►
From YouTube: Oct 14, 2021 - 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
If
we
just
need
to
go
and
add
the
github
action
to
kind
of
push
everything
to
the
the
single
github
pages
site
under
the
ortelius
ortilius
repo.
Does
that
make
sense.
B
Yeah,
it
makes
sense,
except
the
annoying
thing
is
I
found
out.
You
know
how
you
had
put
a
comment
about
the
chart:
repo
url,
linking
back
to
our
tvs,
tortillas
ortiz
thing,
yeah
that
plug-in
annoyingly
works
the
opposite
way.
So
it's
it
doesn't
push
the
individual
repos
to
the
ortiz
tortillas.
What
it
does
is
it
just
ortivius's
index
and
puts
it
into
the
small
into
the
sub
repo,
so
it
it
worked,
the
opposite
of
what
I
intended
it
to
do.
B
It's
it's
almost
like.
I
would
meet
it's
gonna
get
messy,
because
the
way
I
can
see
it
works,
I
get
one
the
first
poly
repo
and
that
gets
linked
to
the
second
point
and
it
chains
all
the
way
down
and
or
tedious
or
there's
the
big
one,
just
picks
the
very
last
one,
but
that
that's
just
not
ideal,
because
you
have
a
chain
and
if
you
break
one
of
the
repos,
it's
going
to
break
the
whole
thing.
So
that's
not
so
not
so
good.
A
Right
so
I
wonder:
if
we're
going
to
need
a
hey,
brad.
C
A
That's
a
so
I'm
wondering
if
we're
gonna
need
to
do
a
create
the
index.yaml
in
each
polyrepo
and
then
do
another
step
where
we
kind
of
push
it
and
merge
it
with
the
the
main
one
on
the
arteriosus
ortulius
side.
C
C
A
Yeah,
I
think
we
can
well
the
other
thing
we
could
do.
Is
I
didn't
look
into
that?
The
github
actions
that's
doing.
The
push
is
probably
off
of
some
open
source
project
or
some
open
source,
git
repo.
B
C
Yeah
I
can
maybe
I
can.
We
can
write
an
issue
and
then
I
can
just
write
some
code
that
will
do
that
because
in
it
I'd
rather
do
it
to
theirs
and
make
my
own
one.
Because
then
I'm
going
to
maintain
it.
A
Yeah
exactly
because
I'm
wondering.
C
C
B
A
D
Am
correct,
correct,
so
where's
cr.
What
what
language
is
that
cr?
That's
just
the
chart
releaser,
I
think,
as
in
the
just
to
the
80s,
which
I
release
it
is
it
the
helm
command
for
chart
releaser
no.
C
I'm
just
gonna
put
it
in
the
chat.
I
know
it's
probably
looking
the
same
thing
as
me.
Well,
actually,
why
don't
we
just
someone
I'll
share
my
screen
right,
yeah
sure
we
don't
have
to.
C
A
No
there's
another
one:
it's
just
cr,
not
cross
dot
sh.
So
if
you
go
down
to
the.
A
C
A
That's
what
we
have
to
figure
out
where
that
cr
is,
if
it's
another
shell
script
or
if
it's
oh,
I
found
it.
So
if
we
go
to
helm
slash,
chart
releaser
is
the
repo.
A
C
I
think
it's
because
they're
using
cobra
for
their
cli.
So
if
I
go,
I
mean
oh
yeah.
They
are
using
cobra
because
it's
got
the
root
command
and
then
you
go
to
the
other
command,
so
that
will
say:
get
the
config,
probably
using
viper,
execute
and
then
the
four
commands
that
they
have
index,
which
is
the
one
we're
concerned
with.
A
C
A
A
A
Oh
okay,
I
didn't
know,
I
didn't
realize
you
could
reference
it.
A
C
C
A
Well,
the
interesting
thing
is:
if
you
look
at
the
the
index.yaml
that
was
generated
for
ortilius
it
references
like
the
releases.
E
C
C
We're
just
talking
about
the
chat
release
for
the
artifact
hub
to
get
all
of
them,
so.
A
So
go
to
so
go
to
ortillius
in
the
github
pages
branch.
C
E
A
It
has
you
know
it's
packaging
up,
like
you,
said
the
urls
over
under
the
downloads
already
yeah
exactly
now.
If
we
run
the
github
action
and
the
polyrepos
we'll
get
the
same,
the
packaging
will
be
there
we'll
just
and
we'll
actually
get
an
index.yaml
over
on
that
side,
but
we
can
ignore
that
yaml
or
we
merge
it
over
on
this
side.
Yeah.
C
C
C
C
A
C
A
C
A
If
and
if
when
we
go
to
go
and
install
ortilius,
my
assumption
was
that
we
would
need
to
have
a
a
parent
helm
chart
and
then
the
children
help
chart.
So
a
charter
charts
basically
yeah.
That's
what
my
assumption
was,
but
if
there's
a
different
way
to
do
it
that
I'm
totally
open
to
that.
But
I
was
assuming
the
way
we
would
have
to
consolidate
would
be
the
charter
charts
route.
F
Yeah,
so
I
think
what
steve
is
saying
is
right,
so
it's
best
to
do
chart
of
charts
and
then
put
a
dependency
in
there
and
dump
in
all
the
other
files.
You
can
also
have
configuration
like.
I
showed
you
yesterday,
where
you
can
enable
each
of
the
charts
or
disable
them
individually
if
based
on
what
is
required
for
installation.
F
C
Yeah,
so
are
we
saying
then,
in
that
case
that
yeah
we're
going
to
have
to
make
changes
to
the
other,
centralized
home
chart
to
have
the
parent
chart
and
then
the
other
ones
underneath
it.
A
C
C
Yep,
okay,
that's
fine,
and
then
in
that
case,
then
let's
say
we
make
a
change
to
this.
Specifically,
you
know
the
the
new
release,
so
it
wouldn't
yeah.
A
C
C
Yeah
so
like
with
this
last
step
like
assuming
that
that
builds
correctly,
maybe
the
last
step
of
this
is
to
do
a
do
a
commit
to
the
waterless
materials
to
change.
C
Yeah,
it
would
actually
be
in
the
chat,
but
you
would
essentially
this
little
script
or
you
know
whatever
would
just
update
to
a
pr
to
that
and
update
the
number
and
then
how
do
we
want
to
do?
That
is
that
we
just
do
it
commit
programmatically.
A
We
could
probably
just
we
could
probably
skip
the
pr
and
just
push
directly
against
the
that
branch.
C
Yeah
on
archers,
because
yeah
we
don't
care
yeah,
we
can
just
yeah,
that's
right.
We
can
just
we
don't
actually
want
it
in
the
github
pages,
it's
isolated
from
our
curry
anyway,
so
yeah
yeah.
So
then
yeah.
Of
course,
we
wouldn't
want
to
waste
our
time
manually,
doing
vrs
yeah.
I
guess
I
guess
maybe
we
can
think
about
some
sort
of
like.
C
A
Right
but
we
could
do
that
on
a
a
github.
C
C
Yeah,
so
when,
when
commit
to
main
or
master,
I
think
it
is
on
this
case.
It
triggers
a
pipeline
just
to
do
a
sending
check
if
it
fails,
and
it
rolls
back
the
change
or
something
like
that.
Yeah.
A
The
other
thing
is,
or
artelias
or
deploy
hub,
can
pick
up
the
changes
to
the
charts
and
do
the
deployments
to
the
azure
cluster.
Well,
and
that's
where
we'll
have
argo
in
place
as
well
to
watch
the
for
the
charts
so
we'd
use
our
our
argo,
kept
it
in
in
artelia's
process
our
getups
process
to
monitor
us
and
push
the
changes
over
to
our
dev
environment.
A
Think
of
would
be
instead
of
on
that
little
script
to.
A
Update
the
you
know,
iterate
the
the
version
number
is
we
just
do
a
merge
of
the
index.yml
and
the
you
know
do
a
full
merge
of
the
index.aml
and
push
over
the
tgz
files
for
the
the
charts
themselves
under
ortulius
yeah
yeah.
It's.
C
Yeah,
it's
not
in
the
source
code.
The
release
is
on
source
code
anyway,
they're
separate,
so
we
don't
really
care
how
much
stuff
we
so
you're
saying
put
all
of
the
all
of
those
artifact
files
under
the
same
release
in
hotels
yeah.
C
A
Source
of
truth,
but
you
can
get
a
little
bit
out
of
sync,
but
I
don't
think
we'd
have
to
worry
about
it.
So
on,
like
the
text
file,
ms,
the
github
action
would
be
just
to
do
a
commit
of
the
chart
itself
underneath
the
artillious
github
pages.
B
A
D
C
I
can
know
if
I
understood,
which
thing
sorry
what
we're
saying,
instead
of
using
the
home
release
package
of
what
we're
doing
now,
to
create
the
artifacts
in
the
index.yaml
for
the
metadata,
we're
simply
just
going
to
move
the
whole
helm
chart
underneath
ortilius
so
that
we're
skipping
the
home
release
apart,
but
we're
just
moving
the
the
manifest
files
for
helm
over
to
the
authorities.
Regrow
that
correct.
A
And
then,
then,
once
they
hit
the
artelius
ortilius
repo
under
the
github
pages,
then
we
go
ahead
and
or
wherever
we
decide,
we
could
do
another
branch
or
whatever
that
we
would
then
create.
The
index
update
the
github
pages
branch.
With
all
the
details
by
running
the
chart,
releaser.
C
I
think
I
I
do
like
that,
because
we
don't
have
to
stuff
around
and
do
github
pages
for
each
polyrevo,
which
is
sort
of
time
consuming
a
manual
yeah.
A
C
C
So
so
what
we're
trying
to
achieve
is
one
one-stop
shop
for
autelius
deployment,
so
we
wanted
to
be
able
to
deploy
its
all.
The
polyrepos
obviously
have
helm
charts
in
those
repos,
and
we
want
it
to
be
easy,
artifact
hub,
pointing
to
just
one
of
these.
So
if
I
want
to
go
to
artifact
tab
to
install
tillius,
it's
points
to
the
mono
revo,
which
modern
poly
doesn't
matter.
It's
just
that
it's
index
in
one
one
place.
Does
that
make
sense.
F
C
So
you're
saying
that
yeah
this
is
the
so
that's
one
way
and
you're
saying
the
other
way
is
that
in
the
helm
chart
we
use
the
url
for
that
ham,
cha
into
the
poly
yeah.
Sorry.
A
C
A
The
issue
with
that
was
having
to
to
iterate
the
version
number
every
time
one
of
the
microservices
gets
built.
C
Yes,
because
you're
still
going
to
have
to
update
auto
necessarily
one
one
way
or
another,
so
both
of
them
are
no
better
than
each
other,
but
it's
like
what
it.
What
is
the
sort
of
I
guess,
cleanest,
neatest
way
to
do
it,
so
we
can
either
update
the
urls
or
we
can
just
simply
copy
the
whole,
which
is
just
small
text
files
right
like
yeah.
This
is
yeah.
It's.
C
Yes,
so
so
with
the
helm
package
releaser
what
that
well,
essentially
that,
obviously
you
already
know
this,
but
I'll
just
say
it
again,
but
the
the
github
action
will
package
the
helm
chart
into
artifacts
and
then
index
that
metadata
into
an
index.yaml
file.
So
it
says
this
is
my
metadata.
C
C
So
it's
like
we
have
that.
The
one
thing
for
a
fact
that
we
need
to
do
at
the
moment
is
update.
The
mono
repo
mediator,
chart,
chart
information,
so
there's
two
ways
of
doing
it.
Probably
more,
but
the
two
ways
we
come
up
with
at
the
moment
is,
and
there's
a
third
option
as
well.
Is
that
we
investigate?
C
Like
the
coolest
way,
I
propose
we
at
least
I
can
make
an
issue
in
in
the
home
release
to
say
this
is
what
we
want
to
achieve.
If
we
get
some
good
feedback
from
that,
if
we
don't,
then
we
simply
just
make
our
own
little
script.
F
Yeah,
that
sounds
pretty
good.
I'm
just
trying
to
look
at
other
artifacts
which
only
provide
this
capability,
like
nri
bundle,
which
I
showed
you
yesterday.
So
I'm
just
trying
to
see
what
they
have
done
right
now,.
C
A
F
So
this
is
nri
bundle,
this
is
new
relics
chart
and
what
they
do
is.
This
is
the
chart
of
charts
where
they
are
including
multiple
charts
and
is
this.
This
is
the
metadata
that
you're
talking
about
right.
C
That's
the
readme,
so
the
metadata
is
a
index
dot
yaml.
So
so
this
is
the
the
hong
cha
yeah
this.
This
doesn't
actually
have
an
index
in
this
case,
so
so
so
they're
not
using.
Probably
it
doesn't
yeah.
So
it
looks
like
what
they,
what
they're
actually
doing
is
they're
making
a
helm
chart
repo
only
for
home
charts
instead
of
having
it
in
the
mono
repo
they're,
just
having
a
helm,
chart
reaper
and
they're
just
simply.
F
Yeah,
so
what
here
you
can
see
that
they,
what
they
use,
is
nri
one
and
they
have
their
requirements.yaml,
which
is
their
dependencies,
and
here
they
mention
all
the
questions
as
we
had
discussed
before,
and
you
can
either
do
it
via
file
or
you
can
actually
do
it
via
url
as
well.
So
this
is
consistently
what
I
found
in
most
of
the
places.
This
is
what
is
used.
This
format.
C
E
C
F
C
Yeah
like
that
ways,
that's
like
option
three,
because
I
guess
how
mono
do
you
want
the
autelius
ortilius
repro
to
be
like
how
much
like?
Is
it
even
a
best
break?
Yeah,
I
don't
know
it
is.
A
one-stop
shop
is
good
and
then,
at
the
same
time,
split
up,
there's
no
right
or
wrong.
A
A
So
we
could
actually
do
this
under
our
website
repo
as
well,
which
may
give
us
some
more
flexibility
in
in
that,
because
we
can,
you
know,
strip
off
version
numbers
and
things
like
that
in
in
file
names,
and
things
like
that,
if
we,
if
we
need
to
so
that
may
be,
another
option
is
instead
of
trying
to
use
github
pages
that
we
would
just
use
the
main
trunk
of
the
website
as
the
as
what
the
point
artifact
hub
too.
A
We
need
a
way
to
just
serve
up
the
the
index.yaml.
F
Yeah,
because
the
nri
bundle
also
has
the
option
of
directly
using
the
charts
in
itself,
so
they
let
me
quickly
share
my
screen
again.
F
A
C
An
it
as
well,
it's
really
the
updating
part
that
yeah,
how
about
I
just
tasked
ben
with
this
this
one
next
week,
and
then
we
just
see
what
he
comes
up
with
like
he
can
present
a
couple
of
ways
too:
okay
yeah.
How
does
that
sound
like.
A
C
C
Yeah
yeah,
he
works
in
my
team,
so
I'm
I'm
like
managing
his
like
work
tickets
and
then
I
manage
as
open
source
stuff
as
well.
Just
so
he
doesn't
go
too,
like
I'm
gonna
spend
all
my
time
working.
It
opens
up
right
yeah.
So
I
can
like
assign
this
to
him
as
well
he's
a
really
smart,
smart
person,
so.
A
C
So
so
in
the
home
packager
to
when
you
create
your
metadata
and
your
artifact,
it
then
said
what
is
the
chart
url
that
you
know
you
want.
We
were
thinking
that
the
index.yet
the
chart
url
you
could
you
could
specify
that
to
any
repo
and
then
that
would
update
the
index
of
one
one
place.
That
does
everything
for
us
and
it
didn't
do
it
like
that,
but
it
got
it
yeah.
A
F
C
A
E
C
That
will
be
a
good
little
task
for,
like
a
like
a
good
first
issue,
type
tasks
for
someone
to
do.
A
Yeah
and
anything
else
on
that
you're
gonna.
C
No,
I
think,
like
we
really
want
to
solve
this
artifact
hub,
because
that's
gonna
make
all
of
that
application
set
so
easy
just
to
reference
yeah,
definitely
and
then,
and
then
we're
going
to
have
a
few
issues
like
once.
We
start
deploying
these
deployments
and
then
we
will
just
make
issues
for
other
people
to
help
us.
You
know
solve
like
like
the
silly
one
like
it
that
values
file
needs
this
image
tag
or
something
like
that.
A
C
A
Real
quick:
this
is
what
sasha's
been
working
on.
A
A
And
then
we
do
do
the
check
in
and
then
captain
kit
gets
kicked
off.
You
do
the
build.
You
do
a
bunch
of
stuff,
then
you
go
over
and
tell
ortulius
to
do
its
stuff
and
then
eventually
it
comes
back.
Ortilius
interacts
back
with
github,
argo
picks
up
those
changes
and
then
once
argo's
done,
we
go
back
and
tell
captain
that
we're
done
here.
So
this
is
still
a
little
bit
of
a
work
in
progress.
A
We're
working
on
rolling
this
into
the
documentation
as
a
proposal
and
then
there's
all
the
the
captain
stuff
as
well-
and
this
is
just
like
page
one-
we're
gonna
do
qa
and
production,
because
the
the
number
of
steps
is
slightly
different.
You
know
whether
you're
doing
builds
or
not
builds,
and
things
like
that.
But
it's
pulling
together
we're
getting
there.
Yeah.
C
A
That's
what
you
get
for
sending
in
austral,
a
south
american
south.
C
C
Yeah,
that
was
yeah
quite
cool,
okay,
cool
and
then
the
only
other
thing
just
to
bring
up
quickly
is.
We
still
do
have
two
clusters
running
just
always
think
about
it.
You
know:
do
we
want
to
ditch
one.
A
I
gonna
get
close
to
that
we're
doing
testing
right
now
and
I
would
say
after
we
get
the
testing
done,
that
that
we
can
flip
over
and
just
get
rid
of
one
of
them.
C
Yeah
because
there's
pros
and
cons
of
my
one
because
my
one's
set
for
production,
so
it's
not
as
easy
for
like
new
developers
to
come,
and
you
know
the
the
api
kubernetes
api
server
is
not
exposed
to
the
world.
So
it's
a
private
cluster.
So
maybe
that's
that's
one
of
the
con
well
pros
and
cons
at
the
same
time
because
it's
safer
but
we're
doing
open
source
anyway.
So,
like
everyone,
can
see
the
code
anyway,
we're
not
exactly
it's
not
in
a
customer
account.
C
Maybe
that
setup
can
be
for
a
customer
for
deploy
hub.
Maybe
that
can
be
the
deploy
hub
version
and
then
the
other
clusters
more
like
the
open
source.
A
C
A
A
C
Yeah
I
tried
to
put
alerts
on
it,
but
it
said
because
it's
one
of
those
like
sponsored
subscriptions
you're
not
allowed
to
so
it
was
a
little
bit
frustrating.
A
Yeah
they
kind
of
hide
things
for
me
I'll.
Let
you
just
go
crazy
and
then
you
then
they
send
you
a
very
big
bill
at
when
you,
your
subscription
runs
out.
A
Well,
I
think
we're
making
great
progress
and,
let's
keep
me,
keep
me
in
the
loop
I'm
gonna
be
at
at
the
linux
foundation-
leadership
summit
on
that
early
next
week.
So,
if
there's
anybody
that
we
need
to
reach
out
over
there,
you
know
like
from
captain
or
argo
or
or
wherever
just
drop
me
a
note.
C
C
Yeah
sorry,
my
internet's
going
bad
is
that
the
linux
foundation,
member
summit
yeah
the
one
in
nappa-
oh
cool,
oh
the
napa
valley,
yep,
okay,
oh
that
sounds
like
what
might
come
next
time.
C
C
List
that's
going
on
my
runoff.