
environments
in
National,
Venture
controller,
a
singular
intro
server.
So
you
can
stand
up
very
a
couple
of
environment
on
a
matter
of
minutes
really
and
you
can
even
make
different
orchestration
technology.
So
you
can
have
a
Q&A
this
cluster
of
production
and
you
can
have
a
talker
or
cattle
cluster
for
for
DevOps,
environment
and.
A
So
and
next
of
also
the
authentication
second
event
sure
those
occurrence
the
activity
currently
integrate
with
Active
Directory,
ldap,
shamu
and
github,
and
there's
also
a
local
authentication
provider
and
the
authentication
always
works
at
the
environment
level.
So
you
can
have
different
in
lab
environments
with
the
different
users,
so
you
can
have,
for
example,
one
stating
environment
there
that
your
developers
effects,
so
you
could
have
a
production
environment
that
you're
up
as
access
to.
So
this
is
really
something
that
is
a
great
value
venture.
A
So
previously
we
had
our
own
user
interface.
Hope
you
be
ready,
because
the
Cuban
is
dashboard
was
really
there
yet,
but
we
felt
that
the
current
state
of
the
cumulus
dashboard
is
very
good.
So
in
rational
104
we
switched
out
our
own
user
interface
to
integrate
the
Cuban
Eddie's,
the
Native
communities
dashboard
and
the
same
is
called
a
catalog.
So
since,
since
the
beginning
we
had
our
own
gratiot
catalog
for
Cuba,
ladies
manifest
and
but
since
helm
came
out
and
getting
widely
adopted.
We
also
switch
that
out
now
by
default
we
are
running.
A
So
why
should
you
use
convention
first
of
all,
very
important?
We
want
to
provide
an
age
of
Cuban
cumulative
experience,
so
we
don't
do
any
propriety
stuff.
You
get.
You
get
full
control
over
your
community's
involvement
with
the
cube
control
command
line
interface
with
the
Cuban
is
except
org,
and
we
also
have
a
web-based
common
line
that
you
can
use
to
inject
with
cube
control,
and
the
next
point
is
that
wrench
provides
are
no
lock
lock
in
solution
and
is
when
I
will
percent
open
source
software
and.
A
Rancher
manages
your
upgrades
and
you
deployment.
So
if
you,
if
you
so
much
you've
ever
deployed
like
a
Cuban
Eddie's
component
of
the
examples
and
the
data
repository
can
see
that
there
are
instructions,
Crips,
basically
for
every
kind
of
scenario
like
an
OpenStack
or
an
AWS,
and
all
those
are
different
right.
So
eventually
you
will
get
the
mounted
Gnostic
deployment
capability.
A
A
Yeah
so
humanities
adventure-
sorry
about
that.
It
starts
by
creating
a
new
environment
and
an
environment
is
defined
by
environment,
template
which
is
customizable,
and
the
first
thing
that
you
choose
in
your
orchestration
engine.
So
in
that
case
would
be
Cuban
Eddie
and
then
you
can
decide
which
of
the
infrastructure
service
with
you.
You
want
to
use
from
Ventura,
so
the
first
one
is
our
health
check
service.
That
makes
sure
that
the
service
is
always
running
and
restart
or
recreate
containers.
A
Let's
fail
the
next
Reggie
DNS
you
can
either
use
reg,
DNS
or
hiding
it
for
people
that
are
already
in
Cuban
Eddie
that
which
tell
them
something
so
the
range
of
the
LS
is
offers
a
simple
dns
based
service
recovery,
where
you
just
weak
theory
for
the
a
records
of
a
specific
service
and
then
get
back.
All
the
IP
addresses
of
the
containers
that
populate
service
and
with
colleen
has
here
european
home
reserve
service
records,
and
then
we
have
a
ranch,
a
meta
data
service.
A
This
is
quite
useful
to
discover
what
services
are
running
and
also
supply
some
runtime
configuration
to
services.
It's
a
http-based
API,
it's
closely
very
similar
to
the
meta
data
service
that
you
have,
for
example,
in
AWS,
and
then
the
network
manager
manages
the
overlay.
Network
runs
the
DNS
meta
data
service
and
just
the
IP
address
management.
A
As
already
mentioned,
we
are
using
CNI
to
supply
network
into
containers.
That
means
you
can
use
any
plugin
that
are
available.
Christian
I
in
the
discus
chrome
steadily
so
I
mean
most
of
the
well-known
plugins
are
weak
and
calico
I
guess,
and
the
rancher
of
the
rash
of
the
cold
plugins
that
are
included
are
Piecyk,
overly
networking
at
the
excellent
implementation.
A
And
finally,
as
I
mentioned
before,
when
sure
provides
persistent
storage
for
your
containers-
and
we
have
a
number
of
implementations
for
that.
Rancher
EBS
is
not
on
the
list
and
also
NFS.
I
should
also
mention
a
Tibet,
so
those
supply
a
persistent
storage
on
to
your
part
or
change
containers.
If
you,
if
you're
using
another
orchestrator.
A
Cancer
yeah
so
have
a
look
at
the
difference.
Deployment
types
for
Cuban
I
discussed
with
the
rancher,
so
the
first
one
would
be
just
a
simple
standalone,
Cuban
and
cluster.
What
that
means
is
that,
basically,
all
the
Cuban
engines,
services
like
the
API,
culet
and
they're
running
on
a
female
of
course.
In
that
scenario,
you
don't
have
any
resilience,
but
it's
great
for
testing
stuff,
like
that.
A
That
would
be
the
that'll,
be
the
most
resilient
approach
for
egg
production,
one
and,
as
already
mentioned
before,
one
of
the
infrastructure
services.
That
van
should
use
to
manage
your
cue.
The
ladies
environment
is
a
health
monitor
of
texture
be,
and
so
this
means
that
your
Cuba
ladies,
the
services
are
monitored
by
three
different
health
check
instances
running
on
different
house
and
when
we
really
find
that
there's
a
problem
with
one
of
the
intrusive,
we
will
either
restart
the
container,
recreate
the
container
or
in
case
of
exit
e.
A
Now,
let's
try
this
place
is
what
I
already
said,
and
that
is
also
like.
So
these
are
the
different
strategies
that
the
AL
check
service
employs.
So
you
can
see
on
the
left
side
the
strategies
just
recreate
the
container,
but
in
the
case
of
actually
this,
this
wouldn't
be
a
good
approach,
because
SD
requires
a
quorum.
So
we
have
another
strategy
that
that
only
recreate
the
container
is,
is
there
still
a
quorum
and
what
we
do
also
for
the
SED
part.
A
So
as
long
as
you
have
that
have
that
folder
tax,
Python,
netbook
storage,
you're,
you're,
you're
good
on
the
backup
side,
and
then
we
make
it
also
very
easy,
in
case
of
a
disaster,
to
restore
the
backup
or
study
for
the
data
blank
for
there
to
be
cluster
yeah.
So,
as
I
said
before,
humanities
pushes
out
releases
very
quickly
and
eventually
we
follow
the
release
process.
We
test
in
features
and
if
they
are
in
our
new
production
ready.
A
We
put
them
in
our
cumulative
distribution
and
then
push
them
out
to
be
arranged
losers
and
how
that
works
is
that
in
your
user
interface
we
would
get
an
upgrade
basically
an
upgrade
parking
that
tells
you
that
there
is
not
paid
available.
You
don't
install
it
automatically.
So
it's
up
to
you.
If
you
want
to
pray
to
this
version.
A
So
I
said
already
ranchers
organized
univ
io
in
environment,
so
the
first
step
would
be
to
create
your
alignment
and,
as
you
can
see,
I
already
have
three
environments
running
there.
So
just
give
it
a
name,
then
you
select.
In
that
case
you
will
collect
a
few
days
of
the
situation
engine
and
this
will
miss,
deploys
just
a
default
template,
but
you
can
customize
this
and
then
you
can
add
additional
like
infrastructure
services.
So
very
I
mean
a
very
popular.
These
integrations
for
external
dns
that
create
DNS
names
for
your
circle.
A
You
just
keep
that
from
nope,
so
what
we
do
in
a
rancher
we
integrate
with
docker
compose
some
of
you
might
have
heard
of
that.
It's
basically
a
provisioning
application
to
create
resources
in
the
cloud
and
the
great
thing
about
it
is
that
it
supports
a
really
big
number
of
both
public
cloud,
but
also
private
cloud
solutions
like
OpenStack
or
Visa,
and
we
leverage
the
their
library
to
create
house
invention.
So
what
we
do
is
tarah
houses.
You
get
this
interface
here
and
all
the
obvious
stuff
is
also
available
through
to
our
CLI
application.
A
A
A
A
A
Okay,
so
you
select
the
DPC
which
opnet
in
which
you
want
to
deploy
those
number
did
by
default.
We
create
a
range
of
machine
security
group,
they're,
basically,
two
ports
that
need
to
be
open
and
you
with
security
group,
depending
on
which
noble
a
network
can
use
it
for
the
IP
site
that
will
be
the
port,
500
total
amount
of
gdp
and
that
allah
that
allows
your
host
to
connect
each
other
and
create
the
overlay
Network
and
be
created
automatically.
But
you
can
also
choose
a
custom
security
groups
here.
A
A
So
you
can
see,
but
these
house
are
getting
provisioned
and
just
depending
on
the
cloud
I
mean
this
takes
a
couple
of
minutes
or
maybe
just
two
minutes.
I,
never
guess
always
seems
to
take
a
little
bit
more
time
to
proving
they're
young
and
let's
switch
to
an
existing
environment
that
they
created
before
and
has
already
a
number
of
homes.
So,
if
you
see
here,
we
have
three
house
in
AWS
and
together
you
see
what
some
of
the
services
that
are
running
here.
A
You
can
see
that
the
genetic
dashboard
has
been
deployed
to
the
communities
cluster.
We
took
a
nap
service
and
killer
for
for
helm,
which
is
the
charge
catalog
that
knowledge
repository
and,
of
course
this.
What
this
doesn't
show
is
the
infrastructure
services.
So,
since
we
deployed
cubed
is
itself
as
containers,
we
pretend
can
have
a
look
at
the
infrastructure
stack
so
an
impulse.
It
respects
the
services
that
I've
managed
to
arrange
for
your
environment.
You
can
see
well,
as
we
have
a
health
check
stack
here,
the
ipsec
back.
A
We
have
Cuban
eighties,
we
have
the
increase
control
and
you
have
netbook
type
of
sex,
so,
for
example,
to
go
to
the
Cuban
at
expect.
You
can
see
that
these
contains
all
the
Neches
cumulative
services,
like
the
API,
like
the
couplet
that
is
running
on
every
host
and
the
proxy,
and
this
is
what
the
ranch
for
managers
and
you
can
see
that
quite
a
number
of
containers
running
here.
You
can
also
see
that
there's
crack
across
all
hosts
with
you
label
show
system
to
get
a
view
on
what
containers
the
running
on
you
chose.
A
You
can
see
I
created
a
zookeeper
deployment
earlier
and
you
can
run
reconsidered.
You
can
see
that
running
there
I
think
I
don't
have
to
explain
the
QA
this
UI,
but
only
hope
you
both
get
some
insights
on
most
of
the
work
load
running
and
you
can
also
create
and
create
new
applications
either.
Let's
fill
out
this
form
or
uploading
a
manager
in
the
Cuban
ad,
so
this
is
just
an
alternative
using
the
tube
control
command
line.
A
So,
as
I
said
before,
you
get
the
natives
experience
of
curating.
So
if
you
want
to
use
the
to
push
or
command
line
utility
on,
you
generate
your
your
configuration
file
here
and
save
it
on
your
desktop.
This
could
hear
your
credentials
and
your
own
URL,
and
you
know
quick
test
or
something
like
that.
A
You
get.
You
can
also
have
a
bed
base
with
also
that
based
and
cute
control,
and
also
also
the
helm
command
line
utility.
You
can
use
that
for
quick
stuff
in
the
engines
which
it
gives
me
you
are
and
yeah,
and
then
they
look.
Let's
have
a
quick
look
at
the
other
infrastructure
power
to
Atlanta
managers.
So
here
the
way
you
add
your
custom,
Tucker
registries,
which
could
be
a
doctor,
appt
k
or
some
custom,
can
always
be
on
registry
that
the
you
have
deployed
somewhere.
A
With
the
last
version,
we
have
experimental
support
for
secrets,
management
intervention,
either
backed
by
venture
itself
or
by
a
bolt,
if
you
know
from
hoshi
co-op
the
wall
solution,
which
is
it's
just
good
management
solution
that
we
integrate
with.
So
this
is
quite
describe
something
as
a
lot
of
people
at
weighted
score
and.
A
Also,
let's
have
a
look
at
the
Cap'n
do
so,
as
I
said
before,
we
did
well
based
access
control
in
venture,
so
this
would
cause
only
be
visible
to
be
the
office
pmo
to
be
happening,
and
this
is
where
you,
where
you
do,
where
you
can
see,
are
the
exit
controls
at
the
moment.
I've
configured
local
access
control
and
I
can
add,
I
can
just
add
the
account
here.
A
Of
course,
if
you
are
integrating
with
you
add
up
all
over
your
ad
that
you
have
on
promised,
you
wouldn't
be
able
to
add
new
account,
but
you
would
be
able
to
assign
privileges
to
existing
users
of
your
elder,
and
another
aspect
as
interesting
to
know
about
venture
is
that
you
can
do
to
our
venture
into
two
scenarios.
How
are
you
to
wave
so
you
can
deploy
just
a
standalone
venture,
which
is
basically
a
single
container
which
has
a
built
in
mysql
database.
A
So
this
is
sliced
for
probaby,
for
testing
or
for
40
blowing
around
chances.
You
know
you
note
the
notebook
to
play
around
this,
but
for
production
environments
we
definitely
recommend
to
to
do
eh,
a
type
of
deployed
which
basically
means
that
you
would.
You
would
deploy
two
or
three
or
four,
a
rancher
and
server
instances
and
a
single
high-performance
SQL
database
and
then
put
a
load
balancer
and
all
of
it.
A
A
The
data
and
application
plane
are
totally
tackling
or
not
totally,
but
almost
every
dimension.
So
when
you
venture
server
goes
down
which
I
don't
have
an
HH
better,
but
you
just
have
a
selection
change
and
for
something
that
goes
down
you
just.
You
merely
use
the
management
capability
for
a
cluster
but
holy
servants
that
you
have
already
deployed
on
your
notes.
A
For
that,
and
that
means
that
you
can
set
up
some
ice,
auto
scaling
from
outside
of
crenshaw
on
basically
Syrian
your
container
metric
that
we
also
expose.
So
if
you,
if
you
go
in
the
container
view-
and
you
click
on
the
given
container,
you
see
that
we
collect
these
metrics
village
map
metric
from
the
container,
and
the
same
is
true,
also
put
the
hosts.
We
also
collect
metrics
from
the
host
and
we
expose
all
those
VIP
I
and
then
you
can.
A
A
A
Presentation
to
see
thank
you
left
here
their
website
before
you
want.
If
you
didn't
try
when
I
get,
it
is
simple
I,
don't
know
if
you
do
that.
Small
simple
as
running
this
poker
run
command,
and
this
will
give
you
a
general
venture
instance
and,
as
I
said,
before,
measures
of
the
chores,
so
our
our
software
is
available
on
github
and
we
also
have
a
flag
channel
that
you
can
connect
to.
A
So
if
you
have
any
trouble
getting
started
with
rancher
or
you
have
any
questions,
you
can
just
connect
to
vanish
and
dash
fusion
at
flexkom,
and
there
will
be
there
will
be
a
people
helping
out
and
also
we
have
a
monthly
meetup
with
our
users.
Also
quite
popular.
You
always
have
around
160
to
200
pics
Thanks
and
those
are
always
sitting
down
specific
aspect,
and
we
be
all
that
most
of
the
times
we
have
at
least
one
demonstration
coma
from
a
rancher
user,
doing
a
demo
on
how
we
use
a
treasurer.
B
B
A
A
Sorry
yeah,
so
that
would
be
the
open
shift,
which
is
basically
right
at
version
of
Cuban,
a
coup
benetti's
rep
by
from
management
stuff
along
it.
But
this
was
the
yeah.
So
then
this
is
another
kind
of
a
different
approach.
They
are
definitely
locking
with
with
Cuba.
Ladies
I
think
I
do
they're
doing
some
great
stuff
there,
but
we've
got
a
lot
of
people
coming
actually
from
the
combat.
You.
B
A
Yeah
yeah,
so
you
need
to
soak
into
panik's
like
from
which,
for
which
version
you
use
a
plate.
So
if
you
do
a
major
upgrade,
that's
a
from
one
at
one
to
one
or
two.
You
definitely
have
to
back
up.
First
have
to
back
up
your
mysql
database.
We
do
have
snapshot
of
the
one
got
one
version,
and
then
you
will
be
able
to
roll
back
together
with
you,
but.
A
B
A
Yeah
so
I
think
the
like.
If
we
talk
about
environment
so
range,
especially
yet
the
the
separation
environment
and
then
each
environment
has
its
own
number
of
nodes.
So
when
we
talk
about
environment,
what
we,
what
I
saw
like
good
for
bigger
deployments,
where
maybe
around
50,
to
60
l
in
a
signal
environment
and
when
Rancher
scales
taste
a
lot
better
on
the
environment
level?
A
A
Yeah
well,
as
I
think
I
don't
know,
I
think
if
you
followed
by
preservation,
I
hope
that
you
get
that
you
got
to
I
mean
that
you
got
from
some
more
inspiration
goodbye.
Why
it's
a
good
idea,
I
mean
well
I
mean
most
people
that
get
into
Cuba.
Ladies,
they
want
to.
They
want
to
use
female
to
to
push
their
develop
development
and
to
be
more
agile
and
pushing
out
new
product.
So
what
I
don't
want
is
to
be
stacked
with
managing
Cuba,
ladies
environment
and
better,
as
I
mentioned
before,.
A
You
Jing
upgrading,
Cuban
environment
is
a
big
problem,
so
most
of
the
usual
flanger
they
actually
what
what
they
like
about
it.
It
makes
really
simple
for
them
to
deploy
cumulative
clusters
that
actually
work
and
get
a
very
stable
and
they
can
start
deploying
their
own
products
of
their
applications
on
top
of
it
and
don't
spend
a
lot
of
time.
Management
o'clock
yourself
and
also
I
mean
doll.
A
So
ugly,
they're
also,
like
other
features
that
we
add,
on
top
of
it,
like
I
managed,
like
I,
said
we
have
a
very
Pineridge
role-based
access,
control,
invention
and
also
we,
the
the
cloud
agnostic
approach.
I
think
there
is
a
flashpoint
prevention
and
the
other
infrastructure
services
like
our
o-line
networking
and
the
storage,
the
Pacifican
storage
seconds.
Of
course,
communities
has
also
has
all
current
equations
for
the
crystal
storage,
but
venture
of
reg
manages
both
both
free.