►
From YouTube: June 24. 2021 - Ortelius Architecture Australia TZ
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
Okay,
perfect
all
right
yeah
so
give
me
so
give
me
an
update
what
you
guys
have
been
up
to.
B
Sure
so
amit's
been
working
on
the
setting
up
sort
of
the
argo
city,
part
of
it
and
he's
been
seeing
best
practices
around.
B
Originally
he
had
it
going
through
the
ingress,
and
then
he
read
somewhere
that
it's
actually
recommended
that
you
don't
go
through
the
same
ingress
as
your
cluster
and
you
have
it
separate
for
troubleshooting
purposes
and
he's
going
to
present,
though
I'll
get
him
to
write
down
that
the
decisions
right
there
of
those
the
reasons
okay.
But
maybe
we
can
point
you
to
a
couple
blogs
he's
read,
but
it
sort
of
makes
sense
to
me
as
well
that
I'll
get
him
to
read
the
blog.
So
so
what
he's
done?
Sergio's
actually
installed
argo?
B
B
A
So
on
that
front,
you
may
need
to
reinstall,
because
the
cluster
name
changed.
A
And
I
cleaned
up
the
old
azure
subscription.
B
A
A
A
Yeah
and
I'm
not
sure
he
sent
an
email
said
he
was
gonna
drop
out
of
the
project,
so
I
don't
know
how
much
we're
gonna.
What
that.
B
B
B
E
Home
so,
but
I
want
still,
I
want
to
listen
at
least.
E
A
So,
just
to
give
a
rundown,
I
was
telling
brad
that
I
got
rid
of
the
old
subscription
and
the
new
subscriptions
up
and
running
the
new
subscription
is
the
one
that
starts
with
like
5a
or
something
like
that.
Let
me
see
yeah
57b,
so
that's
our
new
subscription
and
that'll
carry
us
through
till,
like
june
1st
of
2022.
B
B
A
So
that's
we'll
see
if
it
takes
it
so.
A
When
I
did
the
cleanup,
I
created
a
new
cluster
because
you
it
won't,
let
you
move
from
one
cluster
to
the
next,
so
I
had
to
create
a
new
cluster.
I
redeployed
the
website
and
the
documentation
over
to
the
cluster,
and
also
I
put
those
into
an
artillious
namespace.
A
So
I
did
separate
out
the
artelias
namespace
this
one
I
gotta
delete,
but
so
that's
all
going
to
be
it's
all
good
to
go
and
I
got
the
linux
foundation
to
redo
the
dns
for
the
ingress
into
istio.
So
our
ingress
goes
to
this
seo
address
as
part
of
that
for
the
cluster.
A
So
that's
out
there.
So
when
you
folks
are
ready
to
go
ahead
and
install
argo,
go
ahead,
it's
all
ready
to
ready
to
go.
B
E
B
And
then
argo
is
not
as
easy
as
I
thought
it
was
well
it
it's
not
obvious
how
to
do
things
so
we've
been
doing
tutorials
on
it
as
well,
so
still
going
for
the
same
goal
as
when
there's
a
change
made
to
the
repository
we
trigger
a
payload
to
the
api,
so
a
person,
the
person
in
my
team
is
making
the
api
to
do
that
as
well.
So
he's
made
the
skeleton
api
and
now
we're
just
actually
trying
to
figure
out
how
to
actually
trigger
that.
D
A
On
the
in
the
document,
let
me
share
my
screen
one
more
time.
A
So
sasha
down
in
south
africa
on
discord
was
saying
that
he
was
running
into
some
sprawl
around
get
ops,
so
he
put
it.
I
asked
him
to
put
in
a
little
problem
statement
of
what
he
is
is
kind
of
has
kind
of
set
up
as
his
pipeline
process.
A
So
it's
we
can.
We
can
bug
them
to
to
fill
in
more
details
on
each
one
of
these,
but
this
is
kind
of
when
I
was
looking
talking
with
syme
as
well.
He
was
on
the
same
page
trying
to
figure
out.
You
know
what
is
the
true
one
of
the
the
problems
around
the
the
get
op
sprawl
that
you
run
into
so
you're,
going
to
have
some
other
data
coming
in
here
as
well.
I
don't
know
what
the
time
difference
is
between
you
and
south
africa.
B
A
B
It
shouldn't
be
too
bad,
I
think
actually,
no,
it's
probably
the
opposite.
That's
probably
still
fine.
Normally
I
can
I'm
sort
of.
I
can
either
meet
people
in
the
morning
or
not
so.
A
A
So
that's
a
resource
to
touch
base
with
is
sasha
and
about
what
he's
running
into
the
the
sprawl
around
the
get
ops
world
and
he's
in
there
trying
they've
implemented
argo.
I
think
for
10
repos.
B
A
B
See
as
well
so
so
he's
saying
that
he's
using
atlantis
for
terraform
now
and
that
still
requires
a
shooting
to
approve,
so
I'm
guessing.
He
wants
to
use
argo
instead
of
it
like
this,
maybe
okay
and
then
ortilius.
I
need
you
to
okay,
so
github
reprocreation
for
onboarding
the
application.
So
I
guess
it's
is
it.
Are
we
saying
here
that
this
is
almost
like
a
pipeline
to
build
the
pipeline?
So
I.
A
It
could
be
that
is
a
possibility
of
I've
heard
other
people
talk
about
this
a
pipeline
to
build
a
pipeline,
especially
when
you're
have
jenkins
pipelines.
You
can
kind
of
clone
an
existing
like
template,
repo
and
use
that
to
bootstrap
a
new
application,
especially
if
your
applications
are
all
the
same.
You
know
if
everything's
gonna
be
a
node.js,
that
you
can
bootstrap
the
repo
and
get
like
sample
code
and
stub
code
to
get
you
going.
B
A
Yeah
so
yeah
exactly
that's
going
to
be
the
build
push
now
this
one.
I
don't
know
what
this
this
part
means,
and
I
think
this
is
where
the
in
in
the
argo
world
you
kind
of
they
have
what's
called
an
environment
repo
and
that's
basically,
your
definition
for
your
application.
A
You
know.
So
if
you
have
multiple
microservice
repositories,
you
kind
of
lump
all
the
manifest
files
into
this
environment
application
repo.
So
all
of
all
the
manifests
live
in
one
place
instead
of
the
individual
microservice
repositories.
I
think
that's
what
he's
talking
about
that,
but
that's
where
we'll
have
to
get
some
more
details.
B
Okay
sure
emma
just
messaged
me
who's
going
to
come
now,
so
he'll
be
able
to
explain
a
little
bit
what
I
was
talking
about
before
okay
cool
here
it
is
and
then
okay
they're,
using
database.
A
Yeah,
the
other
stuff
is
just,
and
I
don't
know
what,
how
how
the
the
argo
cd
you
know.
Three
and
six
relate
to
each
other
yeah.
B
A
Yeah
and
then
in
this
in
this
world,
you
know
if
you
have
a
one
secret
for
like
a
dev
cluster,
and
then
you
have
a
different
secret
for
your
production
cluster
in
vaults,
you're
gonna
have
to
go
and
reference.
A
A
The
argo
is
going
to
have
to
know
which
secret
to
go
in
and
pull,
and
that
may
be
where
helm
is
coming
into
play,
where
you
know
saying
that
we're
doing
a
dev
deployment,
go
pull
all
the
dev
secrets
or
we're
doing
a
you
know
in
the
values
file
we're
doing
a
production
deployment,
go
pull
the
production
secrets
we'll
have
to
find
out
more
from
sasha
on
those
on
six
and
three
and.
B
Four
I
mean
I
can
see
for
me:
I
would
probably
have
hashicorp
volt
and
the
devops
cluster
that
it
used
for
pipelines,
etc,
and
then
I
guess
devon
pod
could
just
have
their
own
one
there'll
probably
be
a
few
volts
going
around,
but
yeah
well
I'll.
Ask
you
about
that.
A
Yeah
and
then
on
these
I'm
not
quite
sure,
because
that
sounds
more
like
a
terraform
issue.
Well,
that's.
C
A
A
I
don't
know
this
is
sasha.
He
is
one
of
the
ortiz
contributors
out
of
south
africa
out
of
cape
town
and
on
discord.
He
was
mentioning
that
he
was
getting
get
up
sprawl
and
I
said:
okay,
that's
a
good.
We,
it's
not
a
good
problem
to
have,
but
it's
a
good
problem
for
us
to
solve.
A
B
A
We
need
to
do
some
more
interrogation
of
sasha
and
I
believe
sime
from,
I
think
he's
either
india
or
pakistan.
A
Yeah
so
syme
had
some
interesting
input
as
well.
E
Sorry,
I'm
thinking
that
I
can
do
a
demo
with
github.
I
already
use
argo.
Usually
argo
is
just
like
a
syncing
server.
That
is
like
just
be
sure
that
what
is
on
the
glasses,
what
you
have
on
on
your
repo.
C
E
For
deployment,
usually
you
just
do
the
strategies
about
branching
to
know
where
to
get
like
some
conflict
map
stuff
like
that,
and
the
last
thing
I
use
I
I
do
like
last
week.
I
think
so
I
do
a
demo
with
hashicorp
and
silent
secret.
For
me,
I
surprisingly
think
phillip's
secret
was
make
more
sense
for
me,
because
hatchikorp
is
like
an
external
api
that
is
like
giving
you
on-demand
stuff.
E
F
Absolutely
I
I
kind
of
agree
with
you
on
that
one,
because
I
was
doing
some
research
so
currently
I
use
cubesat,
which
is
not
as
good
as
select
secrets
of
bitnami,
but
it
still
uses
kms
keys
from
your
cloud
provider,
and
you
can
put
in
your
secrets
encrypted
secrets
into
your
git
repositories
and
then
but
the
problem
is,
it
would
do
the
sync
part
of
it.
You
would
still
use
argo
to
do
that.
Sync,
so
yeah
bitnami
is
a
more
widely
used.
One
though
compared
to
cubesake.
E
Yes,
yes,
isn't
like
more
more
on
on
the
more
native
for
saying
some
way,
yeah
so.
A
I'm
wondering
if
I
wonder
sergio
if
you,
if
we
should
schedule
some
time
for
you
to
do
a
quick
demo
of
what
you
what
you
currently
have
set
up,
and
so
people
can
get
a
better
context
of
how
argo
is
working.
You
know
in
the
artelius
group.
E
A
That's
no
problem,
and
you
know
we
could
just
record
it
and
then
we
can
just
play
it
that
way.
So
you
don't
if
you,
if
you
can't
make
a
meeting
or
something
like
that,
because
I
know
you're
really
busy.
We
can
just
do
it
that
way.
B
I
have
a
front
end
if
you
need
as
well
said
joe,
if
you
want
to
deploy
some
code
for
the
front
end,
I
did
it
in
the
good
ops
repo,
just
as
like
a
dinner.
E
B
Yeah
yeah
I'm
keen
to
through
the
week
like
we
don't
have
to
wait
for
two
weeks.
We
can
also
have
sessions
together
where
we
do
some
of
this
as
well.
Maybe
that's
good
collaboration.
There.
B
A
I
just
like
I
just
like
to
see
the
end
results
where
I
don't
have
to
yeah
yeah
real
quick
just
to
bring
you
up
to
speed
on
the
on
the
azure
side.
I
cleaned
up
our
subscription,
and
now
I
renamed
the
cluster
and
redeployed
everything
that
was
deployed
before
so
everything's
all
straightened
out.
You
guys
should
be
able
to
get
in
there
and
do
whatever
you
need
on
the
azure
front.
B
And
do
you
wanna,
you
were
mentioning
to
me
that
was
about
how
argos
shouldn't
go
through
the
ingress.
Do
you
maybe
wanna
share
with
the
group
your
research
there.
F
Oh
yeah,
I
was
actually
going
through
a
couple
of
conferences
and
that
time
I
also
you
know
there
was
different
talks
which
I
went
through
and
that
time
one
of
the
things
that
was
repeatedly
mentioned
was
adversary,
shouldn't
go
directly
to
your
helm,
etc.
So
that
should
be
your
first
point
of
contact
as
soon
as
you
finish
running
your
terraform
script.
To
set
up
your
cluster
and
the
reason
for
that
is
everything
related
to
kubernetes
that
is
deployed
should
only
be
deployed
to
arbor.
So
argo
shouldn't
go
through
your
ingress
controller.
F
It
should
have
its
own
load
balancer.
So
you
henceforth
know
if
argo
failed,
you're
not
trying
to
debug
whether
ingress
failed,
argo
failed
or
what
is
failing.
So
that's
the
reason
you
would
first
put
in
a
load,
balancer
directly
and
then
argo
city
and
then
post
that
kubernetes
resources,
you're
deploying
goes
only
through
our
goal.
A
Okay,
yeah,
that
makes
sense.
You
know
I
could
definitely
see
on
the
debugging
front
that
you
to
be
able
to
and
also
you
could
actually
tie
that
that
ingress
through
a
firewall
to
just
the
the
get
repo
you
know.
So
you
know
you
don't
have
anybody
else
sneaking
in
through
that
ingress
as
well.
F
F
A
So
go
ahead
and
set
up
the
argo
if
you
need
to
create
namespaces
or
new
ingresses
just
go
ahead
and
have
at
it.
E
Yeah
I
I
was
working
on
that.
I
need
a
little
help,
maybe
a
few
tips
because
first
time
using
like
usher
like
from
from
from
end
to
end,
so
I
must
travel
a
little
getting
the
balance
to
expose
the
service
so
to
connect
the
ingress
to
the
balancing
of
usher.
So
if
you
can
give
me
some
some
tips.
B
B
Yeah,
I
know
a
lot
about
that.
It's
it
can
get
frustrating
sometimes
because
the
the
kubernetes
node
for
it
can't
be
it
has
to
be
in
a
different
resource
that
has
to
be
in
a
different
resource
group
through
the
cluster
and
then
the
yeah.
I
can.
I
can
help
you
there
and
maybe
you
can
share
some
knowledge
about
your
argo
skills
as
well,
which
we
would
be.
B
A
Yeah,
so
always
the
only
thing
I've
been
trying
to
do
is
to
put
resources
in
one
of
the
the
low-cost
regions.
So
it
seems
like
u.s
central
is
a
pretty
cheap
region
to
run
in.
A
So
that's
the
only
thing
you
know
it's
not
a
big
deal,
but
if
we
can
pick
a
cheaper
region,
it's
not
like.
We
need
super
high
performance.
What
we're
doing.
A
F
E
I
was
thinking
about
that
and
I
I'm
just
always
when
you're
in
cloud
you
want
to
maybe
shut
down
at
some
point,
your
thing,
so
I
was
thinking
about
scheduling
some
stuff
because
I
don't
know,
but
at
least
me
I
I
thought
I
wasn't
using
argo
cd
for
example.
So
maybe
we
can
save
some
money
when
scheduling
things
scale
down
and
scan
the
app
when
we
needed
something
like
that.
So
the
problem
was
the
documentation.
A
Yeah
we're
running
our
website
and
our
docs
off
of
that
cluster,
so
we
just
need
to
leave
it
up
and
we
have
plenty
of
money
to
back
our
subscription
for
the
next
year.
Last
year
we
spent
like
seven
grand
out
of
the
25
000,
so
yeah
we
have
room
to
play.
F
A
F
Yeah,
I
I
think
if
we
implement
a
node,
auto
scaler
anyway,
so
our
cluster.scaler.
We
should
be
good
because
that
will
take
care
if
the
resources
are
fewer.
You
then
automatically
your
auto
scaler
will
kick
in
it's
anyway.
A.
B
Because
because
you
can
have,
I
think
I
don't
know
if
I'm
talking
about
gcp
here
or
azure
they're
probably
the
same,
but
you
can
have
two
node
pools.
One
node
pool
is
for
like
stuff,
you
like
the
resources
that
you
actually
need,
and
then
you
can
have
spot
instances
for
your
second
notebook
that
you
say
I
want
to
buy
a
fifty
percent
price,
but
you
can
scan.
A
A
F
Nice
cool
sr
also
it's
called
spot
instances
in
gcp.
It's
called
some
other
name
but
azure.
A
A
Yeah
in
what
is
it
in
google.
F
Yeah,
I
I
was
working
on
it,
but
it
I
think
it's
called
transitive
or
something
to
that
effect.
I
I
can't
I'm
dragging
a
blank
too.
E
Run
a
ground
task
like
a
timer
like
heating,
up
from
time
to
time,
to
check
the
activity
on
the
on
the
deployment
and
just
scale
down.
That's
the
way
that
when
you
don't
have
that
by
default,
because
I
usually
work
on
containers
on
training,
so
usually
we
we
don't
work.
We
don't
have
like
a
google
cloud
or
an
amazon
feature
so,
but
you
can,
though,
that
is
not
so
expensive
to
do
or
complex
to
do.
A
Yeah
so
right
now
it's
set
up
to
auto
scale
and
the
minimum
node
count
is
one
and
the
max
right
now
is
five
and
right
now
it's
running
at
one.
You
know
so
scaling
up
the
five.
It
should
take
care
of
all
the
you
know,
that's
the
max
I
just
set
for
now.
We
can
go
higher
than
that.
If
we
need
to.
F
A
B
F
A
B
E
On
a
really
basic
basis,
so
maybe
switching
branches
and
yeah
hooking,
like
change
by
default,
he's
going
to
pass
a
lot
of
triggers
like
by
default
for
okay.
The
healthy
shakes
is
always
checking
that
this,
the
thinking
of
the
component,
so
we
mainly
listen
to
that
to
ensure
like
nobody
is
touching
directly
with
the
cluster
on
the
state,
is
the
original
style
or
the
desired
state.
So
I
I
I
use
it,
but
I
like
a
compliment
for
deployment,
not
focuses,
so
I
never
use
like,
maybe
an
advanced
way
to
use
it.
So
it's
more.
E
F
F
E
A
It
can
either
be
each
argo
cd
or
I
believe
argo
has
like
an
agent,
but
basically
the
the
operator
in
the
cluster
is
going
to
be
looking
for
back
to
the
repo
for
a
particular
change
to
either
like
a
branch
or
a
tag,
or
you
know,
they're
they're
introducing
something
called
our
application
sets
and
that
will
do
some
templating
around
directories
as
well.
A
So
that's
kind
of
what
they're
now
the
the
the
tricky
part
is.
When
you
know
the
problem
set
that
we're
looking
at
with
ortelius
is
you
know
we're
we're
looking
at
managing.
A
You
know
100
to
5
000
microservices,
so
you
have,
you
know
we'll
keep
it
simple,
100
micro
services,
all
the
manifest
files
in
the
same
repo-
and
you
know
you
get
a
lot
of
sprawl
on
that
front.
But
then
also
you
go
through
your
pipeline.
You
got
the
development
and
then
you
got
integration
testing
and
then
you
may
have
qa
testing.
A
One
of
my
one
of
our
customers
has
17
qa
environments
for
one
one
application
and
then
you
have
staging,
and
then
you
have
production,
you
know
they
usually
have
a
uat
or
staging
and
then
production,
so
the
pipeline
it
can
and
managing
all
the
branches
and
all
that
can
get
complicated.
A
A
You
know,
maybe
it's
just
the
the
name
of
the
secret.
You
know
the
qa17
secret
versus
the
qa15
secret.
That
would
be
the
only
environment
change
that
we
would
have.
That
would
push
into
the
get
repo
at
that
level
from
or
the
artillia
side.
So
that's
kind
of
what
I'm
kind
of
envisioning.
As
I'm
learning
more
about
the
the
get
ops
process.
B
B
And
emma
just
a
quick
question
you
mentioned:
argo
should
be
the
first
thing
that
you're
installing
it
it
shouldn't
be
in
the
helm
chart.
Are
you
saying
no.
F
A
I
know
people
have
been
talking
about
customize
as
a
way
to
do
the
templating
and
it's
a
little
bit
easier
than
helm,
because
you
are
basically
working
from
my
understanding.
A
Don't
quote
me
on
this
is
that
you
could
take
an
existing
manifest
look
for
some
key
value
pairs
or
strings
in
those
yaml
files
and
replace
them
pretty
easy
and
keep
everything
intact.
So
it's
a
from
what
I
understand:
it's
a
better
templating
engine
than
the
helm,
where
you
have
to
kind
of
like
code.
What
your
template
looks
like
where
customize
is
more
on
the
fly
world
is
for
when
I
can
get
gather.
A
Yeah
so
like
with
with
helm,
you
have
your
values
file
and
you
take
the
values
file
in
your
home
chart
and
it
kind
of
merges
them
together
with
customize.
A
You
take
existing
manifest
files
that
don't
have
any
templating
like
tags
in
it,
and
then
you
can
go
in
and
do
basically
searching
and
replaces
against
those.
A
It's
like
yeah,
it's
like
a
patch
system,
exactly
okay,.
A
F
B
B
Yeah,
I
guess
sergio
what
what
day
works
for
you,
I'm
really
keen
to
maybe
have
a
session
like
a
technical
session.
What
works
for
you.
E
G
B
B
A
B
B
Yeah,
that's
very
interesting
like
I
was
meaning
to
ask
about
that
because
I
didn't
understand.
A
B
F
B
D
A
Yeah,
because
that
way,
we'll
it's
easy
to
get
a
hold
of
sergio
through
that.
A
That's
why
I
wanted
this
and
you
can
just
do
a
direct
message
to
him
if
you
want
sure.
B
Yeah,
I
think
this
this
sort
of
sprint.
We
can
sort
of
put
all
the
pieces
together
now
we
should
be
able
to
have
some
output.
A
Yeah,
I
think
we're
getting
the
our
ducks
in
a
row
and
kind
of
getting
a
game
plan
together.
So
I
think
we'll,
like
you,
said,
we'll
we'll
start
moving
along
here,
pretty
quickly.
Yeah,
oh
and
sergio's
discord
handle
is
force
acre.
A
Let
me
see
if
I
find
her
yeah
and
you'll,
so
it's
force
acre.
Let
me
just
share
my
screen
real
quick,
so
you
can
take
a
quick
capture,
but
this
is
his
his
handle.
B
Okay
and
did
tyrus,
did
he
he
just
said
he's
completely
out.
Did
he
or.
A
Yeah
he
said
he
he
was
studying
for
some
certifications
and
then
he
had
a
drop
off
for
a
bit.
A
B
A
And
if
so,
if
I'm
still
yeah,
so
when
you
log
in
you'll
there,
here's
the
welcome
channel
that
everybody
keeps
on
you
get
notifications
from,
but
go
to
the
code
contact
one
and
just
click.
This
check
mark
okay,.
D
A
That
one
yeah
and
that'll
open
up
the
rest
of
the
channels
that
are
over
here.
B
D
Cool
all
right
cool,
so.
A
B
A
I
don't
see
it
just
touch
space
with
here.
It.
A
So
it's
a
little
bit
way
a
ways
back,
but
you'll
be
able
to
find
them
and
simon's
out
here
as
well.
They
bring
up
pretty
good
conversations.
B
B
A
If
you
run
into
any
problems
with
the
azure
side
like
permissions
or
anything
like
that,
just
reach
out
to
me
and
I'll
get
them
yeah,
we'll
figure
out
how
to
get
them
sorted
out.
Yeah
sounds
good.
Like
you
said,
the
azure
ui
is
a
interesting
world.
It's
horrible!
It's!
It
is
so
confusing.