►
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
Awesome.
Okay,
so
I
was
wondering
if
I
can
do
a
very
quick
walkthrough
about
what's
happening
on
our
side
of
the
things
and
then
we
can
go
from
there.
So
let
me
just
share
my
screen.
A
So
can
you
see
my
screen?
Okay,
let
me
just
zoom
stuff
it's
here.
Let
me
just
push
it
okay,
so
for
us
in
protect,
we
we
use
the
cluster
management
applications
to
install
the
applications
are
familiar
with
the
cluster
management
applications.
A
No,
that's
fine!
So
there
is
on
this
repository
here,
cluster
application.
A
It
has
a
couple
of
applications
that
are
going
to
be
deployed
using
helm
charts
and
for
this
it's
required
integration
with
kubernetes
cluster.
So
what
happened
is
that
you
create
a
config
file.
I
have
an
example
here.
Let
me
just
open
this
locally
here.
A
This
is
the
template
that
you
have
to
add
in
the
cluster
management
project
and
if
I
show
you
here,
it's
a
very
simple
configuration
so
here,
basically
I'm
I'm
installing
psyllium
and
I'm
installing
ingress
and
then
after
I
run
this
in
the
pipeline,
it's
going
to
deploy
those
shards
home
charts
into
the
cluster.
A
So
from
there,
when
you
install
celium,
you
can
use
something
called
seal
network
policy.
It's
just
a
a
resource.
That's
going
to
be,
can
you
see
my
terminal?
Yes
awesome!
So
if
I
come
here-
and
I
do
api.
A
Those
are
the
resources
that
you
get
over
there,
but
basically
we
have
a
network
policy.
That's
going
to
to
be
to
you.
Can
you
can
set
configurations
between
the
communication
network
communication
between
the
pods
and
that's
what
we
were
doing
before,
so
we
have
the
cluster.
We
installed
the
ceiling
into
the
cluster
after
integrating
with
the
git
lab
and
then
after
deploying
after
running
the
pipeline,
you
have
cd
installed
in
your
cluster
and
then
the
other
feature
that
we
developed
is
that
we
can.
A
We
can
deploy
this
network
policies
from
the
ui
into
the
cluster,
but
that
that's
not
very
important
here.
The
important
step
is
that
we
would
like
using
the
agent
we
would
like
to
be
able
to
cross
information
between
what's
happening
in
cillian
and
also
with
the
information
related
to
the
network
policy.
So
then,
after
having
these
two
information,
we
can
generate
alerts
and
send
back
to
gitlab.
A
So
that's
the
that's
the
intrinsics
of
this.
So
I
have
something
to
show
you
here
so,
if
just
to
give
like
a
overall
picture
of
a
situation
of
this
cluster
here,
sydney
has
a
this
operator.
Here
is
not
important.
A
Sedan
has
the
the
spot
that
that
basically
has
ceiling,
binaries
and
hubble
hubble
is
going
to
be
used
to
monitor
ceiling,
and
this
hubble
relay
here
is
just
a
way
that
sydney
provides
a
centralized
way
of
integrating
all
the
hubble
surfaces
that
are
running
in
the
cluster
in
case
there
is
multiple
nodes
and
then
yeah
and
then
on
the
served
services
side
you
let
me
know
if
it
gets
confused
or
too
fast
or
if
I'm
talking
about
something
that
you
know
already
just
let
me
know,
and
I'm
glad
to
so
here,
the
services
that
we
are
concerned
right
now
is
the
hubble
relay.
A
A
So
then,
what
happened
is
that
hubble
relay
exposed
gerp
grpc
server,
and
then,
if
I
show
you
here
a
basic,
I
played
a
little
bit
with
the
agent
code
and
basically,
what
I'm
doing
right
now
is
that
I'm
just
accessing
this
ga
rpc
code
from
the
agent
and
I'm
I'm
just
reading
that
the
flows
the
flow
for
siemens
is
going
to
be
a
connection
that
went
through
one
end
point
to
another
one.
A
So
if
I
run
here,
I
just
I'm
just
looking
through
the
flows
now,
there's
a
couple
of
them
and
of
course
I
didn't
put
the
code
in
the
best
place.
It's
just
for
me
to
explore
how
to
use
the
agent.
A
So
then
this
is
the
first
step.
The
other
step
is
that
I
can
show
you
the.
B
Rng,
just
to
read
the
rate,
I
understand
so
your
modifications
to
the
agent,
the
code
that
you
added
it
just
calls
the
jrc
endpoint
and
gets
the
data
from
it
and
just
prints
it
on
the
screen.
Exactly.
A
Exactly
yeah
in
the
future,
the
idea
would
be
to
to
match
this
information
that
I'm
having
right
now
with
the
network
policy.
It's
going
to
be
on
other
end
points
that
that
ceo
provide
and
then
after
matching
them.
The
idea
for
me,
as
far
as
understood
of
the
architecture,
would
be
to
send
this
information
back
to
cass
and
then
cass
would
make
a
call
to
the
internal
api
and
get
lab,
and
we
would
have
the
alert
saved
over
there.
B
A
So
so
this
is
basically
like
the
the
very
of
course
that's
not
it's
not
even
in
the
proper
place,
I
just
add
under
refresh
configuration.
I
just
add
this
apply
ceiling
configuration
here,
and
this
is
going
to
basically
just
go
through
it
and,
of
course
this
is
hard
coded
here.
When
we
are
running
inside
the
cluster,
we
can
grab
this
information
or
use
the
the
kubernetes
syntax.
I'm
not
sure
how.
How
would
be
the
best
approach
for
doing
this
for
the
for
the
policies?
A
Just
things
that
take
a
little
bit
of
time
is
that
I
we
need
to
play
sometimes
with
the
files
from
cedium,
and
then
we
need
to
grab
the
endpoints
that
they
provide
already,
instead
of
relying
in
the
default
and
the
default
apis
that
are
provided
by
kubernetes.
So
that's
a
little
bit
of
trade-off
that
takes
time
some
time
to
read
their
source
code
and
get
together
in
this,
and
with
that
said,
I
have
no
experience
with
golang.
B
That
yeah
oh
visit
this
three.
Is
it
a
stream
response?
No,
it's
a
single
single
response.
Right
flow
client,
no
okay,
it's
flow
client
receive,
but
then
you
are
iterating
on
the
status
object
and
what
is
the
status.
A
B
A
And
again,
like
this
code
can
change
completely
in
terms
of
architecture.
My
goal
was
to
show
you
a
little
bit
of
what
we're
looking
for.
So
then
we
can,
because
I
saw
that
you
you,
you
commented
on
the
plugin
architecture
that
you
were
developing
for
the
agent
k
and
then
I
thought
that
if
I
gave
you
like
a
walkthrough
like
this,
it's
it's
easier
to
get
the
use
case
from
our
side
of
the
fence.
B
A
There
is
one
thing
here
this
one
here:
we
don't
care
much.
I
show
you
that
after
we
run
the
pipeline
for
cluster
management
thing,
we
have
a
couple
of
documentations
here
for
how
to
customize
the
custom
management
application
here.
But
what
happens
that
this
project
is
going?
I
think
it's
going
to
be
deprecated
in
the
future,
so,
but
that
that's
a
problem
for
another
day
right
now,
I'm
glad
that
we
we
are
going
through
this.
So
should
I
stop
my
sharing
my
screen.
A
Okay,
okay,
awesome
so
yeah.
So
now
I
would
like
to
know
from
you.
No
pressure,
of
course,
is
that
did
it
give
you
a
good
idea
on
how
to
fit
that
in
the
plug-in
idea
that
we're
having
before.
B
Yes,
I
think
it
fits
really
nicely,
and
the
only
missing
piece
that
you
will
need
is
the
method
to
actually
make
the
request
to
cast
so,
but
everything
else
is
in
place
like
the
structure
the
api
is
in
place.
You
can
start.
B
A
A
So
I'm
gonna
stop
the
recording
now
so
then
we
can
talk
about
random
stuff,
cool.