►
From YouTube: Nov 24, 2022 - Ortelius Architecture Meeting
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
B
Okay,
so
it's
artillius
Thanksgiving
architecture
meeting
and
we're
going
to
go
back
to
Sasha
who's
talking
about
his
ortelius
in
the
Box.
C
Yeah,
so
all
the
containers
come
up
really
nicely
now
and
I
see
they're
all
node
ports
and
my
English
controller
deploys
perfectly
and
it's
using
my
local
IP
address
of
my
machine
as
the
external
IP.
So
it's
it's
functioning
as
a
load.
Balancer
I
see
you
created
a
Ingress
which
is
cool
and
it's
listening
on
I
think
it's
like
a
wild
card
right
just
listens
on
any
for
anything.
C
B
B
So
one
of
the
things
I
had
to
do
is
so
the
because
we
still
have
the
monolith
the
so
this
is
the
monolith
repo
that
we're
looking
at
in
visual
studio
and
in
there
there's
the
the
back
end
Tomcat
server
that
serves
up
the
Java
code.
You
know
like
the
jsps
and.
D
B
Runs
a
bunch
of
the
the
monolithic
Java
queries,
so
this
guy
connects
directly
to
the
database.
The
UI
is
supposed
to
be
more
of
the
UI
front,
end
stuff.
So
all
the
JavaScript
HTML
pages
and
stuff
like
that
it
was
a.
It
was
an
attempt
to
kind
of
separate
the
front
end
from
the
back
end
and
then
the
nginx
part
is
where
we
have
the
the
the
service.
B
So
the
service
is
pretty
simple,
yeah
right
now,
the
yeah
there
it
goes
so
the
service
basically
is
going
to
come
up
on
80
and
if
we
look
at
the
deployment
we're
having
the
images
come
up
on
80..
So
what
ends
up
happening
is
the
Tomcat
server
will
come
up
on
Port,
80
and
we'll
kind
of
do
stuff
on
that
front
there.
B
So
actually,
if
we
look
at
the
deployment
for
the
service
here,
so
this
is
one
of
the
the
80,
the
Tomcat
service
and
so
that's
coming
up
on
80
and
then,
like
I
said,
the
nginx
is
coming
up
on
on
80.
and
the
engine
x
container
that
we
use
is
doing
what's
called
a
a
reverse
proxy.
C
B
Was
this
section
here
to
do
an
Ingress
and
the
missing
part
that
I
was
I.
B
And
this
is,
if
you
do
a
if
we
do
a.
B
Yeah,
so
what
you
end
up
doing
is:
what
is
it
image
class
name.
B
There
it
is
so
the
Ingress
class.
What
ends
up
happening
is
this
I
think
is
installed
by
default
when
you
build
when
the
cluster
is
built
originally
and.
B
B
And
so
that
was
one
thing:
I
had
a
change
on
that
and
then
also
I
had
to
add
the
host
back
in
and
the
reason
why
I
had
to
add
the
host
back
in
is
because
what
I
deployed
to
our
Azure
cluster
I
couldn't
use
a
wild
card
asterisk,
because
we
have
the
docs
in
the
website
are
also
ingresses
and
one
of
the
weird
things
that
if
you
look
at
the.
B
So
the
weird
thing
when
I
deployed
over
to
our
Azure
cluster,
this
is
the
Ingress
that
we're
just
looking
at
in
the
yaml
file.
And
then
these
are
the
two
other
ingresses
and
they
all
came
up
on
the
same.
B
A
A
B
I
actually
had
to
do
a
a
host
name.
There.
C
B
C
B
The
other
thing
that
we
have
in
the
Ingress
on
the
other
side
is
this
top
part.
The
ALB
is
the
Amazon
load,
balancer,
so
I've
deployed
over
to
to
eks.
If.
B
A
B
And
they
wanted
to
use
the
Amazon
load
balancer.
So
we
needed
to
add
in
a
bunch
of
annotations.
C
B
How
to
connect
to
the
AWS
load
balancer
now,
what
we
should
do
is
add
in
gke
and
whatever
other
ones.
We
need
to
do.
A
A
B
Exactly
and
that's
where
you
set
up
your
SSL
routing
and
redirects,
and
things
like
that
so
like
on
this
line
here,
the
load
balance
will
automatically
redirect
all
traffic
that's
going
to
80
over
to.
A
B
So
basically,
the
way
we
with
ortilius
we
do
everything
all
the
certificates
are
at
the
the
load
balancer
level
and
not
at
we
don't
do
SSL
traffic
between
all
the
pods.
B
We
could
do
it,
it's
just
a
headache
with
all
the
certificates
and
stuff
like
that
and
for
what
we
have
it
just
wasn't
worth
the
effort
to
try
to
get
certificates
in
place
and
I've
tried
to
use
like
let's
encrypt
and
things
like
that,
and
it
just
Falls
over
it.
Just
doesn't
it.
B
C
So
in
turning
the
cluster,
the
pods
can
talk
to
each
other
sort
of
there
right.
There's
no
encryption
here
so
I
know,
and
people
like
all
go.
Oh
security
but
I'm
like
if
someone's
gotten
that
far
into
your
network,
I,
think
you've
got
more
problems
in
your
security.
B
Yeah
exactly
so
one
of
the
things
that
we
did
when
you
started
working
on
this
project
was.
B
You
have
you
just
forced
me
to
clean
some
stuff
up
and
the
main
thing
was
the
we
use
a
JWT
token
to
so
once
you
log
in
the
login
returns
a
JWT
token
that
has
your
user
ID
in
it
and
a
date.
B
Basically
you
you
uuid,
so
we
know
that
it's
basically
like
a
unique
session
ID.
So
in
that
JWT
token
encrypted
in
that
is
your
user
ID
and
the
uuid.
And
what
that
allows
us
to
do
is
the
after
the
login
is
finished.
We
save
that
as
a
cookie
and
then
every
transaction
going
back
to
the
server
pushes
that
cookie
across
now.
If
you
go
to
one
of
the
microservices,
we
want
to
make
sure
that
the
microservices
is
validating,
that
you're,
a
real
user
and
just
not
somebody
hack
in
the
system.
B
So
that's
where
that
validate
user
comes
in
at
the
beginning
of
all
the
microservices,
that
validate
user
pulls
the
cookie
and
then
decrypts.
The
the
token
gets
your
user
ID
and
your
you.
The
uuid
goes
and
looks
up
your
basically
in
the
database
to
see
if
you've
been
authenticated
before
and
then,
if
you've
been
authenticated.
B
It
just
updates
the
time
stamp
to
kind
of
like
a
keep
alive
time
stamp
and
it
returns
a
successor
failure
at
that
point
and
then,
if
you,
if
it,
if
you
fail
user
validation,
then
you
just
get
kicked
back.
If
you
are
successful,
then
you
actually
get
to
go
into
the
microservice
transaction
at
that
level.
So
that's
how
the
things
are
kind
of
pulling
together,
so
we
actually
have
the
front
end.
C
B
B
You
know
service
two
Etc,
so
the
that's
kind
of
how
the
networking
comes
into
play
and,
like
I,
said
the
where
to
go.
This
class
that
we're
looking
at
the
nginx
class
gave
us
that
default
initial
low
balancer
route,
so
I
think
when
you-
and
this
is
what
you
should
be
seeing
on
your
side-
is
that
we
have
the
the
Ingress
should
give
you
a
public
IP
address
that,
then
you
could
use
your
little
DNS
tool
to
map
over
to
that
that
IP
address
I've.
C
Actually
found
a
tool,
that's
even
easy
and
there's
no
setup,
oh
cool,
it's
called
local
dot
GD
and
it's
a
guy
that
set
up
a
DNS
in
netley
file
that
always
points
to
your
127.
It's
very
smart
right.
C
So
the
only
thing
a
developer
would
have
to
do
is
add
the
name
create
their
own
I'll
put
it
in
the
readme,
but
they
just
have
to
create
host
names
in
there
in
their
host
file.
Using
that
domain
and
it'll
always
resolve
to
there
and
using
make
make
search,
you
can
make
your
own
SSL
search
and
you
can
test
out
your
application
perfect.
B
B
Because
this
this
level
here
doesn't
have
the
default,
Ingress
doesn't
have
SSL
enabled,
but.
A
B
We
use
like
one
of
the
cloud
providers,
s
ingresses.
Then
we
get
the
whole
SSL
level
yeah.
So
if
we
need
to
do
get
fancier,
we
may
need
to
add
some
more
annotations
or
something
to
the
default
Ingress.
C
B
The
the
networking
part
pieces
get
a
little
weird,
whether
you're
using
mini
Cube
versus
kind
versus
stuff,
like
that,
but
I
think
if
we
can
figure
out
kind,
a
kind
seems
to
be
a
lighter
weight
and
more
popular
than
minicube
I.
C
Like
kind
because
it
resembles
the
actual
kubernetes
cluster
that
you'd
encounter
in
the
cloud
environments,
so
it's
a
like
for
like
environment
and
that's
something
that
developers
or
students
would
be
able
to
learn
very
quickly
right,
they'd
be
able
to
get
the
feel
of
kubernetes
like
it
would
be
in
the
cloud
yep.
So
that's
why
I
chose
kind
two,
and
also
because,
like
you
said
it's
much,
lighter
I
mean
you
can't
just
do
something
like
like
on
what
was
that
other
one?
C
You
just
said:
no
mini
Cube
mini
Cube,
enable
Ingress
and
it'll
just
do
it
for
you
now
you
actually
have
to
understand
it
right
yeah.
So
it's
a
bit
more!
It's
a
bit.
It's
maybe
a
little
bit
more
learning,
but
you
won't
have
to
do
it
if
it's
all
automated,
but
it's
just
it's
like
encountering
the
real
thing
right,
which
is
what
you
want
really
now.
B
On
the
on
this
part,
the
nginx
reverse
proxy
you'll
see
here,
I
went
into
another
folder
that
has
our
entry
point
nginx,
and
this
is
basically
all
the
configuration
files
for
this
reverse
proxy.
Now
you'll
see
that
there's
like
three
or
four
of
them,
four
different
files
and
based
on
the
values
that
you
pass
in
to
your
nginx
you'll,
see
things
in
there
like
SSL,
off
different
options.
B
They're
doing
is
setting
environment
variables
to
let
us
know
what
configuration
files
we
need
to
kind
of
copy
over
to
start
working
with,
because
you
know
you
could
run,
we
can
pass
in
certificates
to
the
SSL
I
mean
to
the
nginx
and
have
we
you
can
you
can
skip
this
in
Ingress
technically
and
you
could
go
from
the
front
end
right
to
the
reverse
proxy
and
that's
the
way
I
used
to
do
it
before
you
started
playing
with
kind
and
we
recognize
it
in
Ingress.
B
But
what
ends
up
happening
is
the
way
the
reverse
proxy
works
is.
This
is
where
we
set
up
just
things
like
how
to
handle
the
transactions.
B
A
B
Are
actually
the
name
of
the
well
right
here?
Is
the
internal
DNS
name
for
that
service?
B
So
this
one
is
bringing
up
the
MS
general
service
on
the
default
namespace
on
Port
8080.,
and
we
associate
that
to
this.
This
Ms
General
back
end
and
you'll
see
here's
the
other,
like
the
validate
user
microservice,
this
one's
a
little
interesting,
because
the
same
service
understands
how
to
it's
the
same
URL
path
that
needs
to
handle
a
post
and
to
get
so
we
had
to
do
some
special
stuff
there.
B
So
you'll
see
all
this
stuff
happens
at
that
level,
and
then
we
have
our
listening
pieces
listening
on
80,
and
this
is
like
a
if
we're
going
to
use
keys.
This
assumes
that
we're
going
to
have
an
SSL.
Let
me
look
at
this.
One
I'll
be
a
little
more
cleaner,
so
this
is
the
one
that
we're
currently
using
in
that
we
just
deployed
is
the
back
end
services
and
we're
just
listing
on
Port.
80.
B
you'll
see
this
statement
here,
the
location-
and
this
is
where
you
actually
do
all
of
the
the
routing.
So
when
we
get
in
this
URL
path,
so
this
is
one
of
the
location
files.
When
you
get
this
URL
path
coming
in
you're,
going
to
capture
that
and
you're
going
to
Route
it
over
to
the
specific
Upstream.
B
B
So
there's
a
bunch
of
them
for
every
single
transaction.
There's
a
location
file
and.
C
C
B
You'll
see
that
there's
a
script
out
here
that
basically,
what
I
did
was
I,
took
them
and
created
a
little
Pearl
script
that
read
this
and
spit
out
the
all
these
files.
For
me,
I
didn't
feel.
B
And
then,
like
here's
like
there's
the
home
page,
this
is
going
to
a
different.
This
is
going
to
the
front
end
the
UI,
microservice
or
monolithic
piece,
and
here
you
can
see
that
we
added
handle
multiple
paths.
So
if
you
wanted
to
load
up
the
CSS,
the
JavaScript,
all
those
URLs,
the
images
we
have
to
capture
and
and
make
sure
that
we
have
the
the
routing
correctly.
B
So
that's
how
this
is
all
kind
of
plumbed
together.
So
we
have
the
the
front
end,
which
would
be
you
know,
like
you
said,
either
the
cloud
provider
or
you
know
the
default
kubernetes
load,
balancer.
A
B
Just
to
Ingress,
it
just
depends
on
how
we
get
that
configured.
B
Could
do
so
much
stuff
with
it,
like
you
know
like
here
we're
overriding
headers
and
things
like
that.
You
know
you
get
to
do
a
bunch
of
fun
things
with
with
the
internet's
configuration.
C
B
So,
at
the
end
of
the
day,
like
you
said-
and
this
is
one
of
the
things
that
we
had
a
had-
a
change
was
originally
the
engine.
X
was
coming
up
as
a
load
balancer,
and
that
was
because
we
didn't
have
the
Ingress
and
then,
when
we.
A
D
C
C
B
Yeah,
it
was
difficult
to
put
it
that
way.
So
this
is
the
the
Ingress
and
you'll
see
that
when
this
host
name
comes
in
so
it's
gonna
match
that
on
that
hostname
and
then
it's
going
to
take
the
anything
that
starts
with
the
slash,
which
is
basically
everything.
It's.
C
B
I,
don't
see
it
showing
the
the
class
that
it
was
that
you're
talking
I
know
what
you're
talking.
B
Why
it's
not
showing
it
maybe
just
just
the
way
it
when
you
do
the
Subscribe
it
just
doesn't
give
us
enough
details
yeah.
C
Because
I
also
got
confused
was
there's
two
there's
nginx
for
kubernetes
by
nginx
themselves:
the
F5,
the
company,
then
there's
an
nginx
increase
controller
which
is
maintained
by
the
kubernetes
community
yeah.
So
you
can
get
really
confused
and
I
read
and
what
I
chose
was
in
the
end.
I
went
for
the
one
by
the
community
because
I
just
felt
like
that
one
was
the
closest
to
Cuban
edit,
because
it's
communities
community.
So
that's
the
one
I
chose
yeah.
B
Now
the
weird
thing
I
can't
figure
out.
So
this
is
the
public
IP
address
and
then
you
have
your
your
routing
and
this
is
going
to
go
to
the
back
end.
But
the
back
end
is
coming
up
on
1.72
and.
D
B
It
I
don't
know
I
I
haven't
had
a
chance
to
figure
out.
Like
you
said,
the
the
routing
inside
and
the
networking
side
of
kubernetes
is
can
be
a
little
strange.
Yeah.
C
It
must
say
it's
something
that
I
really
need
to
what's
taking
me,
I
need
to
understand
it
better.
That's
why
I'm
glad
I'm
doing
this
but
yeah
there's
some
confusing
things
in
here
which
I
don't
like.
Does
this
thing
work
yeah.
B
C
C
C
A
B
B
That's
what
it's
that's,
what
it's
routing
to
is
to
the
the
node
itself.
B
B
So
this
this
Ingress,
this
IP
address,
is
still
one
of
the
actual
nodes.
It
looks
like.
B
Like
you
said,
the
internal
routing
can
be
confusing
yeah.
C
B
So
we're
gonna
one
quick
question:
this
is
gonna
totally
jump
topics
on
the
xrpl
there's
like
I,
think
two
or
three
issues
that
a
couple
people
have
taken
up,
but
they
haven't
actually
done
any
work
on
them.
B
It
doesn't
look
like
does
it
make
sense
for
me
to
just
place
a
comment
in
those
issues
to
see
if
they're
still
working
on
them
and
try
to
get
a
response
because
they're
kind
of
holding
us
up
or
do
you
or
should
we
just
kind
of
take
them
over
and
just
not
give
them
the
option?
What
are
your
thoughts.
A
E
I
think
we
can
create
more
stories,
and
since
we
want
a
few
of
these
stories
to
be
completed
soon
right,
you
can
assign
it
to
people
who
are
active.
B
Yeah,
okay,
so
yeah.
If
it's
it's
like
it
has
to
do
with
like
the
transform
on
the
s-bomb
data,
with
the
with
what's
happening
in
the
ipfs.
You
know
the
nft
storage
and
there's
another
one
with
loading
the
license
data
into.
E
A
B
All
right
so
I'll
go
ahead
and
create
some
more
stories.
B
There
is
a
gentleman
from
Ukraine
I
think
goes
by
Nick
Nik
is
his
handle
I
think
on
Discord,
but
he
does
some
UI
design
so
I'm
going
to
create
a
couple
issues
out
there
around
UI
design
the
see
what
he
can
come
up
with
ideas
for
kind
of
reworking,
some
of
the
the
tables
and
searching
capabilities
that
we're
going
to
need
to
use
kind
of
be
able
to
map
out
the
what
a
you
know
like
a
list
of
components
and
I'll
list
their
applications
and
then
the
other
part
is
a
Time
series
piece
and
I
know
ookar.
B
She
did
some
work
on
the
time
series
at
application,
Level
and
I.
Think
I
want
to
build
off
of
that.
Remember
with,
like
the
3D
chart
thing.
E
Yeah,
that
would
work,
but
if
we
can
get
some
more,
you
know
materialistic
design
and
that
would
make
more.
B
Yeah,
so
so
that's
what
I'm
gonna
get
some
issues
out
there
work
on
those
and
try
to
flush
that
out.
So
everybody
has
a
good
idea,
a
good
good
idea,
the
direction
that
we're
headed
with
the
UI
and
make
sure
that
the
data's
gonna
be
easy
to
navigate
because
I
have
a
feeling.
B
We're
gonna
have
a
lot
of
data,
and
that's
always
a
challenge
when
you
got
a
web
browser
front
end
is
making
sure
that
you
just
don't
sit
there
and
get
the
kill
page
message,
because
it's
taking
so
long.
A
B
All
right:
well,
let
me
let
me
update
those
issues
and
create
some
new
ones.
Does
anybody
else
have
anything
that
I
may
be
missing
for
today.
A
D
C
C
I
can
give
you
guys
two
tools
if
you
get
to
a
client
and
they
already
have
a
cloud
infrastructure
which
they've
been
deploying.
Okay,
you
take
these
two
tools.
You
can
use
one
of
they're,
basically,
the
same
thing:
I,
don't
know
which
one
I
just
found
them.
What
they
do.
Is
they
because
you
it's
very
easy
to
set
up
and
what
they
will
do
is
if
the
people
aren't
using
IAC.
Yet
these
two
tools
will
actually
convert
the
entire
Cloud
infrastructure
into
terraform
code,
for
you,
it'll
do
70
of
the
work
for
you.
Oh.
B
C
Yeah,
so
if
you
want
to,
if
you've
got
a
client
where
they've
got
Cloud,
but
they
haven't
been
deploying
it
in
a
infrastructures
code
in
a
model
yeah
and
obviously
it
makes
sense
to
use
infrastructure
as
code
because
of
all
the
benefits
of
you,
know,
tracking
changes
and
config
drift
and
all
that
just
take
this
tool.
It's
open
source,
I'll,
put
them
in
the
channel
for
you,
guys
now
and
it'll,
literally
suck
in
the
entire
Cloud
infrastructure.
C
A
B
So
one
of
the
things
I
did
to
get
everything
kind
of
hooked
back
together
is
I.
Remove
the
Google
Cloud
build
and
moved
everything
over
to
GitHub
actions
and
you'll,
see
that
I
reworked
the.
B
So
this
is
the
monolith
repo
and
what
it
does
is
it'll.
Let
me
get
to
the
right
branch.
C
B
So
I
did
a
build
push
chart.
Basically
what
I've
changed
everything
to
do
is
we'll
go
through
and
does
our
our
login
to
Quay?
We
do
our
our
build,
our
Docker
build.
We
generate
the
s-bomb
and
then
we
go
and
take
the
information
and
we
push
it
into
the
charts
and
one
of
the
things
that,
like
it
says,
like
chart
version,
chart
version
actually
comes
from
the
GitHub
run
number
in
the
image
version
comes,
you
know,
so
we
calculate
those
at
the
beginning.
B
So
what
ends
up
happening
in
the
model
is
everything
ends
up
having
the
same
chart
version
and
image
tag,
but
they
go
to
their
different
repos.
One
of
the
things
I
try
to
do
is
paralyze.
You
know
make
this
run
in
parallel,
but
the
this
chart
and
releaser
got
confused
when
things
were
being
updated
at
different
times
in
with
commits,
and
things
like
that.
B
So
but
basically,
what
ends
up
happening
is
we
go
through
and
build
everything
create
our
new
chart
and
then
we
go
off
and
Trigger
another
workflow,
that's
in
another
repo
and
that
workflow
then
runs
a
node.js
at
Ben
from
done
Australia
put
together.
B
So
what
we'll
do
is
the
the
JavaScript
will
take
all
the
repos
and
go
find
the
latest
version
of
the
chart
and
it
will
go
through
and
generate
a
new
chart
for
us
and
you'll
see
here
we
have
the
charts
and
I
check
them
in
so
every
time
we
know
which
what's
happening.
So
this
is
the
latest
chart
so
49.
So
if
you're
looking
Sasha
to
find
like
the
latest
one
just
go
over
this,
this
repo
and
open
up
the
chart
and
you'll
it'll
have
the
latest
version
number
in
it.
B
For
you
and
you'll
see
that
it
went
out
and
found
like
the
debt
package
microservices
at
version.
Two,
where,
like
the
scorecard,
is
at
version
43,
so
each
one
of
the
microservices
when
they
build
they
go
off,
create
their
local
chart
and
then
they
call
that
trigger
to
run
the
the
trigger
back
on
this
main
one.
B
So,
at
the
end
of
the
day,
what
we
have
is
our
GitHub
pages
and
on
the
GitHub
Pages
we'll
have
our
all
of
our
releases
that
are
out
there,
because
one
of
the
things
that
the
home
chart
releaser
does
is,
when
you
have
a
the
parent
child
type
of
chart,
it'll
suck
them
all
in
into
one
big
tar
file.
B
So
this
contains
not
only
the
parent
chart,
but
all
the
sub
charts
are
are
pulled
together
into
this
release,
so
it
makes
it
handy
for
us
to
be
able
to
have
the
different
versions
and
then
the
index.yamo,
which
is
what
the
artifact
Hub
uses,
is
updated.
So
you'll
see
here
just
another
rendering
of
the
parent
child
charts
and
asking
me
version
49.
B
But
then
we
go
into
what
was
in
version
48
for
example,
so
we
kind
of
have
a
running
history
of
all
the
changes
as
at
the
chart
level,
so
Sasha
it
just
it
just
simplifies
everything
on
your
side.
B
Be
able
to
to
just
hit
the
one
chart
place,
and
then
this
is
mirrored
over
on
artifact
hub.
C
B
C
B
B
C
B
So
that's
an
extra
trick
if
you
want
the
charts
locally.
Otherwise
you
can
just
refer
to.
You
can
do
the
do
the
deployments
without
having
the
charts,
locally
yeah.
C
That's
true,
I
could
actually
just
just
in
the
beginning
when
I
was
doing
testing
and
changing
values
and
things
I
I
always
add
them
locally,
but
for
the
Dave,
although
Tillis
in
a
box,
you
won't
need
to
I,
don't
think
you'll
need
to
have
them
locally.
Actually,
yeah.
B
The
only
values
that
we
have
are
the
the
database
password
and.
B
A
B
Go
find
it
so
you
could
do
something
like
pre
processing
to
go
grab
the
the
password
if
you
wanted
to,
and
then
this
is
where
the
database
is
running.
B
So
this
one
is
we're
setting
the
the
the
the
hostname
where
it
exists
and
it's
running
it's
running
in
a
the
the
service
name
is
postgres
post
postgres
SQL
in
the
postgres
namespace,
and
that
gives
us
that
internal
connection
automatically-
and
this
came
up.
This
worked
really
nice
because
postgres
is
not,
it
isn't
exposed
to
the
public.
It's
only
exposed
to
the
the
nodes
in
the
cluster.
B
B
It
yeah
postgres
when
Brad
brought
up
postgres
on
the
Azure
cluster.
It
was
running
in
its
own
namespace
and
I
just
reused.
It.
C
B
It
was
just
the
helm
chart,
so
he
ran
the
bitnami
postgres
home
chart.
C
Okay,
because
I
thought
it
would
be
nice
to
have
that
particular
service
running
within
the
other
components
right
they're,
all
in
the
same
cluster.
In
the
same
namespace,
sorry
yeah,
okay,
I
need
to
go
and
change
that
then
I'll
just
change
it.
The
terraform
can
do
that.
That's.
B
Yeah,
basically,
if
you
want
to
do
two
two
pair
for
two
like
deployments,
one
for
postgres,
get
that
up
and
running
and
then
do
a
second
one
for
ortelius.
That
may
be
the
easier
way
to
go,
and
the
reason
why
I'm
thinking
of
that
is,
if
somebody
is
going
to
use
like
a
managed
postgres
like
on
AWS
or
something
that's
external
to
kubernetes,
then
you
can
just
point
them
to
the
the
terraform
for
artillius.
C
C
B
Yeah,
if
you're,
if,
if
your
service
is
running
in
the
in
the
same
name,
space.
D
B
C
B
Doesn't
matter
so
here's
the
the
postgres.
B
And
this
has,
this
is
I,
don't
know
what
the
default
name
space
is,
but
this
is
where
you
can
set
like.
If
you
want
to
know
it
has
tons
of
parameters,
but
I
think
you
can
get
away,
which
is
doing
something
real.
Just
take
the
defaults
and
see
what
happens
yeah.
C
So
what
I've
been
doing
with
the
terraform
of
when
developers
or
people
start
using
it
is
that
I've
been
putting
I've
been
using,
obviously
comments
to
say,
hey.
This
is
otilius
and
this
is
where
it
comes
from,
and
this
is
the
backstage
one,
and
this
is
the
captain
one
and
you
know
with
the
links
in
the
so
the
documentation
is
kind
of.
What's
in
the
code,
yep.
B
B
So
yeah
this
is
I
I
think
this
may
be
the
one
that
Brad
used.
It
could
have
been
this
one
or
there's
a
I
just
saw
it.
A
B
Yeah,
so
this
is
where
it's
the
we.
If
you
ran
this
home
chart
and
you
set
the
password
back
over
on
our
side-
we'd
have
to
set
the
password
here
as
well.
So
this
is
going
to
be
the
same
password
between.
B
B
Sasha,
to
make
things
easier
is
we
could
do
a
default
in
the
values
file.
B
Postgres.Ortelius.Svc.Cluster.Local,
so
we
we
can
assume
that
everybody's
going
to
come
up
in
the
artillius
namespace
with
with
postgres
included,
and
then
then
you
just
skip
this
whole
parameter
all
together.
B
C
I
mean
I
also
don't
want
to
create
extra
work.
I
just
want
to
make
it
as
simple
for
both
sides
of
the
you
know
the
production
and
the
div
yeah
the
I
mean
the
values
is
easy
right.
I
mean
it's
just
a.
C
B
B
C
The
captain
life
cycle
toolkit
doesn't
have
a
Helm
chart.
You
have
to
apply
it
from
the
yamls
okay
that
but
that's
easy
with
terraform
2,
because
I
just
run
Cube
CDL
inside
the
terraform.
To
do
that.
C
Actually,
really
close
now
to
releasing
this
thing.
Actually,
I
just
want
to
be
able
to
understand,
hitting
the
otilious
front
page
right,
making
sure
that
works
yeah,
because
the
most
important
thing
is
having
artelius
running
and
then
obviously,
once
that's
running,
we
can
start
working
on
the
pieces
that
would
gosh
and
everyone
else
needs
for
well,
I
guess
it's
just
good
question
yourself,
like
Nifty
stuff,
getting
a
ranga
up
and
running.
B
Yes,
we
will
need
that
a
Ringo
in
there
and
I
I'm
I'm
guessing
our
brain
go
would
be
a
similar
type
of
setup
where
where's
a
rainbow.
How
do
we
connect
Durango.
D
C
They've
got
a
they've
got
a
cube,
City
a
kubernetes,
it's
called
it's
an
operator
is
that
right,
yeah,
yeah,
Cube
It's
called
a
Cuba
Ranger,
DD
and
I
haven't
played
with
it
at
all,
yet
to
see
if
it's
working,
but
it's
deploying
so
yeah.
D
B
Yep
so
now,
with
this
being
a
default
once
I
check
this
in
we'll
we'll
have
the
the
default
for
the
DB
host
talking
over
to
it.
So
you
don't
have
to
pass
in.
You
can.
D
C
C
For
me,
no
I
think
what
I
would
like
to
do.
If,
if
you
guys
have
got
the
time
and
Steve
too
is
we
have
a
session
like
we're?
Having
now
and
then
I
say
look,
this
is
what
I've
done.
This
is
how
it
works,
and
then,
by
doing
that,
you
guys
will
give
me
will
give
much
we'll
give
great
feedback
and
suggestions
right
to
make
it
better.
C
B
Now
now
one
thing:
you'll,
you'll
notice:
I
have
to
do
here.
Side
changed
the
values,
file
and
I
want
to
commit
it.
But
what's
going
to
happen,
is
we
ignore
that?
Because
the
workflow
is
checking
in
updates
to
the
charts,
so
we
don't
want
to
get
into
an
infinite
Loop.
So.
D
D
B
No
I
cheat
because
I
have
accessed
directly
to
the
repo,
so
I
commit
directly
to
the
main
branch.
That's.
B
That's
actually
going
to
go
off
and
be
running.
A
A
B
And,
like
I
said
it's
going
to
go
through
and
create
all
because
we're
I
made
the
change
against
the
monolith
repo.
We
have
to
build
the
the
UI
front,
end
Docker
image
build
the
back
end.
Docker
image,
build
the
internet's
image
and
then.
B
Been
updated,
we
actually
released
the
aggregate
chart.
A
C
B
It's
it's
you'll,
see
in
our
repos
whoops.
C
C
B
Yeah,
nobody
could
could
agree
on
what's
what's
happening.
Name
wise,
so,
like
I
said.
After
all,
this
shrugs
long
it'll
actually
go
over.
It
takes
about
20
minutes
to
run,
but
the
last
thing
it'll
do
is
go
ahead
and
Trigger
the
workflow
over
here
and
then
this
workflow
will.
Actually
we
can
look
at
one
of
the
past
ones.
B
And
aggregate
together
all
the
the
latest
versions
from
all
the
different
repos
of
the
charts
and
from
there
it'll
create
a
new
like
Consolidated
parent-child
chart
with
the
latest
versions
out.
There
it'll
actually
commit
it
back
to
the
repo
and
then
it'll
run
the
chart
releaser,
and
what
the
chart
release
does
is
basically
the
helm
package
step.
That's
wrapped
up
in
a
an
action
and.
B
Little
bit
more,
it
creates
a
release
on
the
GitHub
side.
So
this
is
where
it's
creating
the
release,
and
then
it
creates
the
index
that
Emma
gets
it
all
checked
in
and
then
kicks
it
off
onto
the
GitHub
Pages
side
of
things.
A
B
Sasha,
you've
definitely
helped
me
force
me
to
clean
up
a
bunch
of
stuff
yeah
I
I'm
liking.
The
way
things
are
are
pulling
together
so.
C
Yeah
yeah
I
think
yeah,
it's
great
I
mean
if
I
could
help
in
that
way.
I'm
glad.
B
And
what
we'll
need
to
do
is
we'll
need
to
walk
through
the
documentation
that
Tracy
put
together
to.
C
C
B
B
Your
CI
CD
process
into
and
hook
it
up
with
ortelius,
so
you
can
actually
start
generating
so
Arvin.
Do
you
have
Jenkins
running
locally
on
your
machine.
C
C
B
Perfect
yeah,
I
think
and
I
have
the
GitHub
actions
piece
done
so
yeah
I
think
if
we
take
some
of
this,
the
stuff
that
arvin's
been
working
on
and
build
upon
it.
B
And
kind
of
take
the
same
code
base.
So
if
it's
like
a
node.js
or
react
that
you're
building
the
arvin's
building,
that
we
do
the
same,
build
in
Azure,
get
up
actions
and
wherever
else
and
we
kind
of
Plum
it
all.
Together
with
a
nice
clean
example,.
D
C
B
B
B
So,
let's,
let's
keep
going
down
and
try
to
wrap
up
the
artillius
in
the
Box,
get
it
to
a
point
where
we're
happy
with
what
it
is
and
then
we
can
start
adding
on
more
features
like
you
said,
for
the
PG
admin
website
and
some
other
like
the
Jenkins
snacks
I
think
we
should
try
to
get
just
a
a
clean
terraform
to
get
us
going.
A
C
B
Yeah
so
I'll
be
around,
so
just
if
you
run
into
anything
just
send
me
a
message
on
Discord.