►
From YouTube: August 19, 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).
B
B
So
go
ahead
and
and
sign
in
just
want
to
bring
some
of
you
that
haven't
been
around
for
a
while
kind
of
up
on
the
up
to
speed
where
we're
at
and
some
of
the
things
that
were
trying
to
wrap
up
so
the
new
ui
is
is
looking
really
well.
We
got
the
for
the
service
catalog
pieces,
so
we
have
the
the
swagger
library
in
there
and
being
able
to
do
a
readme
for
the
component.
We
also
this
week
we
brought
in
the
license
file.
B
Part
of
the
front
end
now
so
we're
we're
going
through
and
grabbing
this
information
from
various
places
and
it's
pulling
together
really
well.
We
have
another
member
team
member
call.
His
name
is
uttarsh.
B
He
did
a
bunch
of
work
around
the
running
cyclone
and
safety
scanning
a
docker
image
and
bringing
that
stuff
into
ortelius
so
and
we're
able
to
get
the
links
going.
Let
me
see
if
I
actually,
which
I
can
show
a
demo
here.
B
Oh
wow,
that's
loading
up,
so
it's
it's
pulling
together,
really
nice,
some
of
the
things
that,
from
a
technology
standpoint
that
were
pulling
together
last
minute,
is
the
pie,
breaker
piece
chroma's,
working
on
some
connection
pulling
for
the
python
flask
in
conjunction
with
a
pie
breaker
and
we're
just
trying
to
get
all
the
code
standardized.
B
B
You
know
what
how
we
need
to
structure
our
code
if
we
lose
a
database
connection
in
the
middle
of
a
transaction,
you
know
how
do
we
recover
from
that
point?
You
know.
We
know
that
that
that
transaction
is
going
to
be
a
dead
transaction
and
it'll
it'll,
throw
back
an
error
to
the
front
end,
which
is
fine,
but
then
how
does
the
that
flask
thread
recover
on
its
own
and
not
crash
the
container?
B
So
that's
part
of
the
the
work
that
we're
doing
to
make
sure
that
it's
gonna
go
and
like
go
through
its
retry
logic.
To
reconnect
to
the
connection
pool
clean
that
up
and
then
when
we
have
that
reestablished
and
the
database
comes
back
up
that
the
next
transaction
that's
coming
in
is
going
to
be
able
to
work,
and
we
have
some.
You
know
graceful
air
handling.
On
that
side,
it
doesn't
look
like
it's
going
to
be
too
bad
to
get
that
to
work.
B
B
Let's
see
so
that's
where
we're
doing
some,
the
the
testing
around
that
and
once
we
get
that
all
sorted
out,
we'll
roll
it
through
all
the
right.
Now
we
just
have
five
micro
services,
five
or
six
micro
services
that
we're
going
to
be
using
for
the
service
catalog
pieces.
B
So,
along
with
that
we're
going
to
need
to
change
the
install
process.
You
know
before
we
had
a
single
monolith
container
that
we
could
just
roll
out.
You
can
run
docker
and
run
it
locally
or
you
could
put
into
a
kubernetes
cluster
as
a
single
container,
but
now
that
we
have
the
microservices
coming
along,
we
need
to
think
about
how
to
kind
of
package
that,
together
as
part
of
that
process,
you
know
so
each
each
each
repo
has
its
own
helm
chart.
B
So
we
may
want
to
look
at
building
up
like
a
app
of
apps
for
argo,
for
people
that
have
argo
installed,
that
they
can
just
point
to
our
our
public
repo
and
grab
everything
that
way
or
you
know
we
may
want
to
look
at
some
other
ways,
maybe
a
a
helm
chart
that
has
nested
home
charts,
so
I'm
open
to
ideas.
If
anybody
has
any
suggestions
on,
you
know
how
we
get
this
rolled
out
easily
to
a
kubernetes
cluster.
A
Okay,
so
when
you,
when
you
do
that,
usually
you
want
to
come
from
from
glitter
to
be
here,
so
I
think
the
the
one
one
app
with
culture
that
can
just
deploy
all
the
components
is
good
to
go
like
for
initial
things.
A
After
that
yeah,
I
will
recommend
go
to
to
an
argo
like
an
upset
configuration
because,
right
now,
on
the
github
side,
it's
more
scalable
that
you
can,
because
at
the
end,
argo
is
going
to
take
a
lot
of
multiple
charts,
helm,
charts
and
it's
capable
to
deploy
it
to
multiples
targets.
So
that's
the
fun
part
so
see.
If
you
already
have
a
helm
chart
you
just
going
to
start
start
to
to
split
that
into
the
multiplication
one
and
just
integrate
it
to
argo,
so
yeah.
A
I
think
that
that's
like
the
the
natural
road
to
that
one
helm
shot
like
a
monolithic
counter
force
in
some
way
right.
That
is
split
in
a
in
a
a
useful
upset
and
mix
it
with
the
configuration
for
argo,
and
I
will
say
that
the
last
step-
and
that
is
more
like
future
desire
and
not
totally
sure
that
can
happen.
But
I
think
I
strongly
think
that
today
that
the
more
ultimate
way
to
work
is
just
having
operators.
A
So
it's
not
it's
nice
to
have
develop
on
an
operator
for
for
telus,
because
yeah
github
is
good
to
go
when
you
are
like
doing
like
updates.
It's
not
so
it's
it's
not
just
a
product
itself.
When
you
have
like
multiples,
multiple
applications
and
at
the
end
you
have
a
solution
that
change
on
base
of
the
the
needs.
Yeah.
You
probably
are
going
to
work
with
the
dogs
and
stuff
like
that.
A
If
you
have
your
control
plane
for
operators,
maybe
you
can
work
with
that,
but
the
operator,
the
control
plane
for
operators
today
in
the
in
the
in
the
public
clouds
and
the
prevent
ones,
is
like
more
common
to
have
it.
Yesterday,
I
don't
know
a
few
months
ago
was
like
maybe
you
can
put
it
if
you
have
the
condition,
but
today
is
more
easy
to
think
that
the
condition
for
for
good
to
go
and
and
deploy
an
operator
is
more
common
than
a
few
more.
B
Where
is,
is
the
opera?
Is
there
a
operator
catalog,
then
you
know
like
they
have
yeah.
B
That's
good
ideas,
because
one
of
the
goals
is-
and
we
run
into
this-
we've
run
it
I've
run
into
it
forever
in
the
software
world.
Is
it
where
people
give
up
because
it's
too
hard
to
install
so
they
they,
if
they
can't
figure
it
out
in
five
minutes,
they're
done
with
it.
So
that's
kind
of
the
the
criteria
is
you
know
a
five
minute
install.
A
Yes,
I,
for
example,
I
don't
know.
Last
week
there
is
an
from
from
rehab
cup.
They
deliver
an
operator
for
managing
certificates.
It
is
sir
utils
operators.
A
Basically
it's
just
one
application
that
read
annotation
on
our
on
a
router
on
an
ingress
control
to
say
in
some
way
if
he
gets
like
this
specific
annotation
he's
going
to
inject,
like
secrets
with
the
certificate
into
the
the
the
router.
So
probably,
if
I
have
to
configure
that
and
doing
tested
it,
I
I
won't
probably
use
it,
but
they
it's
not.
I
want
to
say
it's
not
a
complex
pro.
A
It's
not
like
a
big
platform
that
they
deliver.
It's
just
like
a
new
team,
but
on
an
operator
and
was
just
one
comment
or
one
click
to
get
working
on.
It
get
work
on
it
on
your
curing
cluster,
so
yeah!
It's!
It's
amazing
hope
you
can
yeah
I'm
going
to
use
this
because
it
was
like
that
a
click,
an
annotation
and
it's
working.
B
Well,
that
brings
up
like
bitnami,
you
know,
bitnami
does
a
bunch
of
well.
They
do
kind
of,
like
the
helm,
the
monolithic
helm
chart
on
the
bitnami
side.
If
we
should
be
publishing
out
to
bitnami.
A
Yeah
bitnami
is
like
a
good
example
that
they
always
like
package
things
to
get
you
easily
use
it
on.
You
know,
because
I
don't
know
like
years
ago,
like
having
a
virtual
machine
applying
that
you
just
turning
on
and
works.
Yes,
today,
operators
is
pretty
much
done.
Okay,.
B
And
on
the
operator
side,
I
have
not
ever
worked
with
them.
I
know
I
understand
what
they
are
and
stuff
like
that.
But
do
you
have
to
install
anything
to
make
an
operator
work
or
is
it
kind
of
like
home,
where
you
basically
do
a
cli
and
then
you're
good
to
go,
or
do
you
need
like
a
side
car
in
your
your
cluster?
B
A
You
need
a
control
plane.
Okay,
this
is
yeah.
It
is
just
a
follow-up
that
have
permission
over
the
class
in
simple
way.
B
A
I
already
checked
that
in
and
sure
you
is
just
an
option
of
your
configuration
for
your
cluster
that
you
say
like
I
want
to
have
on
my
object
or
control
plane
or
I
want
to
enable
on
my
cluster
and
that's
all.
Oh.
B
B
Sasha,
what
were
you
going
to
say
or
did
did
sergio
steal
all
your
thunder.
C
But
I
just
wanted
to
say
you
know
we
have
to
use
a
a
helm
chart
that
has
a
dependent.
It's
like
a
there's
like
a
generic
count
chart
with
a
as
a
dependency.
B
C
Sir,
a
nested
home
chart,
that's
what
we
been
given
to
use
to
do
our
deployments
into
the
client
environment.
C
So
it's
got
like
a
version
and
name
and
description
and
all
that
and
then
underneath
the
dependency
on
the
generic
sort
of
microservice
home
chart
yeah
seems
pretty
pretty
good.
C
B
C
C
B
Yeah,
if
you
could
just
out
of
curiosity,
see
if
you
have
the
ability
to
use
operators
as
a
default.
A
I
think
operators
are
well
more
like
every
community.
Your
needs
cluster
or
cloud
is
going
to
do
something
similar.
What
already
red
hat
already
implements
on
on
on
results
by
default,
you
have
the
operator
a
control
plane
installed
and
you
have
the
like
an
optional
like
a
subsystem
or
inside
your
cluster.
That
has
a
connection
to
your
vector
hub.
A
So
pretty
much
you
open
your
your
your
operator,
menu
in
your
cluster
and
it's
connected
with
operator
hub,
and
you
do
like
you
just
you,
you
just
google,
what
you
need
oh
perfect
and
install
it.
So
you
know
yeah,
I
sure,
is
pretty
much
doing
the
same
and
I
think
for
sure
the
rest
of
the
people
is
going
to
do
the
the
same.
A
You
have
like
things
for
ibm
for
microsoft,
for
there
is
a
lot
of
vendors
that
already
have
his
operator
and
you,
and
and
by
default,
on
the
worst
case
that
you
have
like
around
a
community
or
of
something
it's
like
pretty
much
when
like
you're
going
to
docker
hub
is
like
the.
A
C
C
A
Chair
itself-
and
I
think,
a
c
plus
plus
something
like
that-
is
good
to
go.
So
at
the
end,
it's
a
box
with
the
automation
and
scripting
needed
to
deploy
something
in
a
in
a
in
a
standard
way.
But
the
ninth
thing
is:
he
managed
the
life
cycle
so
he's
receiving
like
the
updates.
A
So
you
only
you
go
to
your
operator
install
instance,
and
you
say
like
okay:
I
want
to
move
to
the
next
one
or,
if
you
have
like
a
by
default,
able
to
to
update
whenever
he
wants
he's
going
to
do
it.
So
that's
the
the
good
thing
that
is
is
connected
to
the
provider
and
your
application
is
leaving
connected
online
and
is
getting
updates
stuff
like
that.
B
C
A
It's
a
nice,
it's
a
nice.
B
Way
to
to
wrap
things
up
and
and
make
it
easy
to
one
stop
shop,
for
you
know
solutions
and
it's
really
tailored
around
solutions.
B
It's
very
similar
to
the
so
bitnami
did
something
very
similar
like
sergio
is
saying
used
to
be
around
vms,
where
you'd
be
able
to
download
a
whole
vm
for
wordpress
or
whatever,
and
you
could
get
up
and
going
and
then
they
moved
over
to
supporting
docker.
But
it's
the
same
same
concept:
a
one-stop
shop
and
install
the
database
install
you
know
a
load
balancer,
those
type
of
things
you
know,
and
everything
was
there
ready
to
go
now.
B
One
of
one
of
the
things
I'm
thinking
to
just
start
with
is
when
we
deployed
to
kubernetes
it's
just
running
inch
and
x
is
a
reverse
proxy
as
part
of
the.
So
you
just
have
your
inbound
into
ingress
into
that
reverse
proxy
engine
x
and
then
that
will
proxy
out
to
the
back
end
side.
Just
to
start
there
and
maybe
down
the
road,
have
an
option
where
we
have
a
istio
or
another
service
mesh
option.
B
As
part
of
that,
because
I
don't
know
how
many
people
are
using,
you
have
istio
or
link
or
d
all
configured
and
ready
to
go.
It's
much
easier.
D
B
C
Yesterday,
yeah,
we
use
it
because
it
wasn't
configured
properly
yeah.
A
Yeah
in
in
redhead
and
on
all
the
big
companies
are
doing
the
same
like
pretty
much
all
the
products
of
of
a
technology
company
right
now
that
works
on
kubernetes
is
is
working
with
operators.
A
And
at
this
point
I
am
not
doing
more
like
installations
like
follow
a
guy
for
installing
something
yeah
that
pretty
much
doesn't
doesn't
exist
right.
A
Okay,
maybe
well
usually,
I
might
be
liking
the
time,
but
I
think
if,
if
it
is
interesting
for
you
and
I'll
write,
how
that
I
can,
I
can
ask
for
a
q,
a
cluster
for
like
a
week
for
workshops
and
presentation
objectives.
So,
if
recording
I
can,
I
can
ask
for
this
cluster
and
we
can
do
a
tour
about
operators.
A
Director
have
some
like
new
cool.
We
can
use
git
ups
too,
to
check
how
it's
working
it's
more
like,
not
because
it's
red,
hat
and
you're
going
anything
we
we
shake
on
and
on
openshift
is
something
that
you
can
get
in
any
other
cuban
enterprise
queue
right.
A
So
if
you
want
to
check
it
over
there
at
least
to
see
what
are
the
futures-
and
we
can
do
that
totally-
I'm
not.
I
don't
have
much
time
to
prepare
things,
but
this
is
like
ordered
your
cluster
and
then
having
a
meeting
like
doing
a
tour
inside
the
cluster.
So
it's
not
going
to
take
time
for
me
more
than
the
presentation
itself.
Okay,.
B
C
A
A
Language,
no,
you
probably
you
need
to
go
to
your
cloud
provider
like
just
follow.
Eks.
A
A
B
They
may
have
like
a
prerequisites
on
operator
hub,
but.
A
At
least
in
alberta
to
have
I
I'm
I
inspected,
I
don't
know
I'm
expecting
that
if
you
go
to
the
operator
hub
site
and
you
select,
an
operator
probably
is
going
to
say,
okay
here
is
like
the
reference.
If
you
want
to
install
this
operator
over
some
clouds
right,
let
me
check
that.
I'm
going
to
see
that.
A
C
A
And
there
are
other
things
that,
like
the
operator
development
kit,
the
sdk
for
for
operators
is
a
good
page
at
the
beginning
is
complex
because
the
very
core
not
simple
you're,
doing
like
using
a
you,
your
vienna,
are
doing
development
automation.
So
it's
not
easy,
but
I
think
the
the
object
or
page
for
supporting
you
get
like
the
onboarding
on
on
the
development
on
operators.
I
think,
is
a
really
good
page
see.
Okay,
I
facilitate
a
lot
of
to
start
trying
out
in
implementing.
C
Bricks
aws
has
got
ack
is
an
open
source
project
built
with
with
love.
Okay,
sure
by
aws
aws
controllers,
for
kubernetes
is
that
it
is
that?
What
does
that?
That.
B
C
B
A
B
Yeah,
there's
there's
so
many
there's,
so
many
paths
that
go
down
here.
That's
why
I
wanted
to
sort
it
out
there,
which
way
you
know
just
because
there's
always
new
stuff
coming
out
that
I
just
totally
miss
you
know
so.
D
B
Some
of
the
last
things
I
kind
of
skipped
over
is
dude.
Karsh
was
asking
what
should
be
because
he
did
the
the
cyclone
for
python
and
we
focus
on
python
as
the
the
language
to
scan
for
cves
and
licenses.
B
Does
anybody
have
which
language
should
we
do
a
node.js
or
go
as
the
next
language,
or
do
something
like
at
the
os
level,
like
with
trivia?
You
know,
looking
at
the
rpms
and
packages
and
stuff
like
that,.
B
Well,
this
is
this
is
taking
a
this
is
taking
a
an
application,
a
container
that
has
a
go
application
in
it
and
we'd
scan
for
the
go,
laying
packages
to
see
if
they
have
any
cves
and
if
that's
what
so,
we're
really
not
writing.
Go
we're
scanning
and
implementing
like
cyclone.
That
has
a
a
go
scanner,
I'm
kind
of
leaning
towards
node.js,
since
no
since
js
is
a
pretty
big.
A
B
But
if
you
on
a
node.js,
if
you
do
a
hello
world,
you
end
up
with
1500
dependent
packages.
B
B
A
Yeah,
but
that
well
after
laughing
about
that,
it's
a
pain
because
usually
you
install
one
lead
and
it's
super
common-
that
the
conflict
between
libraries
like
this
library
by
behind
use
some
string
library
on
version
four,
this
other
library
used
the
same
thing,
but
unfortunately
six
and
yeah.
I
I
that's
one
like
like
the
few
things
I
don't
like
all
node.js
too
much
too
much
libraries
and
too
much
libraries
conflicts,
so
I
prefer
python
on
go.
A
I
think
it's
less
likely
that
you're
going
to
have
that
it's
more
simple,
so
well,
the
the
time
I
go
into
the
cyclone
dx
and
for
for
the
the
the
container
we
have
on
our
helios.
A
A
That's
what
like
my
my
feeling
about
it.
So
usually
I
I
will
say:
let's
try
to
keep
it
simple.
B
If
there's
other
open
source
tools
that
are
doing
scanning,
we
can
adopt
those
as
well.
I
haven't
found
that
many.
You
know
you
get
the
the
proprietary
ones
like
snick
or
black
duck.
You
know
those
those
are
the
proprietary
ones,
I'm
just
trying
to
stick
with
the
open
source
side.
To
start
with,
I
have
a
question
yeah.
A
Okay,
you
need
this
for
for
scanning,
but
on
my
head
is
I
usually
I
do
don't
do
that,
but
because
I
like
I,
just
use
my
registry
scan
my
image
and
that's
all
like.
Usually
I
don't
implement,
I
not
code,
like
utilities
inside
my
container,
that
scans
stuff,
I
just
let
it
to
the
image.
So
what
is
the
reason
to
to
have
like
an
inside
library
doing
that.
B
We
can
that
would
that's
one
of
the
other
options
that
we
need
to
look
at
is,
if
you
have
a
default
like
quay.
Does
the
default
scanning
of
your
image?
We
have
to
figure
out
if
there's
a
way,
to
grab
the
raw
data
from
that
scan
and
bring
it
into
ortelius.
So
what
we're
trying
to
do
is
consolidate
all
of
the
that
data
into
one
place
into
our
dashboard.
B
Okay.
So
if
quay
has
has
done
a
scan,
if
there's
a
restful
api,
we
can
call
on
the
quayside
to
just
download
the
raw
json
about
the
cves
and
licenses.
Now,
one
of
the
other
things
that
the
registry
scanners
will
do
is
like
trivial
will
is
a
example.
So,
on
your
github
actions,
trivi
will
go
ahead
and
scan
your
container
image,
but
it'll
only
look
at
rpms
and
your
packages
at
the
os
level.
B
It
won't
look
at
any
of
your
python
modules,
so
you
really
have
a
gap
there,
where
you
need
where
you're
gonna
have
vulnerabilities
in
your
python
modules,
but
not
at
the
os
level
in
the
container.
B
B
So
you
have
no
idea
what
the
license
consumption
is
for
that
container.
A
Yeah,
I
know
I
understand
the
only
thing
I
I
I
have
present
like
I
I'm
I
don't
know.
I
I'm
afraid
that
I
don't
want
to
put
like
complexity
on
the
delivery
side.
For
for
this,
this
scanning
needs,
like
I
would
like
to
have.
I
will
try
to
do.
I'm
thinking
lovely
like
in
this
case.
I
will
try
to
do
something
like
adding
a
image
layer.
A
A
Binaries
on
the
os
and
check
that
it's
okay
and
over
that,
we
should
have
like
a
new
image
that
they
care
about
the
next.
I
don't
know
the
condensing
cell
thing,
but
at
the
end,
hopefully
for
for
the
development
that
is
in
the
layer
of
of
the
image,
and
I
just
have
to
take
care
about
the
development.
B
Right
and
what
we're
doing
is
this
isn't
a
a
hard
requirement
to
use
ortelius
as
part
of
the
as
part
of
your
pipeline
to
go
ahead
and
do
the
scanning?
What
I
like
to
do
is
to
be
able
to
to
leverage
existing
scanning
tools
that
are
out
there
and,
like
I
said,
just
suck
that
data
in
somehow
or
point
to
where
that
data
exists.
B
I
don't
want
to
add
any
more
complexity
to
the
process,
but
that's
one
of
the
things
I'm
recognizing
is
there's
there's
hard,
it's
really
hard
to
get
the
information
out
of
what
you're
looking
for
you
know
for
unless,
like
I
said,
you
go
to
a
proprietary
even
then,
if
somebody's
using
like
the
linux
foundation,
I
believe,
will
they
have
a
contract
with
snick
to
go
ahead
and
scan
all
your
your
images
and
then
you
get
a
report
attached
to
your
git
repo
and
if
we
can
pull
that
in
you
know
where
proprietary
tools
doing
the
scanning
and
for
that
container
we
can
just
suck
that
in
and
make
the
association.
B
That
would
be
great
because
then
we
have
these.
You
know
tools
out
there
that
are
actually
that's
all
they're
designed
to
do
is
look
for
licenses
and
cves
and
we
can
bring
those
in.
That
would
be
my
ultimate
goal
because
I
hate
rewriting
stuff
and
I
don't
want
to
become
a
security
scanning.
You
know
project
I
just
want
to
you
know,
use
what's
out
there.
B
But
that's
that's
kind
of
the
goal
that
we're
trying
that
the
the
goal
from
the
service
catalog
perspective
is
to
show
for
this
component
version
that
corresponds
to
this
new
docker
image
that
it
has
these.
These
are
the
cves
that
exist
in
it,
and
these
are
the
licenses
that
it's
consuming
that's
kind
of
the
goal
at
that
level
at
the
high
level
and
how
we
get
that
information
and
and
where
we
get
it
from
that's.
That's
where
the
the
challenge
comes
into
play.
A
B
If
you
look
at
a
base
image,
that
would
be
like
a
trivia
where
the
base
image
will
go
through
and
make
sure
that
all
the
binaries
you
know
like
the
python
version,
the
url
lib,
some
of
the
open,
ssl
libraries-
don't
have
any
vulnerabilities
in
it.
But
when
you
then
put
on
top
of
a
python
program
that
brings
in
some
python
module
that
is
unsecure.
B
A
And
you
need
like
for
every
flavor
of
framework
and
stuff
like
that,
you're
going
to
need
a
different
thing,
so
it's
pretty
mess.
Well,
I
I
don't.
I
don't
have
idea
about
that,
so
I
will
talk
think
about
it.
People
about
the
the
community
of
practices
that
I
met.
So
maybe
I
can
get
some
good
idea
to
reuse
but
yeah
right
now.
No,
I
usually
don't
care
they
care
about
that.
B
All
right
guys,
we've
gone
over
a
little
bit.
Thank
you
so
much
we
are
going
to
meet
with
the
australian
group
this
afternoon.
My
time
4,
30
mountain
time
to
talk
about
more
about
the
get
ops
piece
so
you're
in
interested
in
that
drop
in
brad
and
ahmet
amid
have
been
doing
a
lot
of
work
around
that
around
argo.
B
So
it's
pulling
together
their
where
they're
at
so
they're
close
to
wrapping
up
their
research
and
gonna,
be
publishing
some
recommendations
on
our
next
step
for
integrating
ortulius
with
at
least
argo.
At
this
point,
okay,.