►
Description
Alejandro (Sasha) Vicente Grabovetsky & Nicola Paoli, AID:Tech
Deploying Hyperledger Fabric to production on Kubernetes is not a solved topic, AID:Tech present their work on designing and open-sourcing Helm Charts.
Rather than developing a monolithic Helm chart, AID:Tech created 4 separate and configurable charts for deploying Hyperledger Fabric components, namely the Certificate Authority, Peer, CouchDB and Orderer.
Sasha demonstrates how to deploy these charts together using Kubernetes on a cloud-based cluster, and connect them to Hyperledger Composer (using a chart under development).
NOTE: If you intend to follow the workshop and run the commands youselves, to ensure smooth sailing, we highly encourage you to:
1) Obtain a Cloud platform (AWS, GCP, Azure, IBM Cloud) account with some free credits, and create a basic Kuberentes cluster with 4 nodes
2) Obtain a cheap or free domain name (either with your cloud provider or elsewhere)
----
Hyperledger – Open Source Blockchain Technologies
Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use
Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory
Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter
Follow-us on Twitter @Hyperledger
Learn about Hyperledger Membership: https://www.hyperledger.org/about/join
#Hyperledger #HyperledgerGlobalForum2018 #Blockchain
B
B
A
Put
in
some
words
like
writing,
stuff
and
I,
think
maybe
you
can
strike
us
all
hello,
everyone,
I'm,
Nicola
and
we've
blockchain
developer.
The
attack
I
mostly
focus
on
developing
the
application
and
I've
been
using
hyper
laser
for
one
year
one
year
and
a
half
now
and
yeah
I
made
some
contribution
with
to
the
hydrologic
code
base
and
more
on
the
on
the
composite
one
on
my
side
and
fabric
and.
A
B
A
way
of
establishing
trust
in
a
shared
record
of
events
between
participants,
you
know
whether
this
is
distributed
and
funnily
enough
trustless,
just
briefly
touch
on
wood.
With
what
etiquette
attack
we
do
platform
for
our
doing
donations,
charitable
remittances,
other
types
of
value
transfer
to
enable
disempowered
and
disenfranchised
people
to
obtain
goods
and
services
and
weave
all
they
stain
about
developing
goals.
That's
what
the
things
that
guide
us
in
our
mission
and
if
you
want
to
find
out
a
bit
more
about
our
product,
we
can
talk
afterwards.
B
In
the
second
house,
we
will
have
actually
a
certificate
authority,
an
ordering
service
using
Kafka
consensus
and
several
peers,
and
that's
brilliant,
we'll
just
discuss
next
steps
and
we'll
have
time
for
questions
and
and
so
on.
So
so
this
is
what
the
actual
clusters
are
going
to
look
like.
Is
that.
B
B
Bandwidth
here,
because
we're
going
if
we
were
using
mini
cube,
we'd
have
alter
downloads
separately,
the
images
fabric
for
the
order
for
the
sieve
Kafka
for
zookeeper
for
the
couch
DP,
and
that
would
be
a
massive
strain
on
there
on
the
network
and
ice,
something
as
I
have
in
the
organization
here.
I
think
it
might
be
too
much
for
there
for
the
Wi-Fi
system,
and
the
last
reason
is
that
mini
cube
is
not
production.
B
Ideally,
we
want
to
be
able
to
deploy
kubernetes
in
production
and
might
as
well
start
with
a
real
faster
on
the
cloud
or
on-premises.
If
you
just
are
connected
to
one
of
your
own
premises,
we're
Nettie's
clusters,
at
least
you
to
have
several
notes
which
cloud
we've
been
asked
to
each
flower
goes
well.
Actually
you
should
not
carry
that.
That's
kind
of
the
point
of
kubernetes
cornetist
is
a
systems
to
unable
to
be
essentially
started.
Most
equipment
have
clusters
on
promises.
B
You
can
use
Amazon
Web
Services,
they
have
a
managed
to
who
are
natives
club
Google
platform.
Also
have
one
Microsoft
Azure
have
one
cloud
digital
ocean
and
so
on
and
supports
choose
the
one
that
you
that
gives
you
nice
credits
and
and
that
you
feel
comfortable
working
with
that's
that's
the
most
important
thing.
A
B
A
A
B
I
will
use
some
standards.
These
two
images,
which
is
clear
just
some
virtual
machines
with
seven
gives
around,
which
should
be
enough
for
the
capital
and
I'll
just
use
the
latest
velocities
version
that
is
available
I'm
sure.
But
if
you're
on
Google's
a
platform,
you
might
have
a
different
one
if
you're
on
it
always
you
might
have
different
one
and
five
know
so
that
we
have
enough
for
the
for
the
kafka.
A
B
Project
fabric
glue
to
it,
so
something
will
know
that
fabric
is
that
framework
under
the
hyper
ledger
project
topology
project
has
number
of
frameworks
like
so
tools,
Iroha
in
the
borrow
and
so
forth
and
hibachi
fabric
allows
you
to
build
permissions
local
networks.
You
can
look
at
these
websites
read
the
docs.
It
has
a.
B
I'll
touch
on
the
core
aspects
of
happenin
fabric.
In
there
you
will
have
three
types
of
nodes
that
will
be
deployed
in
production.
The
first
one
is
the
certificate
authority.
In
fact,
we
don't
necessarily
need
to
use
the
fabric
certificate
authority.
The
fabric
framework
is
quite
modular,
so
you
can
use
any
certificate
authority
that
you
want,
but
the
fabric
certificate
authority
spigot,
and
so
this
is
what
we'll
use
in
this
presentation.
B
B
Packages
transactions
and
puts
them
into
blocks
and
then
Center
disappears,
and
if
you
actually
have
a
Kafka
network,
establishes
some
consensus.
That
is
crush
fault
tolerance,
not
yet
presenting
full
tolerance,
but
quite
sure
that
the
fabric
is
actually
working
on
a
Byzantine
alternate
system
and
we
also
have
Alison
here
who
deployed
bft
is
smart
and
so
that
you
actually
can
also
use
this
system.
B
B
The
fabric
see,
as
I
mentioned,
to
manage
the
identities.
It
first
registers
the
identities
this
either.
You
can
register
that
in
the
name
and
the
password
or
it
actually,
you
can
just
threes
that
enter
name
and
get
the
password
back.
That
is
self
generated
by
the
see
this
ends
up
issuing
is
certificates,
and
then
you
can
renew
these
certificates
or
even
provokes
and
all
of
their
vocation
system
and
on
the
C
is
not
enforced
for
now,
but
it's
something
informal
setting
the
wording
says
services
we
mentioned.
B
It
consists
a
over
set
for
Turing
notes,
either
one
for
development,
with
the
solo
or
as
as
few
as
two
were
as
many
as
like
100
for
Kafka
cluster
and
the
cluster.
There
is
actually
a
walkthrough
on
how
to
bring
up
a
Kafka
cluster,
and
this
is
where
the
information
was
taken
to
actually
setup
cough
cough
cough
cough
cluster
in
kinetics.
B
If
you
were
doing
it
by
hand
is,
it
would
be
quite
a
lot
work
because
you'd
have
to
set
up
at
least
for
Kefka
nodes
and
least
three
zookeeper
knows,
but
Cornett
allows
you
to
do
that
in
single
command
and
enters
the
peers.
They
are
the
actual
node
store
and
maintain
the
block
chain
ledger.
They
do
it
typically
on
disk,
but
they
can
also
attach
a
couch
the
being
database
to
allow
you
to
hold
and
vary
their
actual
current
world
state,
not
entire
looking.
A
B
B
A
B
B
Group,
together,
different
grenades
components
into
templates,
so
these
actors
cookie
cutters
for
different
applications
that
you
might
want
it
to
install.
You
might
have
a
help
chart
for
for
a
Postgres
database
and
you
can
just
click
install
and
then
you
have
a
Postgres
database
running
on
your
cluster
without
having
to
deploy
separately.
Different
ones
like
deployment,
services,
ingress
and
so
on
and
helm
is
the
actual
tool,
together
with
a
server-side
tiller.
That
actually
allows
us
to
install
charts
in
the
future.
B
Basically,
it's
a
good
thing
of
your
net
intercept.
It
allows
us
to
prove
your
resources,
deploy
containers
scale,
the
number
of
containers
either
manually
or
automatically
self-heal.
So
when
we
talk
about
self-healing,
basically
one
of
your
containers
dies.
Then
girliest
notices,
all
this
container
die.
So
let's
actually
restart
it
on
different
machine
because
maybe
you're
one
of
your
servers
got
hit
by
a
metal
meteorite
and
you
need
to
kind
of
keep
keep
your
nice
with
application
life
and
also
allows
you
to
contain
connect
to
these
containers
either
within
the
cluster.
B
B
There's
charts
so
we
talked
about
how
their
bundles
templates
that
define
kubernetes
components,
a
cube.
Ops
house
actually
has
some
and
I'm
just
showing
off
some
of
the
charts
that
we
created
for
it's
a
hybrid
fabric
deployment.
So
there's
that
bitch'll
our
series,
a
CouchDB
and
healthier
in
there
is
the
H
left
order
and
the
H
left
peer
and
those
are
all
available
actually
they're
at
version
1
with
3.
B
Now
we
already
updated
them,
Emporium
we'll
be
updating
them
with
1.4
quite
soon,
once
we
test
them
and
yes,
it
allows
you
to
package
and
you
can
actually
it
allows
you
to
also
create
charts
as
well.
So
you
can
create
a
package
and
then
modify
it
to
fit.
The
needs
of
your
application
and
dealer
manage
the
server
side
stuff
want
actual
cluster.
A
B
A
A
A
B
B
A
B
Mentioned
that
the
basic
development
cluster
that
we're
going
to
work
on
now
is
going
to
just
have
sweet
components,
the
order
itself,
the
CouchDB
that
will
be
part
on
set
here
and
as
it
appears
itself,
which
will
be
bound
to
any
channel.
So
the
order
will
be
running
source
will
be
basically
no
consensus.
Just
the
faction
gloss
and
the
peer
will
be
just
maintaining
their
two
ledger
and
lecture,
and
the
channel.
B
B
So
this
will
give
you
the
latest
tools.
We
have
also
versions
for
1.2
and
1.1,
but
in
this
presentation
we'll
work
with
1.3
and
if
you're
on
Windows
or
Linux,
you
probably
will
want
to
just
use
this
command
with
curl.
This
is
the
official
standard
way
on
the
fabric
docks
to
get
these
very
tools.
It
will
basically
download
for
you,
they're
crypto,
gem
config,
the
exchange
and
a
few
other
tools
that
are
used
for
fabric
and
you
just
need
to
specify
the
version
which
is
1.3
are.
B
So
here
you
basically
in
the
crypto
config,
you
can
define
that
border
organizations,
for
example
the
order
and
give
it
a
domain
because
we're
working
with
the
kubernetes
dns.
We
have
this
essentially
internal
domain
name
service
and
the
cluster
vocal
is
essentially
your
cluster.
These
points
to
the
services
and
these
points
to
your
orders,
namespace
in
kubernetes
and
the
actual
host
name,
will
be
board
one
HL
effort,
because
this
will
be
the
name
filter,
and
this
is
the
knight
name
of
the
actual
charts
that
we'll
be
installing.
B
A
B
Binary
is
the
one
that
we
installed
just
now,
so
it
should
have
been
installed
with
if
you
are
using.
Yet
if
you,
if
you're
using,
not
Mac,
then
it
will,
it
will
have
downloaded
it.
So
you
may
need
to
point
to
it
directly,
so
you
may
need
to
actually
look
at
the
where
it
actually
the
load,
the
folder,
with
our
examples
and
so
on,
and
in
one
of
the
binary
folders
it
will
be
I.
A
B
It
should
be
in
there,
it
probably
will
be
in
there
more
examples
are
in
poverty:
okay,
perfect,
fantastic!
So
sorry,
Sam
husband!
Thank
you
very
much!
Okay!
So
once
you
have
that
you
can
actually
we're
going
to
actually
create
some
kubernetes
namespaces,
because
otherwise
some
of
the
stuff
will
not
work.
So
we
will.
A
B
B
A
B
Certs
started
with
M,
and
you
can
actually
see
that
it
actually
found
the
real
time
there
you
go,
so
you
actually
can.
Actually
the
next
thing
we're
going
to
do
is
we're
going
to
create
the
kubernetes
secrets.
A
great
secret
is
essentially
a
component,
a
obstruction.
It
relates
to
fault
data
that
your
different
ports.
Your
containers,
can
actually
access
in
that
particular
namespace.
B
B
B
A
A
B
A
B
It
we
can
just
quickly
go
through
the
actual
convict,
yet
that
we
have
in
there
repository
we
basically
defined
in
this,
because
it's
very
basic
example.
We
define
just
two
organizations
and
water
organization
which
has
famous
PD,
are
located
here
in
rip
to
configure
the
organization's
orders
service,
cluster
local
MSP,
and
then
we
also
defined
a
peer
organization
which
we
call
a
tech,
and
it's
located
here
in
crypto,
config,
peer
organizations,
peers
of
service
to
clusters,
all
MSP,
and
we
define
an
anchor
peer
for
these
peer
organization.
B
B
B
There
we
go
so
we
just
ruled
out
a
chance
to
walk.
Let
us
do
the
same
stuff
with
the
channel
transaction.
So
again
we
have
configured
this
channel.
We
set
the
profile
to
be
my
channel.
The
channel
ID
will
be
called
those
my
channel
and
then
we
output
and
create
the
channel
transaction
to
a
file
called
my
channels
and
opiates,
and
that's
if
the
channel
transaction
so
we're
in
business,
and
now
we
are
actually
going
to
steal
them
to
a
kubernetes
cluster.
B
A
B
Because
then,
you
can
actually
control
it
with
more
precision
who
actually
has
access
to
them
and
secretly.
You
cannot
is
essentially
immutable,
so
you
you
can
I
mean
you
can
delete
it
if
you
have
access
to
it,
but
you
cannot
change
your
changes
in
place
as
you
would
in
with
a
mounted
NFS
folder.
So
that's
why
we've
gone
for
that
option.
Basically,
so.
A
A
B
B
B
B
B
We
give
it
a
name
of
port
1
and
we
put
it
in
the
namespace
orders
and
we
specify
a
values
file
which
I'll
show
you
just
next,
just
after
we
run
this
command
so
to
understand
what
what
are
the
definitions
that
would
happen
and
I'll
also
show
you
the
chart
itself.
So
file
is
going
to
be
just
the
both
of
the
config,
so
it's
going
to
be
in
there
examples,
helm
values,
slash
board,
1,
dot,
yellow.
So
we
run
that
and.
A
B
So
it
reference
that
file
just
now
right
and
the
file
that
we
referenced.
It
was
this
board
Wanda
Taylor.
Here
we
have
an
image
tag,
so
we
are
using
fabric
version
1.3.
So
you
can
change
that
if,
in
principle,
eventually,
if
you
want
to
use
another
version
like
1.2
1.1,
but
for
the
purpose
of
this
project
we'll
stay
without,
we
also
justify
some
persistence.
So
we
specify
that
we'll
have
an
access
mode
of
readwrite
once
and
the
size
of
4
gigabytes.
This
is
this
is
fine
for
the
development
department.
B
You
don't
need
a
lot
of
space,
but
in
production
you
definitely
want
a
lot
more
than
just
one
gigabyte.
Otherwise
your
network
is
going
to
run
out
of
space,
pretty
pretty
darn
fast.
We
also
specify
that
we're
using
a
solo
order
right
and
we
also
specified
a
membership
service
provider
ID,
which
is
order
MSP.
B
We
also
before
in
the
previous
section,
we
created
the
certificate
authority
certificate
as
a
secret
rule,
and
we
created
also
the
genesis
secret
from
Genesis
global,
as
well
as
the
admin
certificate
for
the
order.
So
all
of
these
are
referenced
here
and
they'll
be
used
by
the
chart.
It
automatically
set
up
the
border
correctly
so
that
you
don't
have
to
copy
and
paste,
like
particular
files
into
your
order
container,
which
would
be
a
nightmare
okay.
B
So
let's
check
her
now.
That's
the
next
thing
that
I
want
to
check
is
the
charts
so
we're
going
to
go
to
our
chart.
Suppose
three
and
I
can
actually
show
you
what
the
order
chart
looks
like
so
that
you
have
a
little
bit
of
an
idea
of
what
is
like
and
I'm
actually
going
to
make
a
bit
of
you
assume,
because
I
think
it's
it's
quite
difficult
to
see
from
down
there.
A
A
B
The
first
thing
that
you
can
find
at
the
chart-
he
is
the
charted
camel
you
see
that
it
has
some
kind
of
net
information.
This
is
an
order,
share
tips
if
they
were
created
by
us
and
their
currency,
not
13,
stated
high
precision
let
your
project,
although
we
hopefully
eventually,
will
make
them
official.
They
have
some
keywords:
some
application
version.
In
this
case
we
support
fabric
1.3,
and
some-
and
these
are
the
maintenance-
does
myself
in
the
colon,
and
maybe
the
other
thing
that
you
want
to
take
care
and
look
at
is
the
values
file.
B
So
this
is
the
default
values
file
the
values
file
that
we
are
if
we
used
in
the
installation,
is
the
one
that
we
use
to
override
specific
values
in
this
default
values.
One,
and
here
we
actually
specify
by
persistence.
We
specify
whether
we
use
TLS
weather,
which
secrets
we
use
resources,
most
lection
toleration
and
if
integers
and
all
the
kind
of
kubernetes
specific
things
will
bit
by
bit,
Gordon
but
I
don't
want
to
now
with
too
much
information
and
usually-
and
there
is
a
helpful-
hopefully
helpful,
readme.
B
So
this
is
the
read
mean
that
explains
how
the
chart
works
and
all
the
different
options
that
are
available
in
it
and
what
they
stand
for
a
maternity
photos
and
now
to
the
most
important
part
of
the
chart,
is
the
actual
templates.
So
this
the
charts
to
use
go
templates
to
enable
you
to
create
essentially
manifests
that
are
specific
to
your
use
case.
So
you
might
have
a
config
map,
for
example,
and
this
config
map
has
a
lot
of
the
order.
B
B
We
also
have
an
ingress
which
allows
you,
if
you
want
to
actually
access
the
order
from
how
the
site,
the
kubernetes
cluster
via
a
domain
name
a
persistent
bullying
plane
which
operationalizes
the
storage.
So
you
actually
define
what
kind
of
storage
you
want
and
how
large
the
storage
should
be,
and
that
is
a
good
upgrade.
This
manifest
and
the
service
which
allows
you
to
communicate
with
the
boat
in
the
container
inside
the
cluster.
A
B
By
running
cube,
CTL
get
votes,
namespace
orders,
we
specify
a
label,
so
we
want
an
application
to
be
Rachel
effort,
which
is
the
name.
A
charm
and
they're
released
to
be
board,
one,
which
is
the
name
of
this
particular
deployment,
always
particularly,
and
we
are
using
JSON
path.
We
are
plude's
in
the
name.
So
basically
you
end
up
with
something
like
this.
Echo
or
supporter
is.
A
B
B
We're
going
to
do
five
steps,
we're
going
to
stall
the
couch
the
beam,
how
important
we're
going
to
stay
here.
Id
from
that
we
created
from
crypto
jump
on
to
prefer
natives.
We're
then
going
to
install
the
pure
thumb
chart
in
a
very
similar
way
as
we
install
their
order
in
your
chart,
will
create
the
channel
and
then
fetch
enjoying
it,
just
as
we
would
in
a
kind
of
in
local
fabric
network
on
your
computer.
B
Again,
we
run
the
helm,
install
command,
use
the
stable
repository
HLF
CouchDB,
we'll
call
it
c
DB
dash,
p
r1
and
specify
that
it
will
be
in
the
namespace
appears
and
again
we'll
specify
a
specific
values
file
to
override
the
default
values
of
this
helm.
Chart
and
I'll
show
you
again
what
these
so
again.
You
can
see
that
it
has
a
deployment,
a
pod,
a
secret
and
persistent
volume
claim
and
the
service,
and
this.
B
B
So
they
CouchDB
chart
looks
like
this
again.
You
have
the
charts
with
this
information
there's
the
couch
DB
instance
has
some
keywords
and
so
on.
Here's
the
values
file
the
values
file
specifies
that
we
again,
we
only
want
one
replica
and
we
want
to
use
the
repository
hyper
ledger
fabric
couchdb
and
by
default.
It
has
a
tag
in
this
case
of
0.49,
but
we
can
specify
a
different
tag
and
then
get
a
different
image.
B
A
B
They
ingress
if
we
want
to
access
the
cow
should
be
from
outside.
So,
for
example,
one
might
want
to
connect
to
it
with
with
some
kind
of
graphical
interface,
so
that
you
can
see
it.
They
persist
in
volume
claim
which
holds.
There
is
a
volume,
and
importantly
the
secret.
So
here
is
where
we
actually
Auto
create
a
password.
If
we
don't
provide
one,
so
you
can
actually
auto
generate
a
random
alphanumeric,
24
lengths,
password
or.
A
A
B
B
B
We
can
connect
to
that
couch
D,
so
you
don't
need
to
manually
specify
all
the
cars
to
be
credentials
when
you
deploy
the
peer.
So
that
makes
it
a
little
bit
easier
to
the
point.
We
specify
also
the
MSP
ID,
which
is
Pierre
MSP
oops.
Again
we
specify
some
secrets
like
the
certificate
of
the
peer.
The
key
of
the
peer.
The
certificate
they
see
is
certificate
for
the
pier
and
the
channel
that
we
created
before
the
period
minute
admin
certificate
and
the
period
min
ki.
B
B
Again,
the
key
things
are
they're.
Basically,
the
values.
You
basically
use
the
hi
pleasure
fabric
pier
to
install
it.
We
specify
the
service
with
a
porch
and
event
requests
they
an
ingress.
If
you
want
to
expose
it
to
the
Internet
and
some
persistence,
as
well
as
the
type
of
database,
whether
it's
gullible
Dubuque
or
cards,
to
beam
as.
B
A
B
B
Now
that
we
are,
we
have
a
peer
that
is
running.
We
can
actually
create
a
channel,
so
we
can
actually
run
the
cube
CTL
exact
command
here
to
execute
a
particular
command
inside
our
container.
We
specify
the
name,
space,
peers
and
the
peer
pod
that
we
are
talking
to.
We
run
peer
channel
quit.
We
specify
the
order.
We
want
to
talk
to
the
name
of
our
channel
and
the
location
of
the
channel
file
transaction
and
we
received
block.
B
B
Q,
CTL,
exec
and
Pierre
our
Pierre
put
again,
and
then
we
run
this
command.
We
need
to
run
actually
bash
see,
and
the
reason
for
this
is
that
we
want
to
actually
specify
an
environmental
variable
that
exists
on
the
actual
container
peer
container.
So
we
do
not
want
this
to
autocomplete
to
our
local
ever
mental
variable.
That's
why
we
specify
this
bio
see
and
then
we
run
pure
channel
joint.
A
A
A
B
The
secrets,
and
now
you
have
a
clean
slate
for
the
next
half
of
the
workshop
full
disclosure.
We
don't
actually
run
these
commands
by
hand
at
8
that
case
we
actually
have
an
internal
tool
that
actually
orchestrates
that
for
us,
and
we
are
actually
going
to
be
releasing
that
later
on,
probably
in
in
a
few
weeks
or
perhaps
months
so,
but
it's
still,
we
feel
it's
still
useful
to
actually
show
all
the
commands
that
go
into
that.
B
So
if
you
understand
first
the
ones
the
complexity
of
their
of
the
deployment,
but
also
so
that,
if
something
goes
wrong
with
the
automatic
tool,
you
can
actually
understand
what
is
going
on
wrong
and
you
can
actually
debug
it.
So
this
is
why
we're
kind
of
going
over
all
these
things,
which
of
which
there
are
quite
lot
and
there's
a
little
petition.
So
please
bear
with
us.
B
Well,
there's
a
lot
of
different
ham
charts
for
the
epilator
foundation.
There
is
also
a
chart
for
borrow
for
sure
and
I
think
the
guys
at
sawtooth
I
think
they
also
have
some
ham
charts,
but
they
I
don't
think
they
release
them
to
the
main
camp.
Shout
repository
I'm,
not
sure
if
they've
open
source
them
or
not.
A
B
The
official
helm,
charts
repository
yes,
but
at
one
point,
iBM
has
a
kind
of
macro
chart
to
deploy
everything
on
their
own
repository,
but
as
far
as
I'm
aware,
it's
not
be
maintained
for
about
six
months.
The
last
time
I
looked
so
I,
don't
know
what's
the
status
of
it
now,
but
we
are
the
ones
who
maintain
these
separate
comp,
charts
and
personally
I
prefer
to
maintain
separation
charts
for
each
component,
because
then
you
can
compose
these
components
together.
B
A
A
B
B
B
You
can
deploy
a
chain
code
in
much
the
same
way
as
you
deploy
in
a
local
machine.
You
can
at
the
moment
we
still
haven't.
What
we
want
to
do
is
to
create
a
CLI
chart
so
that
we
can
actually
easily
connect
and
deploy
a
chain
code
for
now
we're
doing
it
away
in
a
slightly
more
arduous,
more
less
manual.
Processor
Nicola
can
tell
you
a
little
bit
more
about
it.
A
B
B
Next
thing
that
we're
going
to
work
on
is
there
and
the
actual
production
setup
right.
So
the
main
difference
here
is
that
we're
going
to
use
a
certificate
authority
to
actually
create
all
the
crypto
material
instead
of
using
crypto
gem.
So
this
will
allow
us
to
actually
generated
more
less
online
again.
B
B
So
here
we
can
see
that
there
they're
being
created
and
the
things
that
you
won't
run
is
you
here
have
some
notes
here.
So
it
says
the
nginx
ingress
controller
has
been
installed.
It
may
take
a
few
minutes
for
load-balanced
right
key
to
vll
available,
so
you
can
watch
the
status
by
running
QC,
TL,
namespace,
ingress
controller
and
get
services
or
white
nginx
ingress
controller.
So
let's
run
that
command
just
copy
pasted
and
we
get
an
external
IP
in.
B
A
A
B
B
B
We
basically
just
install
helm,
install
with
a
stable
version
of
nginx
ingress,
hey
I,
put
it
in
a
namespace
called
nginx
ingress,
but
some
people
put
it
in
in
cube
system,
which
is
also
completely
fine,
so
I
put
actually
I
put
in
the
next
phase
called
ingress
controller,
and
you
can
give
it
a
name
like
nginx
ingress,
for
example,
and
once
you
have
it
installed,
you
basically
can
get
basically
run
cube.
Ctl
namespace,
whatever
names
rescue
is
told,
you
ingress
a
controller.
B
For
example,
ingress
controller
get
services
wide,
and
this
W
W
W
it
means
watch
so
it
will.
It
will
continue
watching
until
when
there's
changes
you
case,
you
might
see.
Initially,
the
external
IP
are
spending,
so
it
might
take
a
while
until
you
actually
get
the
IP
address
for
the
external
IP
that
you
can
use
with
your
domain
name
registrar,
to
specify
that
you
want
start
domain
name
to
point
to
you,
the
IP
of
your
cluster.
Basically.
B
We
can
also
then
install
the
certificate
manager,
so
the
certificate
manager
actually
is
tool.
There
used
to
be
a
sinkhole
on
kubernetes
called
cube.
Lego
and
the
cert
manager
essentially
allows
you
to
obtain
TLS
certificates
for
free
automatically,
and
it
will
renew
them
automatically
as
as
they
expire.
So
you
don't
need
to
do
that
by
hand
basically,
and
we
just
again
install
the
helmets
tools,
tables,
slash,
search
manager,
also
the
getting
in
the
namespace
cert
manager.
But
again,
if
you
want,
you
can
install
it
in
different
name.
Space
such
as
cubes.
B
B
We'll
need
to
create
some
clusters,
so
the
cluster
issuer
will
allow
you
to
to
deploy
TLS
keys
that
are
there
for
staging
or
for
production
using
let's
encrypt,
so
we
can
essentially
install
this
one
create
one.
Mike
is
already
installed
and
the
second
one
for
production.
We
can
also
create
it
and
again,
in
my
case,
it's
already
installed.