►
From YouTube: Providing Containerized Cinder Services to Baremetal Kubernetes Clusters [I] - John Griffith, NetApp
Description
Providing Containerized Cinder Services to Baremetal Kubernetes Clusters [I] - John Griffith, NetApp & Huamin Chen, Red Hat
Kubernetes deployments running on OpenStack clusters require a full OpenStack: Keystone, Nova, and Cinder services.
This talk presents a more pervasive and simplified deployment architecture by integrating Containerized standalone Cinder services with baremetal Kubernetes. Cinder services offer many storage features that are still missing in Kubernetes. Cinder is supported by many storage vendors, with over 70 storage drivers in its repository. A Containerized standalone Cinder service makes these features and extensive storage products available to Kubernetes cluster.
Key to this architecture is a Kubernetes volume provisioner that provisions Cinder volumes and transparently converts Cinder volumes to Kubernetes supported storage types, such as iSCSI, Fibre Channel, NFS, or Ceph RBD.
Based on the these technologies, the new architecture enables enterprise customers to deploy Container services in a dedicated cluster and consume advanced storage features.
About Huamin Chen
A passionate system software developer, Huamin Chen contributes to open source projects spanning from A to Z: Apache BigTop, Ceph, fio, Gluster, Kubernetes, Tachyon, and ZFS. Huamin Chen is currently employed by Red Hat. Follow him at http://github.com/rootfs
About John Griffith
John Griffith, Principal Software Engineer at SolidFire now a part of NetApp, helped to create the Cinder project in OpenStack. Primary responsibilities are upstream contributions to cloud related open source technologies. Currently active in Kubernetes Storage SIG, CNCF CSI project and various other communities.
John has over fifteen years of engineering experience in both hardware and software engineering. He’s been an active user and contributor to open source for over a decade. In addition to his technical contributions, John also spends a lot of his time talking to people who are interested in learning about various Cloud technologies as well as gathering feedback from current users.
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.
A
Cool
alright
I
think
it's
about
time,
so
late,
Friday
last
day
of
the
conference,
so
thanks
for
being
here
instead
of
napping
or
going
to
the
airport
or
getting
out
of
here,
so
we're
gonna
go
through
some
things,
and
what
we
want
to
talk
about
today
is
using
cinder
as
a
storage
block
storage
provider
for
kubernetes,
so
go
through
and
do
some
introductions
here
and
get
rolling
first
off.
My
name
is
John
Griffith
I've
been
working
on
OpenStack.
The
cinder
project.
A
I
helped
start
that
about
six
years
ago,
when
it
first
came
to
be
a
thing,
so
I've
been
working
on
that
I've
been
doing
some
things
in
the
container
space
as
well
and
all
kinds
of
anything
cloudy.
Basically,
these
days
I
kind
of
work
on
a
little
bit
and
have
various
things
all
my
contact
info
is
up
here.
If
you
ever
want
to
reach
out,
ask
me
any
questions.
Anything
follow
up
on
today.
Give
me
a
holler
I
love
to
talk
so,
and
sometimes
that's
the
only
thing
I'm
really
good
at
so.
My.
B
B
A
It's
been
a
while
so
we're
gonna
kind
of
walk
through
and
for
those
of
you
who
may
not
be
familiar
with
cinder.
I'm
gonna
talk
a
little
bit
about
that
and
go
through
some
of
that
and
then
Juan's
gonna
talk
about
what
the
volume
and
the
persistence
model
in
kubernetes
looks
like
today,
and
then
we're
gonna
kind
of
tag-team.
What
we've
put
together
to
use
cinder
to
solve
some
of
those
problems
and
then
talk
about
where
we're
gonna
go.
A
What
the
future
looks
like
and
then
hopefully
get
some
dialogue
going
with
you
folks,
maybe
get
some
input
kind
of
go
from
there,
so
first
off
in
terms
of
cinder
and
what
it
is.
So
the
idea
of
cinder
way
back
when
was
to
create
a
block
storage
as
a
service
right.
So
the
whole
idea
was
to
give
an
abstraction
to
give
something
that
could
be
used
by
a
cloud,
particularly
OpenStack.
That
would
let
you
plug
in
various
backends
and
seamlessly
consume
them
as
a
user.
The
whole
idea
is,
you
can
scale
this
out.
A
You
can
scale
this
in.
You
can
dynamically
change
things
behind
the
scenes
without
your
users
ever
knowing
right.
So
that
was
the
whole
idea.
One
of
the
other
things
that
some
people
aren't
really
aware
of
is
we
did
have
all
along
this
idea
of
having
it
be
a
standalone
service.
So,
if
you're
familiar
with
OpenStack,
you
may
know
of
a
project
called
Swift,
they
had
sort
of
the
same
model,
it's
an
object
store.
A
It
can
be
used
in
OpenStack
or
outside
of
OpenStack,
alright,
and
so
that's
kind
of
the
same
thing
we
want
to
do
with
cinder.
We
want
to
make
it
something
that
is
useful
for
OpenStack,
and
that
may
be
its
primary
use
case,
but
it's
extremely
valuable,
as
you
can
see
these
days,
because
there's
at
least
eight
or
nine
that
I
count
right
now,
projects
out
there
that
are
trying
to
abstract
storage
devices
into
an
interface
right.
So
we've
already
got
one.
A
It's
been
out
there
for
for
almost
seven
years,
it's
got
over
70
or
80
backends
that
are
supported
and
it's
been
used
in
production
by
really
really
large
clouds
for
a
long
time.
So
it's
battle-tested.
It's
got
a
ton
of
features,
we've
had
thousands
and
thousands
above
bug
fixes
over
the
years,
so
it's
solid
its
rock-solid.
So
it's
a
really
good
choice
and
best
of
all
we've
got
a
great
community
with
with
over
500
contributors,
and
that's
that's
just
code
contributors.
So
it's
a
it's
a
pretty
huge
deal.
A
You
know,
as
I
mentioned,
it's
not
something
that
it
is
just
for
OpenStack.
So
I've
been
doing
this
crazy
demo
thing
on
for
a
couple
of
years
now
on
using
cinder
as
a
standalone
thing
in
docker,
as
well
as
kubernetes
and
swarm
and
everything
else,
and
for
anybody
that
seen
my
demos,
you
probably
know
why
I'm
not
going
to
do
a
live
demo
today,
so
I
used
to
have
good
odds
and
right
now,
these
days
I'm
it
like
I
got
like
a
intention.
So
I'm
gonna,
just
don't
do
him
anymore.
A
That's
okay,
I
put
the
stuff
on
my
blog
and
you
can
check
it
out
there,
but
the
cool
thing
is
one
of
the
things
that
we
did
recently
is
we
went
ahead
and
we
put
stuff
in
the
cinder
repository
in
the
in
the
master
tree
to
actually
containerized
cinder
and
run
it
as
a
service
using
a
docker
compose
file.
So
for
all
those
people
out
there
that
you
hear
you
know,
OpenStack
projects
are
hard
to
deploy
and
they're
difficult
to
upgrade
and
the
whole
idea
behind
this
was
just
to
say:
hey.
A
You
know
that
a
lot
of
that
is
not
true.
So
inside
of
the
contributor
rectory
inside
of
the
cinder
repo
there's
a
compose
file,
you
can
go
in
and
you
can
do
a
make
to
build
the
images,
the
docker
images
and
then
you
can
do
a
compose
up
and
actually
have
a
full-blown
running
cinder
service
on
your
system.
Now
I
say
you
know:
oh,
that's
it
you're
done
right!
Okay,
it's
not
really
that
simple!
A
A
So
these
are
the
steps
you
clone
the
repo
you
CD
into
the
contribute
directory.
You
run
a
make.
What
the
make
does
is
it
uses
a
project
called
low
key
that
builds
all
of
the
images
that
you
need
so
you're
gonna
get
a
cinder,
API
container
or
cinder
volume,
container
controller
container,
a
database
container
and
a
rabbit
and
queue
container.
It's
gonna
build
those
up
and
you're
gonna
modify
your
compose
file.
A
If
you
want
to
run
compose
up
and
that's
it
you're
ready
to
go
so
if
you
want
to
modify
it
and
use
a
back-end
that
you
have,
maybe
you
have
a
set
back
end
or
an
edit
back-end
or
a
hatachi
back
in
some
other
back-end,
that's
supported
in
cinder
that
you
want
to
plug
into
it.
You
can
just
modify
the
comp
file
just
like
you
normally
do,
and
you
now
have
support
with
with
the
standalone
thing
now.
A
B
A
B
Two
types:
the
ones
that
works
for
hypervisors
and
the
ones
that
works
on
palmetto
the
ones
work
for
hypervisors.
What
you
have
is,
what
am
I
d
like
a
a
Tobias
EBS,
you
get
a
volume
ID
and
on
cinder
you
get
another
voting
ID,
it's
an
abstraction
of
the
reference
in
a
database.
That's
only
the
crowd
and
hypervisor
knows,
but,
on
the
other
hand,
you're
working
with
power
metal.
You
got
all
kinds
of
PV
types
that
gives
you
detailed
informations.
How
do
how
to
connect
a
host
to
a
back-end
storage?
B
You
have
the
exports
and
from
that
points
you
can
make
a
month
if
you
go
through
the
ice
caste
and
Augustus
instructors,
you
know
the
end
points
and
you
know
the
past
where
the
volumes
can
be
attached,
and
that
is
something
we
really
want
to
have
from
cinder
and
how
these
magical
works
is
from
the
abstraction,
the
water
mighty
to
the
endpoints,
that's
associated
with
each
of
the
basic
previous
and
that's
the
thing
we
want
to
have
to
make
that
happen
in
this
project.
And
why
do
you
want
to
make
that
happen?
B
We
won't
take
advantage
of
all
the
features
that
cinder
already
provide
to
us
and
that's
our
not
really
present
to
kubernetes.
Yet
if
you
have
one
for
volumes,
backends
and
that
have
other
features
already
works
pretty
well
on
cinder,
but
you
don't
have
these
features
working
on
kubernetes,
yet
there
you
can
just
leverage
cinder
to
make
all
this
happen
to
you,
one
for
storage
and
so
far
we
have
a
lot
of
things
in
kubernetes
the
dynamic
dynamic
provisioning,
the
attach
and
detach
mainly
in
dice.
B
You
can
do
a
third-party
chantey
touch
and
we
also
recently
added
the
warren,
expand
so
called
resize
feature,
and
that
just
happens
in
the
1.9
and
a
snapshot.
Massive
works
off
out
of
tree,
but
almost
something
already
it
works.
Unlimited
storage
back
ends
in
there
happens
to
be
one
of
those
and
thanks
for
what
about
contributors.
B
Before
we
go
into
details,
let's
look
at
how
the
forums
cinder
volume
works
for
in
kubernetes,
so
you
have
the
coconut
master
master
controller
master.
A
bunch
of
controllers
running
over
there
and
things
are
usually
interest-
is
the
provisioners
and
attach
and
detach
controllers.
So,
if
you
create
some
volumes,
you
call
the
cinder
sdk
provision
creates
essentially
equivalent
to
what's
in
the
creates
commands
gives
to
you
and
you
get.
B
B
So
this
tomorrow
we
are
used
to
and
if
you
run
a
crew
practice
and
all
the
Chiefs
on
stack,
these
heist
works
now
we're
going
to
change
the
the
floor,
little
bits
and
just
to
make
sure
the
Batman
hoes
are
the
most
host
machines
in
cast
a
detailed
connecting
info.
So
the
host
can
make
direct
connections
without
OS,
brick
and
with
also
Nova
the
house
works,
and
the
picture
looked
pretty
much
the
same.
We
don't
add
a
lot
of
stuff
except
stairs
improve
us.
B
Magic
office
happens
over
there
before
we
go
to
details
and
how
here
is
the
workflow,
so
we
still
have
their
stock
by
manhole,
kubernetes
cluster
and
nothing
changed,
and
we
still
use
the
cinder
volume
and
average
team
the
stock
cinder
and
it
could
be
a
containerized
or
it
could
be
a
standing
OpenStack
syrup
and
we
still
use
the
existing.
The
storage
types
asked
us
if
I
general,
an
RPD,
nothing
changes
of
they're
at
a
higher
level.
We
own.
The
only
thing
we
are
to
the
picture
is
pretty
much
standard.
B
We
have
a
external
or
cinder
Warren
provisioner,
that's
that's
our
host
Rick
force
and
the
workflow
is,
will
request
force
PVC,
obviously
how
to
define
storage
crust
and
once
the
source
cross
defined
and
your
PVC
use
used
a
storage
across
the
instance
to
our
provisional
and
a
provisional
hard
to
cinder
use
the
synthetic
API
and
the
crystal
volume
just
still
use
the
same.
The
API
and
then
cast
the
connection
info
and
mysterious
is
also
it's
in
the
API,
so
not
intrusive,
pretty
much
standard,
and
from
that
point
we
get
a
not
with
you
not
return.
B
The
cinder
volume
ID,
but
we
returned
underlying
connection
info
is
up.
If
that
is
our
Scotty,
we
will
converse
the
cinder
volume
to
ice
category
if
that's
a
pity,
because
RBD
volume,
so
these
happens
over
there
from
the
users
and
points
and
from
the
goodnights
perspective,
pretty
much
nothing
changed
and
you
don't
need
a
hypervisor
anymore.
A
A
So
we're
gonna
do
here
is
we're
gonna
kind
of
walk
through
the
process,
a
little
bit
and
kind
of
click,
these
slides,
so
I
said
we
raise
questions
and
stuff
at
the
end.
Also,
if,
as
we're
going
through
these,
if
you
have
questions,
we
got
plenty
of
times
so
raise
your
hand
shout
out
and
we'll
kind
of
dig
into
it
right.
So
it's
actually
really
simple.
So
we've
got
this
external
provisioner,
it's
sitting
out
there
and
it's
listening.
A
A
So
let
me
take
that
back
now.
Everybody
copied
us
alright.
So
the
first
thing
it
comes
in
the
provisioner.
It
sees
a
request
for
a
PVC
capacity,
100
gig
access
mode,
read/write
storage
classes,
sender
and
a
volume
type.
So
one
of
the
things
you
can
do
is
so
cinder
has
the
ability
to
specify
volume
types
and
that
can
give
you
things
like
replication
quality
of
service
compression
features.
Deduplication
features
all
that
stuff.
A
A
The
next
thing
it
will
happen
is
the
provisioner
will
go
and
just
call
out
to
cinder.
Just
like
anything
else
had
we
use
the
cinder
api
it'll,
do
a
cinder,
create
request,
give
it
all
the
parameters
that
you
fed
in
and
everything
that's
needed.
Do
that
and
you
don't
come
back
and
it'll
just
give
the
cinder
volume
ID.
So
now
the
provisioner
will
have
a
cinder
volume
ID,
so
it's
got
a
handle.
Who
knows
what
to
do
and
it
associates
it
with
the
volume
handle
that
you
put
in
your
pod
file.
A
So
that's
the
that's
the
dynamic
provisioning
piece,
which
is
fairly
simple.
The
next
part
is
the
attachment
attachments,
pretty
simple:
you
wanna,
okay,
alright,
so
we
do
kind
of
the
same
thing.
The
provisioner
gets
the
request
to
do
an
attach.
So
what
happens
is
it
goes
ahead
and
it
calls
out
to
sender
and
it
does
what
cinder
expects,
which
is
called
a
reserve
which
just
says
hey
mark
this
volume
is
something
that's
that's,
flagged
is
going
to
be
used
and
attached.
A
It'll
then
go
and
it
will
call
an
initialize
connection,
and
this
is
actually
the
important
call.
So
this
call
is
what
includes
all
of
the
things
like
for
I
scuzzy,
for
example,
it
includes
the
initiator
iqn.
The
host
IP
address
all
that
sort
of
thing
right.
So
that's
where
all
the
data
actually
is
and
then
it
responds
I
can
see
you
over
here
on
the
left.
A
It
goes
ahead
and
in
response
back
to
the
provisioner
and
back
to
kubernetes
with
all
that,
and
then
we
call
the
kubernetes
I
scuzzy
mounter
or
RVD
mounter
all
right.
So
then
we
call
that
and
we
go
ahead
and
we
attach
that
volume
to
the
bare
metal
node
right.
So
now
you
have
a
cinder
volume
attached
to
bare
metal
connections.
All
done.
We
notify
cinder
that
we're
finished,
which
then
in
turn
calls
an
attachment
completion
which,
on
the
cinder
side
marks
the
volume
is
in
use.
A
Now
the
thing
that's
cool
about
this
is
as
kubernetes
progresses
as
the
kubernetes
api
is
for
storage
progress
and
things
like
that.
There's
a
lot
of
things
that
you
know.
People
want
to
do
right
and
they
and
they
can't
do
yet
because
it's
not
in
the
API.
But
the
thing
that's
cool
is
now
you
have
this
other
piece.
You
have
this
centerpiece,
so
you
can
go
in
from
here
and
you
can
do
things
like
cloning
and
replication
and
all
of
those
things
that
may
or
may
not
be
in
the
communities
API
anytime
soon,
yeah
Louis,.
A
B
A
B
Being
happy
contributed
by
community
helpers
is
works
pretty
well
on
a
number
of
proteins
and
I
see
an
RPD.
These
are
first
because
I
Scotty,
obviously
guessed
the
password
option
so
far
back
to
risk
question
the
QEMU
and
the
coup
count
and
that
could
potentially
could
be
the
breeder
for
things
like
Craster
and
an
FS
that
does
not
have
the
native
proactive
I
supports.
It
was
thin,
the
essentially
provisions
of
rock
device
and
the
block
device
on
our
face
or
carrasco
it
just
a
file
emulation.
So
the
TCM
URL
runner.
A
A
B
Cinders
rich
set
of
features
and
things
that
happens
happen
to
crop
not
yet
or
haven't
happened
to
the
drivers,
yes
and
if
it
works
for
cinder
out
of
the
box
for
your
back-end.
If
you
have
one
for
ask
us
if
I
can,
this
is
already
working
in
standard,
but
you
don't
really
have
this
natural
control
of
working
on
the
backend
just
bein
cinder
for
that.
If
you
have
because.
A
B
Now
they
just
take
for
another
example.
The
volumes
resized
in
kubernetes
does
not
work
on
individual
drivers.
Yes,
but
it
works
what's
in
there.
If
you
need
this
feature-
and
you
don't
have
that
one-
oh
hi,
Scotty,
proteins
use
similar
and
replication
doesn't
happen
to
kubernetes,
yet
you
still
don't
already
have
it.
So
that's
another
place
right.
B
Si
si
I
know
that's
our
topic
and
where
we
can
go
from
here
and
just
before
this
presentation
we
just
merged
in
the
si
si
driver
for
the
convenience.
It
doesn't
have
all
these
cool
features,
but
wouldn't
be
a
good
starting
point
and
from
that
point
from
there
on,
we
have
have
a
enablement
for
lots
of
drivers
with
us
happen
to
have
individual
drivers
sitting
underneath
and
so
for
those
who
grow
I
know.
There's
already
are
so
many
si
si
presentations.
Yes,
we
haven't
come
to
those
presentations.
B
A
B
A
A
So,
and
it
gives
you
a
little
bit
more
in
terms
of
being
able
to
support
it
and
make
sure
everything
works
and
stuff
like
that
and
then
best
of
all
what
it
means
is
you
can
swap
devices
in
and
out
from
from
behind
this
without
your
users
ever
knowing
without
your
applications
ever
knowing
and
not
impact
your
workflow
right.
So
that's
where
the
advantage
is
granted.
It's
not
for
everybody.
A
You
know
there's
some
people
that
that
definitely
are
going
to
want
to
go
and
write
their
own
CSI
driver
or
whatever
plug-in
stuff
like
that,
and
that's
great.
This
is
just
another
option
for
a
lot
of
people.
That
I
think
has
a
lot
of
benefit,
especially
when
it
comes
to
the
support
and
testing
model
right,
because
this
has
been
tested
pretty
thoroughly.
So
I
think
I.
Think
with
that.
B
A
B
A
B
A
A
So
the
question
was:
have
we
done
anything
with
AZ
awareness
and
cinder?
So
if
you're
familiar
with
cinder,
it
has
availability
zone
type
model?
We
haven't
really
put
any
of
that
into
this
provision
or
yet
because,
quite
frankly,
right
now,
I
don't
know.
Yeah
kunai's
doesn't
have
any
way
to
to
interpret
that
right
now
there
might
be
something
interesting.
You
could
do
in
your
pod
file
to
try
and
distribute
your
your
storage
a
little
bit
that
there
might
be
something
you
do,
but
that's
only
half
of
it
right
so
but
again,
pull
requests.
A
A
It's
not
there
right
now,
but
it
certainly
can
be
so
you
and
I
are
gonna
talk
here
in
a
few
minutes
and
we're
gonna
look
at
doing
that,
because
that
would
be
pretty
cool
right
because
you
could
say:
hey,
run
the
script
to
to
launch
a
cinder
service
inside
of
my
kubernetes
deployment
and
then
use
that
and
consume
it
from
my
kubernetes
deployment.
So
it'd
be
pretty
cool.