►
From YouTube: OpenShift Commons Briefing #71: Keeping an OpenShift Cluster Evergreen Using Cloudsoft AMP
Description
A common repetitive task for OpenShift administrators is keeping virtual machines up-to-date with the latest security fixes, patches and managing updates to the installed software packages while still maintaining application uptime for users. We show how Cloudsoft AMP can automate this process of repaving, helping you to achieve a secure and evergreen OpenShift deployment using the latest releases and updates from Red Hat.
Guest Speaker: Andrew Kennedy is a Senior Software Engineer at Cloudsoft and the founder of the Clocker project. He is a contributor to several Open Source projects including jclouds and Qpid and is on the Apache Brooklyn PMC. Areas of interest include Distributed Systems, Virtualization, Messaging, Information Security and LOLcats. Prior to joining Cloudsoft, Andrew worked for various investment banks as a Software Engineer and Security Consultant and has over twenty years experience in the IT industry.
A
A
Well
know
everybody
and
welcome
again
to
another
openshift
Commons
rethink
its
kind
Klaus
off,
along
with
my
member
of
the
open
shift
Commons
that
I
haven't
gotten
to
give
a
briefing
yet
I'm
really
pleased
that
they're
doing
one
today
and
they're
going
to
talk
about
keeping
open
to
evergreen
using
their
cars
loss.
App
and
they've
done
a
lot
of
work
and
patchy
book
when
I'm
really
pleased
to
have
andrew
kennedy
online.
B
Right,
thank
you
very
much
dad.
So
what
I'd
like
to
talk
about
here
is
often
shift
ever
agreeing
open
chest
and
how
we
can
manage
to
do
that
using
quotes
off
till
first
off
a
bit,
but
myself
I
work
for
a
company
called
quite
soft.
We
are
involved
in
the
open
source
community
as
mated.
Actually
cook
12
is
one
of
the
main
project
that
we
t
forget
to
do
my
workers
as
a
software
engineer,
developing
software
for
actually
Apache
Brooklyn
and
ago.
B
We
are
all
possible,
and
we
also
provide
that
as
a
commercial
product
as
close
of
them
and
with
support
command.
Much
similar,
like
right
after
etcetera,
store,
I'll
begin
with
an
introduction
to
quad
sort
out
just
explaining,
for
it
is
what
it
can
do,
its
application
management
software
so
just
showing
a
bit
about
how
all
of
that
works.
B
Well,
it's
useful
for
because
you
never
met
it
before
and
then
I'll
move
over
to
how
we
can
use
codes
of
them
with
open
shift
and
various
things
that
we
can
do
there
with
the
two
pieces
of
software
deploying
openshift
managing
open
chest
at
one
time.
How
do
we
do
that
management
and
how
that
lets
us
do
that
ever
bringing
managing
to
keep
open
chef
clusters
running
smoothly
and
up-to-date.
B
So
there
are
bits
of
software
that
are
considered
essential
today
that
no
one
had
heard
of
34
years
ago,
a
little
darker
is
the
canonical
example
there
and,
although
they're
essential,
so
complicated,
they're,
not
simplest
of
things
to
get
your
head
round
sometimes
and
because
they're
new
and
exciting
they're,
not
perhaps
through,
we
documented
yet
so
well
the
oceans.
You
will
buy
something
but
they're
still
useful,
so
people
still
want
to
get
these
things
out
there
and
finally,
there
Theodore
instructions:
here's
how
to
install
it.
B
Here's
how
to
get
it
out
onto
your
systems
into
your
data
sang
so
far,
then
you
come
to
the
day-to-day
operations.
The
run
time
management
and
often
that
that's
an
afterthought
so
console
app
is
trying
to
solve.
Some
of
these
problems
make
a
few
of
these
things
easier,
and
so
it
does
that
these
modeling
applications.
So
it
what
you
describe
what
your
application
will
look
like
describe
the
components
it's
built
up
off
and
capture
those
in
a
blueprint.
B
It
can
then
deploy
that
so
instantiate
that
will
print
into
cloud
or
into
a
container
ecosystem
somewhere
into
taco
into
amazon
web
services
into
OpenStack
forever
and
then
Isis
deployed
it
can
manage
that
deployed
application,
and
so
this
isn't
based
on
some
particular
particular
piece
of
software.
It's
it's
a
generic
capability
so
understands
how
to
dense
properties
that
can
go
and
talk
to
rest
api
is.
It
can
run
command
line
commands
over
ssh,
pull
back
attributes
find
out
what's
going
on
with
your
software
and
then
make
changes
to
that
soap
manager.
B
If
perhaps
you're
the
cpu
world
is
too
high
in
across,
though
it
can
scale
that
out
or
things
run,
you
scale
them
down
it
lets
you
make
changes
the
software
as
it's
running,
and
we
can
do
this
pretty
much
anywhere
that
we
have
a
supported
driver
for.
So
we
use
most
public
clouds
their
examples.
They
are
OpenStack
software,
as
your
I
was
a
web
services
that
we
can
deploy
to
public
clouds.
B
We
can
deploy
to
private
code,
so
your
vmware's
with
OpenStack
sand
and
quite
fine
grease,
and
they
can
be
on
physical
infrastructure
sore
ear
metal
machines
that
you
simply
have
credentials
for
or
twice
store
again.
Your
vmware
is
in
your
open
stacks
and
this
is
completely
separate
to
the
application
blueprint,
so
the
application
just
described
to
the
air
and
then
we
can
send
out
after
any
of
these
targets
and
the
things
we
can
create
there.
We
are
philippine
blueprints
for
others.
B
Voice
react
in
dec,
special
multi-color
of
elk
stack,
all
sorts
of
various
useful,
open
source
components
are
available
and
you
can
win
close
together
and
build
them
up
into
complex
apps
of
your
own
building,
your
own
wine
business
software,
your
stuff
written
in
python,
in
goal
in
java,
and
includes
laws
or
enclose
into
you
a
boyfriend
door.
Once
we
have
the
application
deployed
using
amp,
maybe
it'll
just
on
to
some
some
public
cloud
somewhere.
We
want
to
manage
that
at
runtime
and
I
mentioned
policies
and
doing
things
like
scaling
out
from
scaling
down.
B
Don't
we
use
metrics
that
we
pull
back
from
the
app
or
we
can
look
at
just
properties
of
those
machines
that
the
application
is
deployed
to
or
actual
application
specific
things
really
dig
into
database
find
transactions
per.
Second,
we
look
at
web
servers
find
out
the
rate
enci,
and
we
can
use
these
metrics
to
decide
what
sort
of
actions
to
perform
on
your
deployed
application
and
these
actions
again,
though,
their
specific
library
of
actions,
so
how
to
scale
across
the
things
is
not
intrinsically
dependent
on
what
those
things
are.
B
So
you
can
build
up
a
scalable
application
out
of
individual
components
that
you
simply
print
out
of
a
catalog,
but
it's
quite
a
powerful
capability
and
it
like
to
concentrate
on
designing
the
app
rather
than
worrying,
about
high
availability
policies
and
mechanisms
and
worrying
about
what
sort
of
monitoring
software.
These
are
just
components
that
can
be
added
in
to
your
boyfriend.
B
So
this
lets
you
use,
concentrate
on
the
application
more.
It
lets
you
think
about
how
your
business
is
using
the
app
so
you're,
not
you're,
not
worried
about
infrastructure,
and
it's
that
I
think
the
important
it
will
give
you
design
the
app
you
think
about
the
app
you
don't
worry
about
the
underlying
colleagues.
You
don't
worry,
but
we
underlined
bare
metal
machines
really
take
care
of
that
those
or
two
before
we
print
your
you're
able
to
send
that
off
to
some
college
somewhere
and
finally,
as
well
as
being
to
deploy
applications
to
the
cloud.
B
So
that's
all
well
and
good,
deploying
something
like
all
contesting
that
we
would
need
to
do.
But
then
the
applications
you're
going
to
deploy
they're
no
longer
going
to
the
clothes
are
actually
going
to
open
sisters.
Their
containers
or
not
software
running
on
virtual
machines,
door
and
Brooklyn
are
also
able
to
deploy
to
container
infrastructures,
we're
able
to
stand
up,
aku,
Pilates,
docker
swarm
or,
as
I'm
going
to
show
you
an
open
shift
infrastructure.
Don't
that
out
provision
it,
but
then
your
applications
that
you
want
to
run
plc
deployables
to
working
shift.
B
The
whole
context
is,
then,
not
just
the
thing
you'd
Apollo,
but
it's
also
a
target
and
you
can
even
construct
applications
where
some
part
of
those
who
is
running
for
the
inside
the
open,
safe
environment
and
other
components.
Your
run
oracle
database
is
running
on
a
bare
metal,
ething
machine.
Second
mixture
in
your
data
center.
Don't
we're
able
to
deploy
and
deployed
to
containerized
application
environments?
So
I'd
like
to
next
go
how
you
can
apply
all
that
to
open
shift
are
we
can
deploy
and
manage
an
open
shift
costs?
B
B
B
Don't
pretend
to
know
better
than
better
how
to
deploy
openshift
the
canonical
and
best
practices
are
described
in
the
ansible
playbooks
provided
on
github
by
right
app,
so
we
simply
use
them
and
we
can
configure
that
and
visiting
a
quite
powerful
configuration
language
apart
with
ESL
that
lets
us
describe
how
we
want
to
build
our
cluster.
So
we
can
configure
all
the
component
parts,
but
amp
is
also
been
able
to
introspect
into
it
with
deploy
your
purchase
and
work
and
see
what's
happening
to
these
virtual
machines
in
real
time.
B
What's
happening
to
the
operative
processes
that
are
running,
River
who've
been
a
key
component
to
ensure
the
docker
components
and
we
can
build
up
some
quite
complex
mechanisms
to
do
some
very
useful
things
to
this
deployed
coster
from
some
basic,
primitive
actions.
You're
right
increase
the
number
of
nodes
in
the
cluster
or
execute
this
command
on
this
Lord
of
this
cluster.
B
Don't
we
take,
don't
think
this
is
a
very
simple
blueprint
that
describes
how
to
deploy
open
to
the
origin
template,
and
this
doesn't
contain
the
nuts
and
bolts
of
the
deployment,
but
is
rather
the
configuration.
So
is
there
the
settings
you
the
ansible
playbooks
or
you
can
see
the
opposite
very
soon.
You
can
see
the
size
of
the
mast.
Acosta,
there's
two
with
3x
QB
machines,
five
little
to
a
good
they'll
applied
node
cluster
and
some
specifications
for
the
machines
that
we're
going
to
create
for
these
with
Reshef
Nora
to
a
number.
Of
course.
B
My
program
makes
this
and
with
nice
graphical
you
I
that
letters
other.
We
could
what
we're
deploying
so
here
you
can
see
just
exactly
what
we've
shown
in
that
bit
of
yamel
before,
but
perhaps
in
a
slightly
friendlier
fashion,
doing
see
those
are
open
costs
that
has
been
highlighted
and
you
can
see
the
table
of
properties
down
there
to
the
west
and
we
would
simply
click
deploy
or
you
can
save
it
to
a
catalog,
but
in
this
case
we're
going
to
deploy
that
and
we
get
a
running
applications.
B
So
this
is
the
part
of
the
application
that
contains
that
500
costco
certified
open
ship
notes
there
until
there
was
a
little
open
ship
or
go,
and
we
could
also
look
at
the
old
cost.
Also,
there's
a
tree
structure
here
in
see.
There
is
also
masters
and
the
heb
part
that
we're
not
just
creating
a
single
entity
that
represents
Custer
and
that's
it.
That's
in
a
pitch
black
box,
you
can
introspect
right
down
into
the
individual
chains,
so
I'm
going
to
try
and
flip
across
and
I.
B
Look
at
this
wife,
so
I'm
not
going
to
show
the
point
because,
to
be
honest,
that's
quite
boring.
You've
got
to
sit
away
well,
young
updateable
bunch
of
packages.
You
put
a
second
wait
while
things
are
freed,
Duncan
remote
labels.
This
is
one
that
I've
already
deployed
and
you
can
see
the
structure
but
sees
here
two
starts
here.
So
we
can
see
we
are
open
to
an
origin.
We
have
a
little
entity
that
manages
our
horse
trials
so
that
the
individual
machines
could
talk
to
each
other.
B
This
clog,
it
could
be
the
place
without
find
server
if
you
want
eh,
but
purposes
of
input,
Emeril's,
I'm
just
using
XE
roast
and
we
have
an
ansible
server
that
contains
the
link.
It
proved
down
the
pollutants
we
github
and
that's
really
excuse
the
clans
to
deploy
and
create
open
its
origin
costo.
You
can
see
the
main,
the
URI
with
our
different,
if
not
walk,
all
the
project-
name,
it's
not
actually
in
the
public
dns.
So
if
we
drill
down
a
little
bit,
you
can
see
there's
a
world
balancer
masters,
HED
costo
and
some
notes.
B
Xdd,
we
have
babies,
the
Masters,
it's
a
pair
of
Master
servers
and
a
single
rule
bouncer,
and
we've
already
seen
that
the
five
mode
openshift
node
cluster,
don't
go
back,
look
I
group
again
into
workmanship
nodes.
We
look
at
sensors
and
what
sensors
are
there?
The
mechanism
vamp
and
Rick
will
use
to
pull
data
back
when
things
as
deploy,
so
we
can
then
use
them
for
one
time
management
doing
T.
B
It's
got
roast
address,
which
is
on
doing
this
on
a
VPN
to
the
old
private
10
dot
addresses,
rather
than
public,
there's
information
that
we
were
installed.
Software
and
I'll
pull
back
the
whole
llamo
document
that
describes
an
open
shift,
no
its
status
and
in
fact
this
can
also
be
processed.
So
this
is
simply
the
first
part,
so
this
is
a
llamado
enjoy
the
you
can
breathe
in
here
live
24
capacity,
the
amount
of
memory,
and
that
can
then
be
fed
into
policy
that
will
extract
out
the
correct.
B
No,
perhaps
the
memory
status,
perhaps
out
of
this
message
is
it
will
be
strike
those
out
and
the
policy
can
then
decide.
Actually
I
need
to
do
something.
This
notice
unhappy
I
need
to
restart
the
service.
I
need
to
maybe
delete
it,
shut
down
all
the
parts
running
on
it
and
go
and
go
and
start
Amelia
to
pick
up.
We
can
see
here
it
opens
its
status
as
happy
it's
running
services
up
this
is
it
happen,
cost
of
it.
It's
doing
the
right
thing.
B
If
we
would
get
the
masternodes
again,
we
can
see
the
same
sort
of
thing
and
we're
pulling
back
information
about
versions
of
which
particular
bit
self
or
ended
up
being
installed
by
that
ansible
entities
played
ok
and
the
cost
of
itself
I'll
just
open
the
open
ship
console.
So
it's
it's
a
pretty
standard
open
shift
with
running
yet
we're
writing
a
single
regus
hard
on
this
and
I'll.
Come
back
to
that,
but
we're
just
playing
a
single
part.
B
It
is
not
not
doing
anything
particularly
interesting
at
the
moment
till
that
are
runtime
managed
and
the
provide
open
shift
costs.
Other
amp
has
created
all
of
this.
In
fact,
finder
thing
that
OSHA
was
the
activity
going
on
until
there
came
from
so
we
provision
this
or
a
blue
box
open,
OpenStack
environment,
so
I'm
created
the
virtual
machines,
have
provisioned
them
on
OpenStack,
and
it's
then
executed
a
series
of
actions.
B
It's
installed
software
customized
it
and
launched
it
and
then
the
same
thing
with
an
ansible
serigraph,
which
goes
we
provisions
a
noticeable
server
and
it
then
executed
that
waiver
can
deploy
the
terminologies.
Not
two
machines
into
that
running
would
shift
cluster.
Is
that,
but
you
saw
there
on
the
console
the
open
to
until
you
I,
don't
I'm,
going
to
leave
that
there
for
more
will
come
back
to
this
one
time
management
and
actually
execute
some
one-time
commands
on
it,
but
I'd
like
to
explain
a
bit
more
about
how
I
actually
works.
First.
B
So
I'm
going
to
go
on
to
automation
and
describe
how
we
can
automate
action
derivative,
so
we're
not
just
deploying,
but
this
is
automating
tasks
that
you
might
want
to
do.
While
the
cluster
is
running,
don't
we
want
to
monitor
the
cross.
Selectivity
saw
our
nation's
policies
that
look
at
say
the
number
of
pods
so
suppose
I
have
five
nodes
or
that
they're
configured
for
a
maximum
of
ten
poles
each
need
to
monitor
and
go
back
once
I
hit
some
water
mark.
B
Save
your
pods
I
know
that
actually
I'm
going
to
buy
a
space
soon,
because
I'm
going
to
hit
my
limit
of
50,
so
I
need
to
do
something
about
that.
So
the
first
thing
we
need
to
do
is
monitor
the
costs
that
week
at
these
sensors,
look
at
the
property
of
the
cluster
and
pulling
back
into
policies.
That
can
do
something
useful,
then
woody.
To
actually
do
that
useful
thing.
So,
adding
you
notes
and
that's
reasonably
simple
for
amp,
because
as
well
you
so
I
can
deploy
to
the
cloud
so
adding
a
new
North
Augusta.
B
Well,
we
simply
provision
an
adductor
machine
and
helpfully
right
have
provided
probably
boot
for
ansible
that
we
can
use
to
extend
a
cluster.
So
there
is
not
just
a
playbook
for
deploying.
There
are
perks
to
scale
up
and
scale
back
so
I'm
Tim
to
orchestrators
and
execute
these
further
flavor.
If
your
scale
out
of
any
HD
other
things
we
might
need
to
do
is
likely
to
replace
machines
so
perhaps
or
a
call
centers
field.
B
Maybe
if
you
know
the
distance
video,
it
is
a
bad
network
connection,
we
don't
know,
but
our
cost
of
five
has
become
a
customer
for
go
to
regain
that
capacity
with
me
too
again,
it's
people
adding
a
new
node.
It's
just.
When
does
it
happen?
We
monitor
we
monitor
for
failures
and
we
do
replace
the
field
machines
and
either
delete
the
field
ones
or
keep
them
and
investigate
later
or
what
was
it
but
happened?
Was
it
will
someone
hacking
in?
Was
it
just
a
dead
on
arrival,
machine
or
some
natural
cosmic
rays?
B
Something
like
that
so
and
we
can
grow
the
coaster,
still
adding
you
notes
when
we
need
to
because
there's
excessive
world,
but
perhaps
we
also
want
to
anticipate
that,
so
we
can
provide
a
policy
that
forwards.
The
Sun
will
sort
of
speak
that
nine
till
five.
We
increase
the
size
of
our
custard
because
we
know
that's
when
most
certain
jobs
are
running
by
the
clock
happens,
we
seal
it
back
down
to
some
minimal
level
and
we're
not
wasting
money
or
not.
B
Spending
our
Amazon
compute
bill
is
not
growing
enormously
with
when
we're
doing
nothing
with
the
costco.
And
finally,
we
might
want
to
update
blue
team's
in
the
cost
that
we
might
want
to
update
the
operating
system.
We
might
want
to
deploy
patches
or
perform
some
kind
of
operation
and
maintenance
of
the
underlying
infrastructure,
and
hopefully
do
this
without
upsetting
customers
that
are
using
the
cluster
without
affecting
the
applications.
We
simply
want
to
make
sure
you
have
the
right
type
of
security
patches
were
up
to
date
with
the
latest
services
running
or
costs
are
so.
B
B
It
is
five
children,
though
those
nodes
running
in
there
and
each
of
those
words
has
not
just
the
sensors
to
find
on
it,
but
it
also
has
vectors
these
vectors
where
ways
to
perform
tasks,
so
each
entity
can
have
certain
tasks
performed
on
it
and
which
tasks
you
perform
is
based
on
the
type
of
entity.
Sure
there
are
some
things
it
makes
sense
to
do
or
an
open
shift.
Node
there's
some
things.
It
makes
sense
to
be
on
the
master
and
other
things
that
it
only
makes
sense
of
Leon
say
the
answer
server.
B
So
only
the
ansible
server
understands
how
to
run
the
skin
lipolytic.
We
were
starting
and
stopping
the
open
shift
process.
That's
running
a
node.
Well,
that
only
makes
sense
to
happening
on
the
open
to
ignored
or
management
reports
that
only
makes
sense
rods,
the
Masters
that
would
give
it
to
do
that
wrapped
buying
40
command
line,
commands
or
executing
some
HTTP
REST
API
request
against
the
master
or
against
your
bouncer
and
resizing
at
the
coaster
is
well
the
thing
that
knows
how
to
do.
That
is
that
the
cost
of
component
itself?
B
It
knows
how
to
add
another
one
of
the
things
that
is
managing
across
the
rope,
so
the
master
of
cost
understands
how
to
add
a
new
master.
The
north
cluster
can
add
a
new
openshift
note
and
so
forth.
So
all
these
effectors
are
defined
on
these
individual
nodes
and
they
perform
sometimes
sinful
actions,
sometimes
a
little
bit
more
complex,
but
in
general
I'll
show
you
example
here.
This
is
the
definition
from
the
actual
underlying
blueprint
and
it's
just
it's
a
gal
file.
B
It
describes
the
structure
that
tree
structure,
but
it
also
describes
these
effective
actions
and
this
one
that
happens
to
be
selecting
the
oldest
Lords.
But
you
see
the
underlying
way
it
does
that
it's
a
real
CD
command
and
we've
worked
out
the
correct
parameters
to
passing.
We
want
the
key
all
this
nodes,
so
we
passing
this
dorky.
Oh,
we
can
build
up
some
quite
complex
things
from
just
very
simple
components,
so
we
can
defining
unloads
all
of
these
little
effectors
that
do.
B
There
is
quite
simple
things,
but
you
then
get
a
very
powerful
capability
out
of
that,
because
you
can
compose
them
together,
so
direct
orders
nodes
once
we
selected
them.
Well,
maybe
we
want
to
do
something
with
them
and
in
fact
the
evergreen.
The
whole
evergreen
process
is
simply
it's
a
composition
of
a
lot
of
these
very
small
vectors.
These
primitive
actions
putting
together
and
bundled
together
to
perform
their
useful
operational
actions
on
your
cluster,
and
you
know
ever-growing
is
just
one
of
those.
This
is
it's
just
an
example.
B
You
could
do
pretty
much
anything
you
wanted,
so
we're
able
to
combine
these
tasks
likes
like
the
oldest
and
we
can
do
different
cause.
We
could
say
sick
on
some
task,
so
do
a
then
B,
then
C
great.
We
could
compose
them
together
of
support
to
a
take
it
out
glute
and
see
that
into
big
dick.
These
I
put
that
in
to
see
whooping
else.
So
here's
a
list
of
nodes
just
do
extra
chopper
may
be
exes.
We
start
the
service
that
we
work
through
the
rest
of
no,
it's
restarting
the
service
choices.
B
So
if
the
service
is
unhealthy,
they
do
X
otherwise
dubai
transformations
and
replacements.
So
we
can
transform
the
content
or
the
output
of
some
command.
Maybe
we
get
the
output,
it's
an
IP
address
and
we
want
to
find
the
horse
named.
Something
like
that.
Don't
all
these
little
compositional
mechanisms
can
be
used
to
build
these
useful.
B
That's
how
in
fact
that's
how
we
build
a
task
like
repave,
so
we
want
to
keep
our
costs
a
fresh
and
we
say
that
the
phrase
of
your
evergreen
in
the
cluster.
So
we
do
that
and
really
simple
primitive
components
and
aunt
letters.
Bill
goes
up
into
a
task
like
we
gave.
So
what
exactly
do
I
mean
by
repaving
evergreening?
B
So,
if
you're
not
familiar
with
it,
evergreening
is
the
process
of
making
sure
that
you
cuz
ur,
doesn't
get
scale
that
the
operating
systems
that
the
pad
sets
that
the
security
op
fixes
that
are
applied,
that
you're
always
up
to
date.
So
one
way
of
doing
that,
one
way
of
evergreen
in
your
cluster
is
to
ring
pave
nodes.
Don't
repaving,
especially
it's
a
simple
set
of
steps.
We
got
here,
so
we
resize
across
those,
so
we've
got
say
five
nodes.
B
We
resize
it
so
we've
got
six
and
that
sixth
node
is
going
to
be
created
using
fresh
operating
system
and
when
we
do
the
update
or
we
call
an
update
command,
it's
going
to
pull
down
the
wake
of
security
patches.
It's
going
to
pull
down
the
latest
packages,
though,
were
we
know
that
when
we
create
a
new
virtual
machine,
we
get
the
fresh
operating
system,
the
fresh
pad
set
and
the
fresh
packages
that
we
need
to
update
the
cluster,
but
that's
just
the
virtual
machine.
B
So,
yes,
we
resize
the
costly
we've
got
a
new
fresh
vm.
That's
got
the
latest
set
of
hot
pictures
on
it.
We
then
need
to
scale
out
open
shifts
or
go
out
on
to
this
new
VM,
and
that's
where
I'm
still
comes
into
play.
Ansible
helps
us
by
giving
us
the
scale
I
playbook,
and
we
kick
that
off
and
the
new
fresh,
butchered
machine
becomes
an
open
shift
mode.
But
now
we've
got
a
six
nor
dog.
We've
got
a
bigger
cost
so
that
we
want
it
dog.
B
That's
where
that
select
oldest
little
task
sure
comes
in.
We
want
to
get
rid
of
some
of
the
old
goals
because
we
don't
meet
too
many
work.
We've
got
a
nice
new
fresh
one.
Do
we
find
the
oldest
and
then
we
run
the
open
shift?
Commands
we
say
lights.
Don't
please
nodes
take
them
out
of
the
cluster
top
all
the
tasks
on
them.
So
they're
running
some
pods
they're
running
some
processes.
B
We
stop
those
processes,
stop
pods
or
rather
tell
open
shift
to
do
that.
An
open
shift
will
help
we
start
from
and
we
start
them
on
some
machine
that
actually
has
space
for
them.
So
open
typical
drain
tasks
and
move
them
over
to
you.
Hopefully
our
new
virtual
machine-
and
we
can
then
get
rid
of
these
stock
nodes
that
aren't
running
anything
anymore
and
I,
just
taking
up
space,
kicking
up
cpu
and
memory
and
our
cluster
and
adding
to
our
monthly
bill
on
AWS.
B
Don't
that
laterz
repave,
a
single
or
up
to
a
couple
of
nodes,
and
if
we
want
to
make
sure
on
top
of
clusters,
is
evergreen.
We
simply
edge
of
this.
We
say:
let's
run
this,
we've
got
a
name,
Lord
custard.
Let's
run
it
any
time,
so
five
no
cost
to
run
at
five
times
once
a
week
and
that
five
times
will
clean
up
one
note
at
a
time
and
of
course,
though,
every
week
we
know
it's
evergreen
and
if
there's
a
problem
well,
we
can
stop
after
the
first
one.
B
We
can
say:
oh
okay,
there's
something
wrong
with
this
and
we
can
go
back
and
we're
not
going
to
build.
The
coaster
is
all
going
to
crash
and
burn
because
we
just
changed
everything
all
at
once.
It's
a
similar,
similar
mechanism
to
say,
Bluegreen
deployment
that
the
same
sort
of
concept
we're
doing
one
at
a
time
and
making
sure
that,
yes,
that
was
okay
and
we
can
carry
on
so
I'm
going
to
kick
off
a
demo
of
this
and
show
you
how
repave
actually
work
so
first
off
here
are
the
effectors.
B
So
we
have
repaved
cluster
on
the
open
sea
authority,
but
I'm
not
actually
going
to
use
that,
because
what
this
would
do,
it
would
repaid
all
five
of
the
machines,
all
five
of
these
open
shift
modes
and
that's
going
to
take
a
bit
more
time
than
we've
got
available
to
us
to.
Instead,
we
can
execute
the
repaid
nodes,
which
is
sitting
here
and
repave
nodes,
simply
retakes
k
where
k
is
some
parameter
and
we
can
repaid
single
nodes,
and
that
will
happen
a
bit
faster.
Rightly.
B
This
is
a
good
place
to
work,
because
you
can
see
in
this
list
of
effectors
the
things
that
an
ansible
server
key.
Do
you
can
see
the
plants
are
all
set
up
and
also
to
scale
out
be
like
nodes.
It
knows
how
to
stop
machines,
how
to
update
nodes,
and
if
we
look
further,
the
Master
will
find
that
the
other
things
I
talked
about
disabling
share,
doing
evacuating
roads,
selecting
older
stay
away
from
the
open
ship's
master,
because
it's
able
to
run
open
shift
commands
across
the
entire
cluster.
B
So
before
we
go
and
start
or
repave
or
I'd
like
to
do
here.
Is
our
offensive,
console
or
repair
flips
to
the
open
shift
web
console
I've
got
a
demo
project,
as
you
saw
before,
used
arias
door,
but
look
at
the
deployment,
so
there's
only
one
replica
there.
What
I'd
like
to
do
before
this
is
to
give
this
up
to
five,
because
we've
got
five
notes.
B
Oh
no
Michael
collisions
employment
application
so
that
list
of
parts
yes,
so
we
can
see
we
had
our
initial
running,
reduce
and
we're
going
to
five
regus
services
running
and,
if
I
up
to
the
command
line,
if
a
risk
of
knowledge,
so
we
can
see.
Here's
all
I
know
till
I
started
this
cluster
hour
ago,
and
I'm
also
going
to
look
at
the
list
of
pods
and
we
can
see
here
is
our
1
2
3
5
pods
different
X,
so
we
can
actually
work
with
you
to
the
pods
running.
B
B
B
But
I
don't
think
it's
that
lesson
instructor
to
watch
this
as
I
have
a
sort
of
visual
diagram.
That
shows
that's
more
clearly
how
exactly
this
is
all
happening.
So
we
started
with
the
initial
openshift
cluster
you
can
see.
There
are
no
each
more
telling
dr.
imaging
and
open
shift.
Are
we
have
pods?
Okay,
there's
a
lot
more
pods
running
here
than
on
my
little
demo
coaster.
Basically,
the
light
green
ones
here,
some
application
that
is
spread
across
all
five
nodes.
Don't
first
thing
that
happens
is
resizes
that
cluster.
B
B
These
are
the
ones
we've
gone,
dotted
lines
run
them,
so
amp
is
able
to
run
that
little
bit
of
shell
command.
That
I
showed
you
and
it's
selected,
the
two
oldest.
Wouldn't
you
have
notes
so
next
we
want
to
stop
anything
further
being
the
point
of
those
which
the
old
ones
we're
going
to
get
rid
of
them.
So
we
turn
off
k,
killing
which
is
again
single
command,
and
then,
after
that,
we
drain
odds
on
those
nodes.
So
you
can
see
the
dotted
lines.
B
We've
gotten
rid
of
the
pods
running
on
these
old
nodes
and
open
shift
has
helpfully
started
them
up
on
the
new
nodes
that
we
created.
This
is
all
possible.
We
get
remember
the
way
open,
chick
likes
things
happen.
You
know,
you're
running,
stateless,
pods,
you're,
running
stateless
images,
so
we
can
stop
the
start
and
lose
things
around
and
our
end
users
are
going
to
be
none.
The
wiser,
though
they're
going
to
single
application
continue
working
though
they
still
have
capacity.
B
There's
no
loss
transactions
Kingdom
early,
but
like
that,
as
long
as
your
applications
are
built
as
stateless
it
was
microservices,
it
was
applications.
Open
ship
was
able
to
juggle
them
alone
as
it
sees
fit
and
move
these
pods
to
the
waitest,
precious
nodes
that
finds
in
the
system,
because
of
course
it
can't
shed
your
anything
anymore
to
these
old,
the
notes
of
the
dotted
lines
around
them.
Don't
hang
up
then
again,
so
numbers
called
open
shift.
Let's
reshape
your
these
pods
to
another
node.
We
then
turn
off
the
open
ship
cortex.
B
We
don't
need
open
shift
anymore.
The
SSAT
game
stops
the
processes.
There
is
nothing
happening
anymore,
those
processes
of
knowing
of
any
interest
to
us,
but
we
still
have
the
virtual
machines.
Don't
wast
ask
and
please
to
go
and
release
those
virtual
machines
put
them
back
into
the
full,
maybe
talking
snack,
we
get
rid
of
images
so
shut
down.
Maybe
it's
AWS,
but
essentially
they
were
just
taking
up
space
they're,
taking
up
to
be
able
o
to
of
CPU
been
causing
that
you're
increasing
our
Amazon
Web
Services
their
original
bill
forever.
B
So
we
voiced
those
virtual
machines
but
disappear
and
were
left
with
bring
Paige
open
chef,
kostow
right.
This
is
simple
to
you
and
you
should,
if
you
imagine
we
do
this,
but
one
at
a
time
or
we
do
it
once
more
but
say
set
the
boundary
to
free.
We
end
up
with
five
old
kostner.
Every
virtual
machine
has
been
weird
replaced
and
applications
are
still
running
quite
happily
or
pods
are
still
quite
happy
and
end
users
of
one
wiser,
and
yet
they
have
a
more
secure,
more
up-to-date
piece
of
under
eye
infrastructure.
B
So
we've
been
able
to
do
all
this
to
improve
our
infrastructure
without
inconveniencing
the
end
users
running
applications
there.
Oh
I'm,
not
sure
how
much
I
can
take
a
note
with
this
time
tonight.
I
started
running
through
that
little
animation.
So,
let's
see
okay,
no,
that's
looking
good!
So
with
resize
the
notes.
We
then
updated
them
so
I'm
selected.
All
this
happened.
We
get
a
bit
of
50
things,
conducting
the
spring
to
the
list,
I'm
going
to
stop
the
oldest
ones,
let's
go
back,
and
then
we
actually
stopped
machines.
B
Those
were
running
on
so
bit
of
keeping
again
resize
by
Delta,
sorry
sizes,
five
minus
one,
and
what
that
does
it
says:
let's
do
away
the
machines
that
are
unwanted,
so
that
would
be
any
machine
that
had
open
shift
stopped
on
it
so
but
tapping
the
two
is
still
got
five
modes
and
our
cost
I.
It's
not
actually
agent,
Lee
intuitively
visible
here.
What's
happened,
though:
I'm
gonna
flip
across
back
to
our
command
line.
So
let's
look
at
two
things:
I'm
going
to
run
that
get
pods
command
again
and
we
can
see
there.
B
We
go
seven
minutes
all
seven
when
it
sold
seven
minutes
all
three
older
pods
I
with
21
and
1
minute,
all
aboard
for
the
fun
run
that
way
and
yes,
I,
ran
it
with
a
size
of
one
so
that
there
was
the
initial
pod
that
we
started.
Then
I
scaled
out
to
phi,
which
will
be
seven
minute
old
boards,
and
then,
where
is
this
prime
new
one
which
has
just
been
started
it?
B
So
this
is
the
third
part
that
got
moved
on
to
the
new
evergreens
node
and
in
fact,
if
you
look
at
the
risk
of
nodes,
we
can
see
that,
yes,
that
all
one
hour
old
except
years,
I've
got
14,
6
and
yep.
That's
exactly
right!
That
new
node
is
what
of
my
new
pod
is
running
on.
Don't
you
see
that
one
load
has
vanished
a
new
one
has
been
created
two
minutes
ago
and
the
pods
have
been
shuffled
around
and
our
application
still
as
five
replicas,
it's
still
quite
quite
running.
B
There
got
to
be
able
to
see
if
we
look
at
our
console.
In
fact.
Yes,
we
can
see
so
there's
our
two
men
sold
this.
This
little
little
media
trying
here
go
activities
and
the
only
other
thing
that
might
be
a
bit
useful
to
going
to
hear
is
yes,
we
can
do
that
manually
and
that's
great.
Oh,
we
want
to
do
this
automatically,
so
here's
our
list
of
things
that
happened,
but
if
I
look
at
policies,
we
see
periodic
effector,
and
this
periodic
effector
is
running
the
screen
David
to
Lourdes
every
day
or
we
could.
B
We
did
the
entire
cost
or
we
could
run
this
div
cost
so
once
a
week.
The
principle
is
that
we
were
able
to
execute
these
vectors
automatically.
These
effectors
and
tone
run
sequence
of
further
primitive
commands
so
resizing
starting
stopping
scary,
alright,
and
that
happens
on
some
kind
of
schedule.
Schedule
policy,
Sheffield
done
program.
B
So
I'm
going
back
to
the
ride
smoke
and
go
just
summarize
what
we
see
here
and
why
it's
useful.
So
first
off
we
talked
about
arm
and
pumpkin.
Do
go
out
like
to
manage
your
applications,
go
your
applications
scale
and
dynamically.
Depending
on
the
work
goes,
it
gives
you
a
common
management
platform
so
that
mechanism
that
we
saw
to
manage
openshift
infrastructure
is
exact
through
the
same
front-end,
an
API
and
user
interface
that
we
use
to
manage
applications
running
on
or
purchased
so
completely
the
same
piece
of
tooling,
clearly
agnostic.
B
So,
although
we
across
platforms,
although
we
on
OpenStack
there,
it
could
run
on
a
SS,
it
could
run
on
private
or
public
college
doesn't
matter.
It
can
also
run
on
containerized
platforms.
And,
finally,
it's
a
simple
and
consistent
management
mechanism.
I
think
we're
to
build
up
from
certain
components
into
complex
apps,
using
a
library
of
effectors
and
actions
and
a
library
of
entities
and
templates
and
blueprints.
So
you
can
orchestrate
not
just
your
application
and
the
mechanism
for
defining
your
application,
but
also
the
operations
that
happen
to
your
application
at
runtime.
B
So
we're
able
to
deploy
opens
with
origins.
We
then
are
able
to
identical
to
deploy
applications
to
open
shift
in
the
same
way
that
we
can
deploy
openshift
itself.
We
can
manage
a
running
open
shift
to
look
at
running
virtual
machines
with
properties
of
those
go
to
machines
and
we
can
orchestrate
actions
across
the
whole
application.
So
we
can
make
things
happen.
The
ring
aid
of
an
entire
cluster
and
use
that
to
Evergreen
open
shift.
B
So
with
this
mechanism,
we
can
deploy
your
consist
and
make
sure
that
at
all
times
we
have
the
latest
version
of
whatever
security
patches
and
fixes
are
required,
and
this
all
happens
automatically
in
the
background.
I
I
just
add
that,
of
course,
we
this
is
more
solution
or
alternative
to
your
heart
rate
comes
out
and
you've
got
to
fix.
Everything
literally
I
feel
we're
not
recommending
it.
Oh
it's,
okay.
It
will
be
repaved
next
week,
but
you
can
use
the
same
mechanism.
You
we
could
simply
have.
No
heart
rate
is
out
right.
B
B
A
A
Think
I
have,
I
have
a
couple
just
so
maybe
clarification
which
and
I'm
thrilled
that
we
finally
captured
will
this
information,
and
I
can't
wait
to
get
the
feedback
from
people
who
are
using
it
and
what
they
think
of
this
approach.
I
see
the
following
a
lot
of
problems.
I'm
have
to
say
to
me
this
is
the
first
time
I
heard
the
phrase
repace
I've
heard
evergreen
before
that
repaid
is
it
doesn't
mean
to
me
that
I'm
Duncan,
looking
which.
B
Yeah
so
I,
don't
look
you
notice,
but
I
did
I
look
at
the
same
sheet,
but
I
the
readers
I
deployed
was
the
reduce
with
the
federal
storage.
I
didn't
choose
the
persistent
volume,
and
essentially
you
wouldn't
want
to
use
or
localized
volumes,
but
you
can
be
paid
if
you're,
using
volumes
that
are
coming
from
an
external
south
coaster
or
something
like
that.
Then
there's
no
problem
at
all,
because
cousin
at
ease
and
10,
so
pretty
smart
enough
to
restart
the
pod
somewhere
else
and
reconnect
cut
volume
from
your
san
yeah.
B
If
you
reuse,
a
local
storage,
of
course,
you've
got
a
problem.
Your
app
is
going
to
have
a
bad
time
because
when
it
started
separate
can't
find
any
of
its
data,
but
I
think
so
always
you
follow
best
practices
and
everything
should
be
okay,
and
that's
one
of
the
next
steps
that
I'd
like
to
work
on
with
this
openshift
appointment
or
you'll
notice
that
it
gave
them
to
contain
a
deployment
of
San
door.
Though
there
wasn't
a
storage
costs
low
point
there.
B
So
I
think
that
would
be
next
step
is
to
modify
the
blueprint
to
deploy
a
storage
cluster
and
make
sure
that
we
can
repave
and
redeploy
the
application
is
connected,
two
volumes
on
that
storage
cluster
and
also
think
about
what
to
do
to
repave
and
update
the
machines
are
actually
running.
The
storage
costs
are
potentially
Atleo.
You
have
to
worry
about
never
rain
and
availability
of
your
storage
and
that's
another
exciting
problem:
oh
yeah,
that's
it!
The
district.
A
Exciting
problems
out
there,
you
know
where
I'm
hoping,
maybe
some
of
you
guys,
will
be
coming
to
read
at
summit
in
Boston
and
will
pop
into
the
open
shift
common
gathering
on
the
day
before
it
makes
a
first
in
Boston
there'll,
be
a
lot
of
real
world
production
scale,
customers
talking
and
who
we
have
nine
or
ten
of
them.
Now
who
are
going
to
talk
about
their
sack
and
their
their
implementations
of
open
ship,
and
it
would
be
good
to
see,
because
I
think
that
there's
this
is
the
PDC's
and
system
storage.
B
To
run
the
upgrade
play
with
no,
but
that
is
one
of
the
things
I
had
in
mind
as
being
one
of
the
capabilities
of
these,
these
effectors
so
yeah,
absolutely
that's
a
further
vector
that
would
be
quite
simple
to
design
based
on
those
work.
That's
been
done
here,
yeah,
instead
of
calling
scale,
we
call
upgrade
and
we
move
from
your
workmanship
origin,
whatever
to
X
next
week's
yeah.
That's
that.
A
A
Useful
to
have
and
have
it
tested
and
have
those
have
that
implemented
here,
because
that
really
is
Cooper
Nettie's
being
one
of
three
months
for
release
cycles
and
an
open
shift
origin
being
rapidly
coming
up
on
its
heels.
I
think
we're
always
one
point
behind
that.
We
can
take
three
months
clearance,
sale
and
on
the
power
projection
versions
of
these,
but
that
I
think,
on
the
upgrading
process
is
something
that
is
always
top
of
mind
for
everyone.
Who's
in
operations
using
open
shift
pretty
much
any
more
yeah.
B
C
A
I
think
that
we,
if
you're
interesting
it
and
that's
kind
of
why
I'm
thinking
that
you,
if
you're
not
coming
to
Red
Hat
summit
or
the
open
gathering
it
right
out,
because
I
think
over
80
80
sessions,
moms
that
are
merely
just
people
who
are
using
it
in
production
and
that
their
feedback,
words
I,
think
informed,
and
they
are
several
hearts
on
real
world
use.
For
this
approach.
I
probability
of
showing
that
the
good
kind
of
to
get
together
and
introduce
you
to
some
of
these
folks
is.
C
A
A
Shift
org,
slash
gathering,
we
do
three
is
any
years
looking
at
each
coupon
and
noun
1
the
day
before
we
write
have
some
us
and
we
love
I,
think
I
really
didn't
you
ought
to
be
in
the
room
and
can
be
part
of
that
conversation
cuz.
That's
where
you're
going
to
find
people
like
Dale
and
others
I'm
going
to
help
you
out
visit
and
I'll
shoot
you
guys
a
note
and
code
couldn't
figure
out
this
group,
Whitaker
mofos
I
mean
if
anybody
else
wants
to
I'm
happy
to
endorse.
A
A
It
is
because
I
don't
think
that
the
face-to-face
time
that
really
makes
the
difference
in
quality
briefings,
I,
do
and
I
and
there's
one
of
the
people
who
to
watch
them
afterwards
and
people
who
joined
us
today
and
I'm
thankful
for
that.
So
she
would
ask
a
few
more
questions,
but
kidding
I
think
we
get
a
very
good
job.
So
maybe
that's
why?
A
There's
not
too
many
questions,
but
I
think
that
a
lot
of
long,
a
lot
of
good
synergies
here-
and
this
is
very
useful
and
it
and
for
me,
I,
think
today,
the
first
time
Andrew
and
I
think
you
did
a
really
good
job
at
explaining
the
value
proposition
for
what
you're
doing
at
cloud
soft.
So
it
was
very,
very
big
formal
to
thank
no
thank
you
and.
A
And,
as
I
said,
we
gained
all
these
will
be
up
on
our
YouTube
channel,
because
the
page
different
shapes
on
uses
and
I
am
behind
a
few
as
some
organisms
I,
usually
post,
each
of
them
as
a
blog
and
I've
caught
up
in
a
few
other
things
gone
behind
about
three
or
four
sessions,
so
I
will
get
beat
up,
hopefully
by
the
end
of
the
beast
on
in
the
backlog
and
they'll,
be
up
on
blog,
open
shift
or
my
comrade.
It
won't
be
there
soon.
So
thanks
again
Andy.