►
From YouTube: Webinar: Enhancing your Control over Kubernetes with Spinnaker and Continuous Delivery
Description
Developers have to deal with a huge number of tools that all have to work together in order for you to deliver software. By orchestrating resources such as containers and services, Kubernetes goes a long way towards decreasing the pain of creating applications, but some of the items that are out of scope, such as notifications and approvals, making sure the right artifacts are going to the right place, and orchestrating deployment strategies above the Kubernetes object level are non-trivial issues that can make or break your ability to provide Continuous Delivery. Fortunately, Spinnaker, the open source Continuous Delivery tool, ties together many of the tools we use every day, such as Kubernetes and Jenkins, to provide enormous power over your software pipeline to solve these problems. In this webinar, you will learn:
- How Spinnaker concepts such as clusters and server groups relate to their Kubernetes counterparts
- How Spinnaker deployment on Kubernetes works, including integration with additional tools such as Jenkins, Helm, Terraform or Prometheus, to simplify and standardize operations
- How to deploy Kubernetes resources with Spinnaker
- How to create a Spinnaker pipeline that integrates automated and manual approvals to ensure the quality of deployments
- How to use Spinnaker to automatically deploy to multiple Kubernetes (and non-Kubernetes) clusters simultaneously to easily create multi-cloud solutions
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
Join us for KubeCon + CloudNativeCon in San Diego November 18 - 21. Learn more at https://bit.ly/2XTN3ho. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
A
Welcome
everyone
to
our
regular
CN,
CF
webinar,
serious
spit,
that
is
regular.
Continuation
of
our.
It
appears
after
the
cube
control
native
cone,
one
of
the
biggest
events
that
has
happened
in
in
the
communities
in
the
chlorinated
world,
and
this
happened
last
week,
and
today
we
have
our
presenter
Ernie
chase
from
Miranda's
who's
been
described
as
how
to
enhance
your
control
over
cabinets
with
spinnaker
mass
delivery.
B
B
So
today
we
are
going
to
be
talking
about
spinnaker
and
continuous
delivery
and
how
that
relates
to
kubernetes
before
we
go
on
I
just
want
to
kind
of
give
you
a
quick
heads-up.
If
you,
if
you
are
interested,
we
have
a
beta
of
a
product
that
is
based
on
spinnaker,
but
we're
not
going
to
really
get
into
that
here.
I'm
just
gonna
kind
of
drop
that
as
a
link
all
right.
B
B
If
you
are,
if
you
are
already
familiar
with
continuous
delivery,
I'm
going
to
ask
you
just
to
indulge
me
for
a
minute,
particularly
because
a
lot
of
there
are
a
lot
of
people
who
have
heard
of
it
and
they
think
they
know
what
it
is.
But
if
you
really
dive
in,
but
if
you
really
dive
in
you
find
out
that
they
are
not
necessarily
familiar
with
how
it's
really
really
done.
So,
let's
just
take
a
minute
and
refresh
so
you're
a
developer.
B
But
the
idea
here,
the
CI
part
of
CI
CD,
is
that
you
build
your
code
and
you
don't
just
throw
it
over
the
wall.
You
build
it
and
you
have
an
automated
pipeline.
That
tests
it
you've
got
your
unit
tests.
You've
got
your
infrastructure.
Your
integration
tests
you've
got
acceptance
testing.
All
of
that
should
be
done
in
an
automated
way
in
order
to
make
sure
that
every
time
you
commit
changes,
you're
not
breaking
things,
that's
the
whole
idea
behind
continuous
delivery.
B
Is
you
one
fix,
doesn't
take
the
whole
thing
down
just
because
somebody's
working
on
something
doesn't
mean
that
the
whole
thing
is
unavailable
until
they
are
finished
with
that
now
that
build
and
test
process
creates
an
artifact.
It's
some
binaries,
it's
whatever
it
is,
and
those
are
passed
into
the
actual,
continuous
delivery
part
of
the
equation,
and
that's
where
it
gets
handed
off
to
operators.
These
are
the
people
who
make
sure
that
the
code
can
go
where
it
needs
to
go,
and
normally
that
is
taken
care
of
with
scripting.
B
So
you
want
that
to
be
as
automated
as
possible.
Okay,
now,
once
you've
got
that
script,
that
script
typically
deploys
with
a
particular
release
strategy
and
those
strategies.
Are
there
there's
any
number
of
different
types?
There's
we
you've
got
them.
You
see
them
listed
here:
blue
green,
also
known
as
red
black
we're
going
to
talk
a
little
bit
about
that
later,
Canary,
which
you
know
where
you'd
roll
lab
a
little
bit
and
make
sure
that
it's
working
and
then
gradually
roll
it
out
more
rolling
updates
this
kubernetes
developers.
B
You
are
likely
familiar
with
that
where
you
know
you,
you
roll
out
a
little
bit,
and
then
you
take
down
the
old
one
and
so
on,
but
you'll
notice
that
these
release
strategies
are
independent
of
where
it's
actually
going.
That
is
the
ideal
situation
where,
as
a
developer,
you
just
need
to
build
your
code
and
then
it's
gonna
end
up
wherever
it
needs
to
end
up.
Maybe
it's
going
to
kubernetes.
Maybe
it's
going
aw
ask.
B
B
B
On
top
of
that,
kubernetes
applications
have
a
lot
of
moving
pieces
in
terms
of
the
infrastructure,
ideally,
and
really
honestly,
if
you're
not
doing
this,
you
can't
really
call
it
continuous
delivery,
but
ideally
your
infrastructure
itself
is
treated
as
code,
so
all
of
this
has
to
be
tested.
It
all
should
be
scripted
now,
there's
a
lot
of
problems,
though
a
lot
of
this
is
handled
in
an
ad
hoc
way
in
a
lot
of
organizations.
B
They
do
it,
and
now
the
problems
with
that
is
you've
got
security
issues.
First
of
all,
as
kubernetes
developers,
we
work
with
a
lot
of
containers,
they've
done
studies
and
show
to
show
that
I
forget
what
the
actual
percentage
was,
but
a
significant
percentage
of
images
on
docker
hub
have
preexisting
have
pre-existing
vulnerabilities
on
them.
I
think
I
want
to
say
it
was
an
average
of
like
16
vulnerabilities
per
image,
or
something
like
that.
B
But
don't
quote
me
on
that,
because
I'm
not
sure
what
the
actual
number,
but
it
was
an
eye-opening
number
now
the
way
to
get
away
from
that
is
to
produce
what
are
known
as
golden
images,
so
your
operations
team
might
put
together,
okay,
if
you're
going
to
use.
You
know
this
Python
base
image
great,
but
you
need
to
use
this
one
here,
because
we've
already
made
sure
that
it
doesn't
have
any
vulnerabilities
on
it.
We've
already.
You
know
locked
down
this,
that
the
other
thing
and
so
on.
B
B
Which
you
know
if
you
haven't
heard
of
it
by
now,
you're-
probably
not
in
this
webinar,
huge
huge
security
and
compliance
burden,
but
that's
not
the
first
time
it's
come
up.
There's
there
was
so
many
sarbanes-oxley
if
you're
in
the
u.s.,
you
undoubtedly
have
heard
of
HIPAA
and
all
the
security
involved
there.
So
when
it
comes
to
your
applications,
you
really
need
to
know
what
is
running.
B
Where
is
it
running
and
who
said
it
should
run
there,
so
you
need
to
be
able
to
track
all
of
that,
and
all
of
that
involves
a
great
deal
of
coordination.
You
need
to
know
the
who.
What
when,
where
and
how
is
this
all
going
to
work
and
the
problem
with
@ha
is
well,
it
is
ad
hoc,
so
one
of
the
one
of
the
things
about
spinnaker
is
it
takes
all
of
this
and
puts
it
into
a
single,
a
single
auditable
interface
for
continuous
delivery.
It's
incredibly
powerful
it
takes.
B
B
The
the
pace
of
development
is
really
quite
rapid
to
the
point
where
you
go
to
the
bi-weekly
meeting
and
they
say:
oh
yeah,
we've
got
a
new
version
and
it's
no,
it
blinks
an
eye.
So
yes,
all
right
anyway.
Let's
do
a
quick
tour
here.
I
have
I
have
a
problem,
because
this
interface
is
in
my
way.
Okay,
here
we
go
so,
let's
just
kind
of
do
a
quick
tour.
This
is
not
a
vanilla,
install
of
spinnaker
for
various
reasons,
technical
that
have
to
do
with
my
laptop
and
how
the
technology
gods
hate
me.
B
But
this
is
basically
just
a
look
at
Moran
chess
application
platform,
which
is
amaranthus
branded
version
of
spinnaker,
with
some
things
on
top.
So,
but
all
of
this
is
going
to
apply
to
generic
spinnaker.
Ok,
so
you
can
search
for
things.
You
can
group
things
by
projects,
but
mostly
we're
going
to
look
at
applications
within
applications.
You
have
clusters
and
pipelines
and
tasks.
B
B
B
So
this
is
kind
of
an
example
of
what's
going
on,
you
have
load
balancers,
which,
or
the
equivalent
services
will
we'll
get
into
that
in
a
minute,
but
just
kind
of
to
give
you
an
idea
of
where
we
are
so
all
of
this
comes
together.
Now,
let
me
go
back
to
my
slides
okay,
so
how
spinnaker
helps
with
all
of
this
is
again
it
provides
you
a
single
source
of
truth
for
all
of
these
deployments.
B
If
you
want
to
know
how
something
works
and
how
its
deployed
that's
where
you
can
go,
if
everything
is
always
deployed
through
spinnaker,
then
there
is
no
question
of
spinnaker
being
being
up-to-date,
because
if
it's
not,
then
it's
simply
not
going
to
work.
It
gives
you
on
an
audit
ability,
because
you
can
see
who
did
what
and
you
can
create
these
sort
of
templated
pipelines
that
use
these
verified
images
and
code,
those
golden
images
that
we
talked
about
before
all
right.
So,
let's
kind
of
jump
in
to
spinnaker
for
kubernetes
speakers.
B
Okay,
so
we've
got
if
we
start
at
the
bottom.
Pods
are
known
as
instances
you
don't
do
much
with
those
directly
through
the
interface,
but
you
can
see
them
in
server
groups.
We'll
talk
about
that
in
a
minute.
Server
groups
would
be
the
equivalent
of
a
replica
set
okay,
so
it
would
be
the
actual
code
and
all
the
configurations.
How
many
do
you
need?
Where
does
it
go
and
so
on
and
so
forth?
B
B
Okay,
so
moving
sort
of
to
the
outside
a
kubernetes
service
would
be
a
load
balancer.
So
that
would
be
how
we
can
direct
traffic
to
particular
instances
within
spinnaker,
which
in
our
case
would
be
pots
security
groups
I.
You
have
probably
heard
of
those
in
terms
of
you,
know,
class.
You
know
how
cloud
providers
refer
to
them
for
kubernetes.
We
would
refer
to
them,
obviously,
as
network
policies,
all
right
so
sort
of
putting
that
all
together
with
this
interface
out
of
the
way,
so
instances
are
grouped
together
into
a
server
group.
B
Server
groups
are
grouped
together
into
clusters,
and
clusters
can
be
grouped
together
into
applications
all
right
now,
that's
all
great,
but
how
do
we
actually
do
the
delivery?
How
do
we
actually
do
the
deployment?
So
what
we
do
is
we
create
pipelines,
and
we
look
at.
We
looked
at
those
pipelines.
A
minute
ago,
each
of
these
pipelines
is
made
of
stages.
B
Okay,
so
each
of
these
is
individual
stages
and
if
we
were
to
say,
let
me
go
to
a
different
application
here,
because
I
don't
want
to
so.
Let's
just
kind
of
take
a
quick
look
here.
So
if
I
were
going
to
create
a
load
balancer,
let's
say
I
wanted
to
put
an
application
on
kubernetes.
So
I've
got
my
coop
local
account,
which
is
my
kubernetes
my
company's
cluster,
as
in
kubernetes
cluster,
not
spinnaker
cluster
see
I
hate
that
they
do
do
that,
but
you
get
the
idea.
B
B
I
might
have
webinar
production
demo,
I
can
set
the
ports,
you
see
note
port
I
can
add
a
new
port
and
so
on.
You
are
you're
recognizing
a
lot
of
these
I'm
sure
so
cluster
IP
node
balance
or
node
port.
These
are
all
you
know.
These
are
all
familiar
to
you
as
kubernetes
developers.
We
can
add,
you
know
we
can
add
external
IDs
service,
annotations
and
so
on.
So
I
can
go
ahead
and
create
this
and
there's
a
task
view
that
shows
all
the
tasks
and
their
various
their
various
progress.
B
I've
got
a
couple
here
that
I
stopped
for
various
reasons,
but
you
can
see
everything
that
was
done
and
you
can
see
who
did
it.
So
that
is
a
very
important
piece
of
a
very
important
piece
of
the
puzzle
we
have.
So
if
I
go
over
here
to
clusters
and
I
want
to
create
a
server
group,
so
now
I'm
ready
to
go
ahead
and
create
some
infrastructure
on
which
to
put
this
application,
so
I
can
I
can
decide.
You
know
where
I
want
to
I'm.
B
B
And
I
mentioned
that
the
technology
gods
hate
me
today
and
then
I
can
say:
I
can
define
which
containers
that
I
want
now.
What
shows
up
here
under
containers
is
going
to
be
based
on
the
repositories
that
are
configured
into
spinnaker
when
you,
when
you
create
it.
Okay,
when
you
install
it
rather
so,
for
example,
if
I
wanted
to
say
created,
Jenkins
Jenkins,
server,
group,
okay,
so
I'm,
saying,
okay,
yeah
I
want
to
make
it
a
deployment.
Here's
a
rolling
update,
I
want
my
max
surge
to
be.
B
You
know
three
and
max
unavailable
one
now,
but
where's
it
going
to
get
served.
Okay.
Well,
here's
our
here's,
our
webinars
staging
demo
load
balancer
that
we
created
a
minute
ago.
We
can
set
up
auto
scaling,
you
can
see
here
we
can
base
it
on
a
CPU
target
and
and
so
on
and
so
forth.
So
you
can
see
that
all
of
these
all
of
these
options
are
available
to
you.
In
the
container
itself.
We've
got
the
image.
We've
got
the
name
for
the
container.
We
can
change
that
if
we
want
basic
settings
arguments:
environment
variables.
B
All
these
things
are
familiar
to
you
from
your
deployment,
yeah
Mel.
You
can
set
resource
limits
and
requests
again.
These
are
all
things
that
should
be
familiar
to
you
as
a
kubernetes
concept,
but
you
can
go
ahead
and
encode
them
into
spinnaker
so
that
you
know
what
they
what
they
should
be
and
they
are
not
sort
of
ad
hoc,
as
I
said
hey.
How
is
this
going
to
run
and
so
on?
I
didn't
really
pay
much
attention
to
what
I
was
putting
in
here.
B
So
I
have
no
idea
whether
this
lets
gotta
come
up,
wrap
early
or
not,
but
you
can
kind
of
see
what
happens
here
and
we
can
go
ahead
and
just
kind
of
monitor
this
from
the
from
the
task
bar.
If
we
want
to
do
all
right,
so
that
kind
of
gives
you
a
little
bit
of
a
feeling
for
that
and
pipelines.
We
can
go
ahead
and
we
can
create
new
pipelines.
B
B
But
don't
worry
about
that
right
now
they
come
over
here.
Okay,
so
you
can
see
the
different
kinds
of
staging,
so
you
can
create
a
pipeline
that
triggers
on
various
on
various
things.
Cron
jobs
changes
to
get
Jenkins
jobs,
Travis
jobs,
the
completion
of
another
pipeline,
perhaps
pub/sub
from
google
web
hooks
changes
to
the
docker
registry.
All
of
these
can
trigger
a
pipeline.
B
Okay,
let's
see
here.
So
let
me
just
kind
of
look
at
if
you
want
to
add
a
stage.
I
want
to
show
you
here,
so
what
you
can
do
is
you
can
have
stages
that
do
various
things
you
can
check
for
preconditions.
Did
a
particular
script
come
through
in
a
particular
way,
did
a
particular
pipeline
execute
and
so
on
and
so
forth.
You
can
deploy
you
can
deploy
in
two
different
ways.
This
shows
how
to
deploy
previously
baked
images.
B
B
From
the
interface
okay,
so
I'm,
apparently
running
behind
so
I'm,
going
to
jump
right
through
here
deployment
strategies
we'll
see
that
as
we
go
along
all
right
now,
so
there's
various
steps
here
for
deploying
on
on
kubernetes
you've
got
halyard
is
the
is
the
configuration
tool
you
use
it
to
configure
the
environment
and
spinnaker
and
then
to
go
ahead
and
deploy
it.
You
can
you've
got
three
different
choices:
local
debian
and
local.
Get
this
basically
means
check
it.
Local
git
means
basically
check
it
out
and
run
it.
B
B
Okay,
there
are
various
providers
that
you
can
use
here,
as
you
can
see,
I'm
not
going
to
go
through
all
of
them,
obviously,
because
the
only
ones
we're
interested
really
are
kubernetes,
but
just
know
that
they
exist
again.
Kubernetes
legacy
is
what
we
were
looking
at,
kubernetes
be
to
manifest.
Based,
you
can
store
your
you
have
to
have
a
persistent
storage.
B
B
So,
basically,
when
you
go
ahead
to
deploy,
let
me
just
kind
of
grab
this
here.
You
can
see
what
you've
got
configured
with
halyard
and
I'm
just
going
to
kind
of
show
this
here,
so
you
can
see.
You've
got
all
of
your
various
pieces
like
I
have,
in
this
case
a
kubernetes
account,
that's
just
using
my
current,
my
current
current
koop,
config
and
so
on,
and
then
you
can
just
do
how
deploy
apply.
B
Okay
so
and
then
that'll
go
through
and
run
and
how
it
actually
runs
is
going
to
be
based
on
what
your
provider
is
and
so
on
and
so
forth.
So
I'm
not
going
to
sit
there
and
watch
that
because
the
game
we're
running
a
little
bit
behind
but
okay.
So
let
me
just
kind
of
jump
back
here.
Okay,
there
are
various
versions
and
the
simplest
way
to
deploy
is
to
use
what
is
to
use
this,
the
the
quick
deployment.
This
is
going
to
be
obviously
the
most
kubernetes
friendly.
B
As
you
can
see
here,
coupe
cutoff
I'm
going
to
actually
get
out
of
this
right
now,
just
because
I
want
to
show
you
the
coupe
cuddle,
really
quick.
This
is
the
simplest
way
to
do
it.
You
do
not
have
to
go
through
a
million
steps
and
believe
me,
there
can
be
a
million
steps.
You
notice
here
it
just
went
ahead
and
it
did
everything.
Okay,
so
and
then
I
can
check
it.
I
can
get
pons
for
the
spinnaker,
namespace
and
I
can
see.
B
There
we
are
okay,
so
just
quickly
to
show
what
other
tools
can
be
integrated
with
it.
You've
got
Jenkins
and
Travis
4c
ICD
various
artifacts,
so
you
don't
have
to
define
specifically
the
binaries
or
the
code
that's
going
to
go
in,
but
you
can
define
where
to
get
it
from
okay,
so
Google,
Cloud,
storage,
obviously
github
or
just
directly
from
HTTP
for
monitoring
and
metrics,
which
becomes
obviously
very
important,
even
if
you're
not
using
canary
deployments,
but
obviously
also
for
canary
deployments,
Prometheus.
B
Obviously
stackdriver
our
data
dog
are
all
available.
You
would
configure
them
with
halyard
and
notifications.
What
we're
going
to
see
is
you
can
you
can
actually
set
manual
steps
so
that
something
is
not
going
to
go
live
until
somebody
manually
approves
it,
and
you
can
configure
spinnaker
to
send
notifications
for
that
to
slack
and
email
and
have
chat
or
via
a
text
message
security.
You
can.
Obviously
you
want
to
control
who
gets
in.
You
can
use
google
groups
github
team's,
LDAP
and
sam'l
from
that,
and
you
need
to
be
able
to.
A
B
Where
it's
done
and
where
to
find
the
artifacts
okay,
now
you
can
use
various
triggers
and
so
on.
Canary
wheat
on
canary
briefly
canary
deployments,
are
when
you
roll
out
to
a
small
bit
to
make
sure
that
it's
all
working
and
if
it
is
you
roll
out
two
more
so
you
can
configure
all
of
that.
There
are
various
providers
involved.
B
B
Here
we
go
here,
we
go,
we've
got
a
video
of
the
demo
just
so
we
can
kind
of
make
things
go
faster
if
necessary,
which
obviously
is
necessary
at
the
moment.
So
here
we
go.
So
if
you
notice
here,
we
have
these
pipelines
and
the
pipeline's
are
set
up
to
trigger
when
there's
a
change
to
to
the
source.
In
our
case,
we
use
Garrett,
but
you
could
also
use
I
mean
you
could
use
raw
Jenkins
get
whatever.
B
So
in
this
case,
where
we
have
an
application
that
holes
pulls
data
from
Twitter
and
makes
a
cloud
a
tag
cloud.
So
what
we've
done
just
now
is
all
we've
done.
Is
we've
changed
the
version
of
Python
to
Python
3
and,
as
you
will
see
in
a
minute
that
will
automatically
trigger
the
build
pipeline
that
builds
the
base
image.
B
B
B
B
So
let's
say
when
we:
if
we
look
at
the
security
scan,
you
can
actually
get
the
console
output,
which
shows
that
in
this
particular
image
we
have,
we
do
have
the
vulnerability
that
comes
that
comes
with
double
you
get,
so
you
can
see
that
particular
information
right
there
in
the
console
and
then
you
can
make
a
decision.
You
know,
do
I
want
to
move
forward.
This
is
using
Claire,
which
is
the
core
OS
open
source
API.
B
B
We're
going
to
go
ahead
and
change
the
code
for
the
application
itself
in
this
case
and
we're
just
going
to
make
a
change
to
the
actual
text
that
goes
on
the
page.
You
can
see
here
he's
putting
in
the
date
and
I'm
just
going
to
kind
of
move
forward,
so
he's
gonna
go
ahead
and
approve
that
change
and
that's
gonna
kick
off
a
Jenkins
job.
B
Okay,
so
you
can
see
that
that
so
we've
got
all
these
jobs
in
here
and
that
Jenkins
job
is
going
to
kick
off
a
different
pipeline,
which
is
in
a
different
instance
of
spinnaker.
In
this
case,
so
we're
gonna
go
east,
go
ahead
and
pull
that
up
and
a
second
there
we
goes.
There's
the
Twitter
app
and
there's
that
triggered
build
right
there.
B
Now
this
one's
a
little
bit
different,
because
you
could
see
it's
going
to
deploy
to
staging
it's
going
to
run
the
tests
on
staging
it's
going
to
scan
the
image
and
before
it
can
go
to
production,
it's
going
to
have
to
be
manually
approved
by
someone.
Now
this
is
kind
of.
What's
called.
This
is
what's
called
a
red-black
deployment
or
a
Bluegreen
deployment
where
we've
got
well.
I
shouldn't
say
that
this
is,
but
this
is
similar
to
that.
In
that
we've
got
one
set
of
we've
got
one
set
of
instances
that
are
running.
B
If
we
decide
we
like
it,
we're
just
going
to
switch
all
the
traffic
over
to
that
set
and
get
rid
of
the
other
one.
So
one
way
that
some
people
implement,
that
is,
they
have
two
identical
environments,
they
have
their
staging
environment
and
they
have
their
production
environment
and,
if
staging
goes
well,
they
just
shunt
all
their
live
traffic
over
staging
and
that
becomes
the
new
production
and
the
old
production
environment
becomes
the
new
staging
environment
and
so
on
and
so
forth.
B
So
this
here
this
here
shows
that
we've
got
the
request
for
whether
this
can
go
to
production
or
not.
It
included
a
link
to
the
live
version,
so
they
went
and
looked
at.
This
is
staging
yes,
it's
there
for
Python
three!
So
now
do
we
want
to
go
ahead
and
promote
that?
Yes,
we
do
we're
going
to
continue.
So
now
it's
going
to
go
ahead
and
deploy
it
to
production
and
shut
down
staging.
So
you
can
see
that
you
know
we
work.
We
already
have
the
security.
We
already
know
about
that
and
so
on.
B
All
right
so
and
it
shut
down
it
shut
down
the
original
staging.
So
there
we
all
right,
okay,
so
that
brings
us
to
the
end
of
the
demo.
I
know
that
we
kind
of
blitzkrieg
through
that,
but
we
still
have
a
few
minutes
left.
So
let
me
just
kind
of
get
back
to
my
slides
all
right,
so
you
can
see
what
we
did
here
is
we
have
the
application
services
team,
otherwise
known
as
the
operators.
B
They
created
all
these
golden
images,
they
put
it
in
the
registry
and
they
created
these
pipelines,
which
the
dev
team
can
then
use
to
deploy
to
staging
or
production.
So
that's
what
we
just
saw,
and
that
brings
us
to
the
end
again.
You
can
sign
up
for
debt
for
the
beta.
If
you
want
to
see
that
demo
in
a
way
that
is
not
spoken
so
quickly,
you
can
go
ahead
and
sign
up
for
the
next
one,
which
is
on
May,
16th
and
they'll
be
doing
it
live.
So
you
can
actually
ask
questions
at
that
time.
B
A
B
B
The
reference
section
typically
is
going
to
show
you
how
this
works
and
it's
going
to
give
you
a
gazillion
halyard
commands
for
backing
up
all
this
stuff.
I
didn't
have
a
chance
to
really
cover
that
too
much,
but
I
do
want
to
just
kind
of
give
you
one
and
then
I'm
just
going
to
jump
over
to
the
API.
B
But
if
you
look
here,
if
you're
adding,
you
can
go
ahead
and
add
kubernetes
as
a
provider
I
wanted
to
cover
this,
but
we
sort
of
ran
out
of
time,
but
you
can
add
it
as
a
provider
and
you
can
see
what
you're
adding
in
here
you're
adding
you
know
where
are:
where
are
the
secrets?
What's
the
context
and
so
on
a
lot.
B
B
B
A
B
B
Oh
IIIi
for
a
minute,
I,
wasn't
kind
of
clear
on
what
the
actual
question
was.
I
think,
what's
being
asked,
is
if
the
object
and
the
if
the
object
is
deleted,
you
know,
if
you
do,
you
know,
coop
cuddle,
delete
dash,
F,
whatever
will
spin
ik
or
pick
that
up
I,
don't
know
the
answer
to
that.
I
think
it
is
probably
no
but
I
would
have
to,
but
I
would
have
to
check
on
that.
A
B
B
B
Yeah,
so
what
you
do
to
version
things
is
you
can
define
an
artifact,
you
define
an
artifact
so
like
let's
say:
I
wanted
to
use
something
that
was
in
github,
so
I
would
define
the
github
artifact.
Then
you
can't
then
do
this
a
couple
of
ways.
First
off
you
can
trigger
a
pipeline
changes,
but
the
other
thing
that
you
can
do
is
you
can
create
a
pipeline.
That
specifically,
is
going
after
a
particular
container
image.
B
B
B
B
A
B
So
if
you
are
trying
to,
if
you
are
trying
to
orchestrate
kubernetes
on
OpenStack
you're,
actually
using
the
kubernetes
provider
and
the
cluster,
that
you
are
the
kubernetes
cluster,
that
you're
targeting
just
happens
to
be
on
OpenStack,
so
the
OpenStack
provider
itself,
I
am
NOT
a
hundred
percent
familiar
with
how,
with
what
it
provides
to
you,
I,
don't
know
that
it
provides
OpenStack
rather
directly.
I
can't
answer
that
definitively,
but
I,
don't
think
so.
So
I
could
be
wrong
if
I'm
wrong.
You
know,
please
somebody
pipe
up
and
say
so.
A
B
Absolutely
definitely
because,
in
fact,
that's
one
of
the
advantages
of
using
these
of
using
these
pipelines,
because
what
should
we
do
in
that
case
is
first
of
all
you
can.
You
can
put
in
a
precondition
on
the
pipeline
that
the
pipeline
itself
is
not
going
to
run
unless
specific
criteria
is
met
and
that
criteria
may
be
checking
to
see
whether
the
Prometheus
operator
has
been
deployed.
It
can
be,
although
well
yeah
I
mean
that's.
That's
that's
one
way
to
do
it.
B
You
could
include
the
deployment
as
part
of
the
operator
deployment
as
part
of
the
pipeline.
The
main
thing
is
making
sure
that
everything
making
sure
that
everybody
knows
that
they
need
to
go
through
this
process
rather
than
everybody
just
sort
of
you
know,
being
the
Wild
West
out
there.
It's
usually
people
there,
the
problem
and
not
the
technology
in
a
lot
of
cases,
but
it's
definitely
possible
to
do
that.
A
Perfect
and
how's,
the
question
is
about
the
clock
support.
So
what
is
the
feature
of
the
digital
ocean
support
and
what
are
the
requirements
of
what
provisioning
support
is
needed
to
add
new
cloud
providers
to
the
providers?
Spinnaker
support,
for
example,
add
node
support.
Would
you
need
strong
terraform
support,
kubernetes,
Cloud,
Controller
manager,
ansible.
B
Yeah
well,
if
the
question
is
about
kubernetes
support
on
digitalocean,
then
it
seems
to
me
that
that
would
just
be
the
kubernetes
provider.
If
the
question
is
non
kubernetes
digitalocean
support.
How
do
we?
Basically?
How
do
you
add
a
new
provider
for
spinnaker?
I
believe
that
that
is
actually
covered
on
the
on
the
spinnaker
website
itself.
I
think.
B
B
Would
I
would
definitely
go
to
the
spinnaker
website
here
we
go,
develop,
perhaps
use
stage
expansions
yeah,
there's
I,
think
the
place
to
go
to
ask
is
going
to
be
either
slack
or
stack
overflow.
There
is
an
active
slack
community
for
spinnaker.
Also,
another
good
place
that
you
can
go
is
community
dot
spinnaker.
Do
they
have
a
it's?
It's
a
pretty
active.
It's
a
pretty
active
community.
So
I
would
ask
that
question
there
and
see
what
they
have
to
say.
A
B
A
B
B
There
are
really
no
places
that
I
can
think
of
where
spinnaker
kind
of
falls
down
on
the
job.
In
terms
of
helping
you
work
with
kubernetes,
it's
a
little
bit,
it's
a
little
bit
convoluted
and
it
takes
a
little
getting
used
to.
But
then
you
know
cast
your
mind
back
so
did
kubernetes
when
we
first
got
started
so
I
I
am
NOT.
I
am
NOT
up
on
all
of
the
tools
that
are
out
there,
but
I
would
say.
Actually
the
spinnaker
probably
fills
the
bill.
In
that
case,.
A
B
B
B
A
B
B
B
A
Alright
great
suggestion
neke,
so
we
are
almost
out
of
time
thank
unique
antolia
team
for
using
webinars
and
demos.
I
thought
it
was
really
available
for
everybody
who
is
interested
in
integration
of
spinnaker,
who
is
kubernetes
and
about
CIT
D
processes
at
all.
Thank
you
for
joining
us
today.
We
I'd
like
to
also
remind
you
that
this
session
was
recorded
and
it
will
be
available
shortly
on
the
CNC
update
a
website
in
there
recorded
events
section,
and
we
are
going
to
have
our
next
webinar
next
week.