►
From YouTube: PSGraph: Creating Dynamic Graphs and Diagrams on the fly with Kevin Marquette December 5th, 2018
Description
Imagine being able to dynamically create diagrams of your infrastructure, business processes or workflows! All on the fly!
Let’s build some graphs with Powershell and PSGraph! Kevin Marquette will be presenting his tool for building and presenting graph data on demand using nothing more than PowerShell!
Let's be honest, some things are just hard to see in tables of data and are much easier to understand when visualized. The problem with most diagrams is that they are out of date as soon as they are created.
Kevin will show you through the use of his toolkit how you can keep your documentation, your workflows, and your business processes up to date via dynamically created content that runs from the command line!
Resources:
Check out the content for this presentation at Kevin's Github: https://github.com/KevinMarquette/PSGraphPresentation
Content from presentations can also be found at our github:
https://github.com/rtpsug/Meetups
Watch our previous meeting about AD and Regex:
https://youtu.be/xqjIUH0zwvE
SUBSCRIBE to learn about PowerShell!
https://www.youtube.com/channel/UCiTuB6HZiupE8CJ71vkMFnA/example?sub_confirmation=1
LET'S CONNECT!
-- https://twitter.com/rtpsug
-- https://www.meetup.com/Research-Triangle-PowerShell-Users-Group/
A
A
B
A
So
he
has
brass.
Is
it
supposed
to
let
us
build
graphs
with
using
just
PowerShell
and
I've
always
been
a
fan
of
Visio
I've.
Just
never
been
a
fan
of
automating
Visio
like
I've,
never
found
the
right
framework,
but
let
me
build
the
diagrams
that
I
really
wanted
to
build,
and
so
through
my
hunt,
I
found
an
open
source
module
or
program
called
grasp.
A
Is
ahead,
its
own
text-based
language
for
describing
graphs
and
good,
generate
images
of
various
formats
and
I
have
been
writing
little
wrapper
scripts
around
it
for
a
while
and
eventually
formalize
it
into
an
actual
module
here.
So
a
long
fade
me
saying
that
my
PS
graph
is
really
just
a
front-end.
Affirming
interface
to
the
graph
is
engine,
and
if
you
need
more
in-depth
details
on
topics,
sometimes
the
the
reference
information
off
their
website
will
translate
very
very
closely
to
what
I've
done
with
with
with
PS
graph.
A
So
when,
in
my
demo,
script
I've
got
the
first
couple
steps
to
bootstrap
your
system.
However,
if
you're
using
chocolaty,
you
can
do
the
find
package
and
install
graph
is
off
the
repo
if
you're
on
Linux
or
Mac
I'm.
Sure
there's
some
others.
If
you
download
and
load
into
your
default
locations,
the
my
module
should
auto
discover
it.
A
A
Then
I'm
going
to
walk
through
each
of
these,
but
this
module
is
implemented
as
a
DSL
write,
a
domain-specific
language,
so
I
use
commands
that
are
very
specific
to
what
you'd
use
for
generating
a
graph
like
a
graph
and
edges
and
nose
I'll
kind
of
explain
what
these
mean
as
we
go
through
them.
But
the
command
lips
is
actually
fairly
short
here
and
we
can
do
quite
a
bit
with
that.
A
Let's
grab
the
most
basic
of
graphs,
so
you'll
notice,
I
use
like
a
castor
like
syntax,
where
I
define
a
graph
as
a
script
block
that
runs
a
couple
commands
and
we're
going
to
make
a
node
named
home.
A
node
named
work
we're
going
to
draw
one
line
between
the
tool
we
pipe.
The
output
to
my
export,
PS
graph
and
I
can
save
it
to
any
location
that
I
want.
So
let's
go
ahead
and
first
generate
our
graph.
A
A
Now
the
nice
thing
about
graph
is
is
I
only
need
to
call
it
a
node
if
I'm
doing
something
special
with
it.
So
if
I'm
just
drawing
edges
lines
from
nodes
it
realizes,
it
needs
to
put
the
note
on
the
graph.
So
here's
a
case
where
I
basically
draw
an
edge
from
home
to
work
and
then
work
to
home
and
I
use
a
shorter
syntax
here
with
just
show
graph
to
save
the
step
of
saving
it
to
a
file
so
running
this
one
and.
B
B
A
B
A
I
feel
that
I
clarify
the
definition
right,
so
my
in
for
this
module
there
base
like
node
edge
based
directional
graphs
yeah.
So
it's
not
like
the
bar
charts
or
whatnot
you
get
out
of
Excel,
but
more
I'm,
defining
relationships
between
items
and
then
graphing
them
visually
I
will
step
into
some
like
concrete
examples
of
doing.
You
know,
like
processes
and
dependencies
or
like
network
connections
as
to
get
a
good
visual
of
what
you
can
do.
I
want
to
make
sure
we
have
like
so
long.
I
start
with,
like
my
basic
syntax.
A
B
B
A
Graph
is
that
just
shows
the
types
of
graphs
that
could
be
generated.
You
know
get
like
state
machines,
org
charts.
You
can
do
here's
I
suppose.
Here's
the
models
that
is
supported
by
graph
is
that
are
non
infrastructure,
related,
I,
guess
and
my
examples
that
I'll
dive
into
will
hit
more
closer
to
home.
A
A
Take
a
peek
at
this
edge
so
for
drawing
edges
now,
I
can
I
draw
from
individual
nodes.
I
can
give
it
an
array
or
collection
of
nodes,
and
in
this
example,
it
would
draw
an
edge
from
A
to
B
to
C
to
D
and
then
back
to
a
so.
It
kind
of
gets
own
a
little
circle
there
and
and
connect
the
connections,
the
collection
serially.
A
A
We've
mixed
up
prettier
by
using
attributes
so
like
I
support
a
really
short
syntax,
so
I
can
actually
skip
to
in
the
from
so.
If
I
have
a
first
node
in
a
second
node
and
it
will
do
the
lines
to
them
but
from
to
or
if
I
specify,
attributes
I
can
give
it
the
edge
itself
a
custom
label
or
give
it
colors
and,
as
you
make
heavy
use
of
positional
parameter.
So
while,
if
you
love
being
more
verbose
or
like
the
shorter
syntax
of
just
specifying
your
attributes,
both
are
handled.
A
It's
generally
the
name
of
it
and
you
draw
edges
between
those
unique
identifiers,
but
sometimes
you
may
actually
want
to
have
beginning
identifiers
for
the
note
itself,
but
give
it
a
nice
friendly
and
label
when
presenting
it
to
the
end-user,
so
I'm,
quite
often
using
labels
to
give
it
custom
text
and
changing
the
shape
to
adjust
the
way
the
way
it
looks
so
in
this
case
here
I'm
going
to
add
the
my
house
label
to
home
and
I'm
gonna
label
work
as
loan
Depot
and
then
later
on.
I
can
create
my
edges.
A
A
A
A
You
can
create
zones
where
you
can
actually
click
on
links
and
it'll
it'll
interpret
those
correctly
so
tons
of
options
built
into
it,
and
if
you
ever
need
to
to
see
the
full
list
you
can
actually
go
to
the
graph
is
documentation,
documentation
directly,
which
I
have
linked
in
here
as
I
scroll
to
this
giant
list
like
there's,
it
feels
almost
endless
a
number
of
things
that
can
be
customized
as
far
as
for
edges
and
nodes.
They
can
give.
B
B
A
A
But
if
I,
but
the
important
thing
to
note,
is
it's
like
top-down
processing,
so
it's
more
like
I
have
nodes
without
attributes
and
I
want
to
change
the
default
halfway
through
the
default
will
only
apply
to
the
stuff
further
down
in
the
list,
so
it
processes
stuff
in
order
that
and
as
those
as
it
goes
along.
So
if
I
run
this
one
here,
we'll
see
that
the
first
node
home
should
still
be
an
oval
or
circle,
because
this
node
default
doesn't
come
into
place
until
halfway
through
the
script.
B
A
B
B
A
On
a
small
tangent
here
so
I
do
score.
A
good
range
of
the
output
I
mean
I
may
say
these
passing
parameters
back
to
a
graph
is
on
what
formats
it's
of
course,
and
so
Rapids
has
been
a
really
robust
application.
That's
been
around
for
quite
a
while,
so
they
have
added
quite
a
bit
of
features
to
it,
not
Windows
or
space.
B
A
My
quicklist
here
is
I
that
I
know
that
added
math
auto-completion
mappings
for
JPEG
PNG
gifts,
it's
actually
like
the
IMAP
and
I'll
do
a
demo
at
the
end.
We're
actually
you
create
some
clickable
images.
Post
court,
PDF
playing
SVG
I,
know
graph
is
supports
many
more
than
this.
So
if
you
find
one
that
you
need,
it's
just
a
pull
request
away
for
me
to
add
it
to
the
foundation
set
for
the
most
part.
A
A
So,
while
I'm
making
very
friendly
edges
and
nodes,
you
know
to
me
it's
right.
This
is
basically
just
a
hash
table
and
whatever
Keys
you
put
in
there
I'm
going
to
pass
them
to
graph
it
graph
is,
as
you
present
them.
So
if
you
do
something
that
it
doesn't
support
graph
is,
will
come
back
and
say:
I,
don't
support
this
key,
but
that
also
means,
if
they
have
our
update
graph,
is
and
add
new
attributes,
I
kind
of
gain
support
that
automatically
like
I'm
I'm,
trying
not
to
filter
it
in
the
way
of
somebody.
B
So
Kevin
there's
a
few
guys
that
are
asking
about
and
I
kind
of
I'm
foreshadowing
I'm,
better
you're,
going
to
tell
us
just
wait
and
see,
but
some
guys
are
asking
about,
like
you
have
a
success
story
that
you
that
you
can
speak
about
that
like
help,
sell
this
to
management
or
anything
like
that
and
I
have
a
feeling.
You
probably
have
something
like
that
in
later
demos,
I
don't.
A
B
A
A
So
mm
just
points
that
were
like
duplicating
nodes
right,
so
I
mentioned
that
the
name
is
special
and
if
I
want
to
have
a
note
on
there
more
than
once
like
in
this
case,
I'm
drawing
multiple
edges
and
I'm
gonna
get
this
nice
graph
that
shows
from
home
to
work
for
work
to
lunches
back
to
home.
This
is
the
circle
bit
I
might've
mentioned
want
wanted
to
have
work,
to
have
be
like
the
circle.
Instead
have
them
show
up
twice
and
I
mean
the
short
answer
to
make
that
happen
is
well.
A
A
A
A
Ok,
so
that's
it
for,
like
general,
stacks
and
I,
want
to
step
into
something
might
be
getting
a
little
closer
to
home
right.
So
here's
an
example:
the
super
basic
of
doing
a
look
at
server
infrastructure,
if
you
will
right
like
I,
have
a
diagram
where
I
want
a
load.
Balancer
that
has
a
connection
to
the
website.
A
Website
connects
to
the
API
and
the
API
connect
to
database.
So
it's
almost
like
a
you
know
like
a
multi-tier
application,
so
they
run
this.
You
know
I
see
exactly
what
I
described
and
that's
where,
as
missing
and
Visio
was
be
able
to
describe
what
I
wanted
using
a
simple
syntax
and
actually
getting
it
to
generate.
Whereas
looking
for
and
continuing
with
this,
you
know
what,
if
we
wanted
to
add
servers
to
lists
right,
so
I'm
ants
are
adding
arrays
for
my
visit
nodes
and
edges.
A
So
here's
the
case
where
maybe
my
server
name
is
web
1
and
web
2.
Where
I
have
two
servers,
let
me
rank
them
together
and
they
have
my
load
balancer
connect
to
my
web
server
list.
My
ap
is
my
website's
connect.
My
AP
is-
and
my
API
is
connect
to
my
database-
all
right
so
I
run
this
we'll
see
all
of
those
nodes
that
cross
link
and
look
at
each
other
or
come
to
each
other,
and
this
is
where
one,
where
some
of
the
features
I've
added
two
PS
graph,
start
giving
you
value
over.
A
What
graph
is,
does
natively
traditionally
you'd
have
to
draw
it?
Well,
you
have
to
mark
out
each
of
those
edges
that
you'd
want
shown
where
I
know
in
this
scenario,
here
this
edge
I'm
saying
that
I
want
I
want
to
try
cross
multiply
these.
If
I
could,
where
everything
here
should
connect
to
everything
here
and
it's
going
to
do
one,
they
go
one
two
and
three
and
web
two,
two
one
two
and
three,
and
it
generates
that
list
for
us,
like
the
raw
syntax.
A
A
So
now
I
can
use
variable
to
reference
these
items.
Web
servers,
api's
and
databases
and
I
go
back
to
my
graph
by
PS
graph
diagram
and
I
can
go
back
almost
that
first
first,
in
fact,
I'm
saying
you
know,
my
node
for
web
servers
connect.
My
load
balancer
to
my
web
servers,
connect
web
servers
to
API
servers,
it's
very
natural
language,
descriptive
of
what
I
want
to
generate,
and
it
is
the
exact
same
graph
as
before.
I
believe.
A
But
I
had
a
few
more
nodes
and
now
that
I'm
pulling
it
from
an
array,
I
can
quickly
add
a
fourth
and
a
fifth
website,
or
you
know
six
or
seven.
You
know
API
to
list
or
we
just
make
it
all
data-driven
right
and
that's
where
you
start
to
get
real
value
out
of
the
module
like
this
is
I'm
not
going
too
hard
to
put
that
list.
I'm
gonna
go
pull
for
my
my
my
source
of
truth
right
so
in
this
case
I'm
pulling
from
a
CSV,
but
this
could
be
an
Active
Directory
groups.
A
A
A
A
A
And
then
adding
IP
addresses
right.
I
have
IP
information.
Let's
get
that
on
the
diagram,
so
the
one
change
I
make
is
at
the
very
bottom
here.
I
move
over
all
the
servers.
I
create
a
node
reference
like
I,
say
for
this
node
and
I'm
using
the
computer
name.
As
my
unique
identifier,
I,
add
a
label
I'm
eligible.
This
shows
both
the
computer
name
and
then
the
IP
address
the
slash
end
is
to
make
it
a
new
line.
So
it
breaks
it
up
into
two
lines
on
this
diagram
so
running
this.
A
Same
set
of
data
we're
just
that
extra
line
if
you
are
crowded
but
to
see
IP
addresses
on
a
siren
or
any
other
information,
you
want
to
present
right.
Look
I'm
just
doing
IP
addresses
as
the
quick
demo,
but
you
could
do
small
tables
in
these,
like
that.
Html
is
supported
because
it's
dynamically
presenting
this
information
I
think.
B
A
B
So,
but
if
you're
playing
with
this-
and
you
can
do
that-
and
you
would
use
this-
you
could
just
import
it
as
an
attribute
yep
and
then
and
what
the
absolute
you
looking
for
is
you
do
reference
out
back
to
graph
is
in
order
to
get
that
data?
Absolutely
okay.
So
if
we
wanted
to
do
it,
we
could
do
it,
but
we
just
pass
it
back
out
the
Brassica
Krakus
there's.
A
Also
different
layout
engines,
I
think
there's
like
circular
so
graph
is
itself.
Has
multiple
layout
engines
the
default
one.
Does
a
hierarchy
and
I
mean
this
one
makes
sense
to
do
a
hierarchy
but
I
often
as
my
graphs
get
large
I,
especially
with
with
nodes
that
reference
multiple
nodes,
like
almost
like
circular
type
references
I,
will
lock
my
list
of
layout
engine,
so
I
find
the
one
that
said
moving
the
best
look
spring
model
small/medium
for
the
most
part,
the
default
one
works
the
best
in
most
cases,
but
there's
an
exit.
B
A
A
A
A
A
B
A
So,
like
those
aligned
with
stuff
I
probably
be
doing
in
video,
I
saw
the
other
classic
example
is
the
organizational
chart,
and
this
is
the
scenario
where
you
have
a
set
of
data
that
defines
its
own
relationship
right
before
I
was
defining
roles.
You
know
web
talks,
API
and
API
dr.
database
pretty
hard
coded
in
the
graph,
but
in
this
example
here
I
might
just
have
an
organizational
chart.
A
A
A
So
let's
run
this
graph
here-
and
this
is
the
first
truly
data-driven
diagram
and
I
can
see
all
the
reports
how
it
flows
dynamically
and
automatically.
So
you
know,
maybe
it's
your
Active
Directory,
you
could
be
walking
the
shows.
You
know,
managers
and
reports.
This
one
here
is
pretty
like
top
level.
Heavy
I
can't
see,
isn't
enough
staff
or
management
and
I
can
see
that
now
that's
what
I
love
about
diagramming
and
visualizing
stuff
is
it.
A
B
A
Do
you
have
an
alternate
syntax,
so
I'm
going
to
keep
the
same
list
of
people
I
found?
This
pattern
was
so
common
in
all
my
graphs
and
charts
that
I
made
a
a
simpler
syntax
for
me
for
handling
this
scenario.
So
this
same
thing,
I,
actually
support
giving
node
the
full
list
of
objects
and
providing
a
couple
script
locks
for
it
to
pick
out
what
should
be
the
ID
and
then
on
the
label,
giving
a
script
block
for
to
process
for
each
node,
how
to
label
it
and
the
exact
same
thing.
A
A
A
So
I
run
this.
You
know
the
table
just
flew
by,
but,
as
you
can
probably
already
guess,
let
me
use
this
is
the
node
name,
the
ID,
the
parent
ID
is
a
relationship
and
then
give
the
label
of
the
process
name
and
I'm,
using
that
newer
syntax
here
so
I'm
going
to
say
in
this
case,
left
to
right.
I
felt
like
this
looks
better.
That
way,
give
me
the
shape
of
a
box.
A
The
node
for
every
process
use
the
process
ID
as
the
node
name
and
then
for
each
attribute
the
process
name
and
then
my
edge
below
it
do
the
same
thing
and
map
those
properties.
So
now,
when
we
run
this
guy,
we
get
a
diagram
here
of
these
all
the
processes.
On
my
system,
let's
zoom
in
on
something
like
oh
yeah,
here's
vs
code
I
can
see
it's
calling
vs
code,
this
loading,
vs
code
and
I
got
PowerShell
running
in
it.
I
can
scroll
up
to
something
else,
Chrome
how
about
the
chrome
tabs
open.
A
A
A
Is
my
whole
net
set
object
I'm
using
my
like
the
full
collection
of
objects
defining
from
two
a
my
attributes,
this
time,
I'm
actually
going
to
show
the
local
port
and
the
remote
port
that's
connected
to
my
system.
So
this
one
now
I
run
this
and
look
at
that
I've
got
a
live
diagram
of
all
the
systems.
I'm
connected
to
I
can
see
here
whatever
system.
This
is
I'm
connected
to
it
twice
like
there's
two
different
entries
and
it
drew
a
line
for
each
one
of
them.
A
A
B
B
A
So
and
the
thing
of
it
is
you're
only
bound
by
your
creativity
right
then
that's
abuse,
ISM
is
all
really
doing
a
thing.
I
have
a
node
I'm
connecting
to
things
you
can
just
imagine
saying
so
you
can
connect
and
relate
to
each
other.
That's
a
diagram
for
you.
Can
they
can
create
I
took
awhile
to
find
these
demos
so
I'm
like
okay?
What's
a
good
repeatable
demo
that
I
can
do
that's
live
and
just
these
attributes
of
the
local
system
are
beautiful.
A
A
A
A
So
when
this
one
runs,
this
is
now
the
process
is
on
my
system
and
what
they're
talking
to
so
I
can
see.
I
got
you
know
three
tensions
from
link.
You
know
silence
it's
knocking
this
server
Chrome.
These
are
the
sites
I
five
open,
my
browser
and
again
I,
trying
to
figure
out
what
processes
is
talking
to
what
remote
system
over
what
ports
mom
like
that
for
troubleshooting.
This
is
like
a
dream.
Come
true,
really
and
then
reflect
on
that.
You
know
it's.
So
the
diagram
doesn't
look
any
more
complicated
before
so.
A
I
walk,
Oh,
I'm
jumping
here,
they're
gonna
jump
give
each
node
an
ID.
You
can
give
the
buffest
name,
which
is
pretty
standard
and
then
I
walk
the
net
set
list
to
map
the
owning
process
to
the
remote
address
instead
of
the
local
address,
there's
time
and
because
the
own
process
ID
and
this
ID
are-
are
the
primary
keys,
fill
the
Lowe's
map
together
and
that's
really
what
you
want
to
find.
We
want
to
think
about
what
you
want.
A
A
A
Maybe
I
just
have
to
just
destroy
it
systems
that
use
a
different
syntax
for
the
same
thing
you
know
like
I
might
have
load
balancers
to
say
I'm
in
this
network
segment,
but
my
servers
are
in
this
data
center
and
I.
Could
you
can
use
like
a
hash
table
to
like
fill
that
mapping
and
what
I
mean
by
that
is
the
from
scripts
on
this
I?
Can
I
should
say,
like
okay
I
have
a
map
if
I
give
it
an
owning
process,
it
will
give
me
the
value.
A
B
A
So
that's
that's
an
option.
If
that's
the
type
of
data
you
want
to
see
and
present
you
can,
but
the
nice
thing
about
services
well
I
started
down
this
path
is
that
they
do
define
who
they
depend
on
so
doing
another
hierarchal
example
here
I'm
going
to
do,
get
all
my
services
and
I'm
going
to
map
who
we're
getting
more
more
creative.
Here,
all
my
services
using
the
node
name,
the
label
is
the
display
name,
this
time,
I'm,
giving
it
a
color
based
off
the
status.
A
If
it's
running
or
not,
blue
blue
will
be
running,
red
will
not
be
running
and
then
my
link
services
I,
am
finding
the
services
that
have
it
depends
on
then
I
draw
edges
for
each
of
them,
using
the
from
name
to
service,
depends
on
dot
name
and
like
I'm
still
using
a
heretical
way
out
here.
This
is
the
defaults.
I
know
why
I
have
it
in
here.
Maybe
I
was
experimenting,
trying
to
find
a
better
view.
I'll
just
trim
that
off.
If
I
run
this
now,
we
will
show
the
diagram
of
all
the
services.
A
That
so
much
depends
on
remote
procedure
call
and
in
this
one,
I
like
this
diagram
too,
because
it
also
lets
me
show
one
more
gotcha
with
leveraging
graph
is,
is
that
the
IDs
are
case
sensitive.
So,
if
you're
the
case
where
some
are
saying
where
a
procedure
call
in
all
lowercase
right,
we
depend
on
and
then
I
have
another
spot
here.
A
I'm
pretty
sure
is
on
the
same
chart
where
they
decided
to
use
all
caps,
and
so
by
default,
it's
case
sensitive,
but
I
do
have
a
way
to
handle
this
scenario,
because
I
knew
that
I
run
into
various
node
formatting
scenarios,
I
added
what
if
I
click
a
custom,
node
format.
Scripts,
if
case
sensitivity,
is
your
problem,
you
do
to
lower
to
lower
case
everything
for
the
node
ID.
A
So
let
me
run
this
to
a
quick
example.
If
I
run
this
now,
like
actually
manage
from
this
without
first
show
my
problem
set.
So
here,
I've
got
my
cases
missed
mismatched
right,
like
I
hand,
side
the
eye
fat
finger,
that
it
critical
nodes,
so
I
actually
want
this
to
be
one
node.
So
I
add
the
two
lower
on
here.
A
This
is
important
if
you
ever
have
special
characters
in
your
ID.
That
graph
is
tends
to
care
about
and
complain
about.
You
could
at
a
custom
mode
formats
grip
to
do
like
a
substitution
for
the
ID
to
still
make
it
unique
and
use
what
you
want,
but
if
you
have
you're
using
like
I
know,
quotes
or
colon
actually
is
a
weird
one
that
sit
sore
smug
laugh
anyway.
There's
options
just
be
aware
of
to
tweak
that
so
doing
this
full
graph
now
setting
the
node
form
a
script
to
lower
with
my
same
dependent
services.
A
We
should
now
get
everything
one
instance
of
remote
RPC
and
they
shall
point
to
each
other.
So
now,
when
I
zoom
in
to
get
EMS
and
it
renders
for
me
everything
is
hide
in
here
and
I.
Don't
have
those
hanging
references
like
I
had
before
any
of
them.
I
do
have.
That
are
hanging
happens
to
be
something
that
says.
I
depend
on
this,
but
I
actually
have
a
service
with
that
name
to
get
a
label
on
it.
A
A
Just
so
I
love
seeing
stuff
visually,
because
there
will
be
things
you
just
will
not
notice
or
ever
see
until
you
do
present
it
in
a
visual
nature,
and
this
is
one
of
those.
So
basically
what
I
just
did
is
I
walked.
All
the
help
is
actually,
if
I
do
a
get
help
on
something
you
know
get
help
help
on
get
help.
A
A
Why
you
see-
and
you
see
stuff
cluster
together,
like
you,
see
stuff,
that
that's
parked
the
same
family
of
commands,
get
service
to
spend
service
I
didn't
realize.
That
was
a
thing
there
before
stop
service
step
service
and
then
the
fun
one
is
the
bug
that
you
find
by
zoom
in
here
to
get
content.
I
notice
it
actually
reference
itself
twice
like
is
that
kind
of
weird?
So
if
actually
look
at
get
help
on
the
get
content,
no,
not
that
one
get
help
on
get
content
and
I.
Look
at
this
list
here.
A
B
A
A
Trade
borders
like
there
is
a
wealth
of
setting
from
the
attributes,
color
gradients,
these
HTML
colors
versus
the
string
friendly
names.
Here's
all
the
ends
you
can
use.
So
if
you
need
an
empty
versus
wrangle,
whatever
you
want
and
there's
lots
of
documentation
for
somebody
to
work
through,
if
they
really
want
to
do
some
cool
stuff,
there's
a
little
bit
above
down
to
you,
can
tweak
how
the
lines
they
could
be
more
right.
Angles
versus
loopy.
A
A
A
A
A
A
A
Yet
so
I'm
not
saying
this
is
the
best
way
to
diagram
your
get
history
but
more
showing
what's
possible
and
I
teach
you
guys
with
this
before
we
can
create
HTML
links,
and
this
is
a
scenario
where
I'm
going
to
add
a
URL
to
this
node
that
basically
prefixes
it
with
you
know
the
commit
path
if
I
inject
in
the
commit
ID
I'm
github
will
take
me
to
that
commit
right.
So,
if
I
adjust
this.
B
A
Run
this
whole
thing
here:
actually
this
first
time,
I
save
to
a
variable
I'm,
actually
generate
two
graphs.
One
is
a
PNG
file
to
show
the
image.
The
second
is
a
see
map,
X
I,
don't
do
it.
I
had
no
idea
what
that
was,
but
it's
basically
like
a
a
a
mapping
showing
hot
zones
over
an
image.
So
if
I
just
generate
this
map,
X
file
and
then
I
actually
do
a
little
of
HTML
magic
to
actually
just
create
an
HTML
file
that
contains
those
things.
A
A
So
this
could
be
like
I'm.
Just
doing
you
know
get
chemist
here.
This
could
be
your
server
dashboard.
That
goes
to
your
performance
page
or
you
know
it's
like
they
give
you
a
health
check,
something
simple,
but
when
you
click
on
it
it
can
give
you
you
know
some
webpage
you
already
have,
especially
if
you
have
like
a
like
rest
site
or
like
oh,
the
new
admin,
Center
anyways,
but
it
even
craft
a
URL
to
go
to
something
automatically
and
inject
that
in
here.
A
A
Well,
that's
that
takes
me
as
far
as
I'm
going
to
take
PS
graph
I
do
need
to
plug
a
second
module
of
mine,
which
is
PS
graph
plus,
which
has
a
couple
handy
commands
that
take
a
lot
of
the
demos
I
did
tonight
and
put
them
into
commands
that
you
can
run
as
needed.
So
if
you
feel
like
you
need
to
see
the
process,
the
eld,
what's
connecting
to
what
you
can
actually
just
run
this
command
and
Indic
those
settings,
but
but
wait,
there's
more
I
actually
have
support
for
targeting
a
remote
computer.
A
So
we
actually
put
our
name
in
here.
It
will
create
a
PS
session
to
that
computer
run
what
it
needs
to
pull
the
data
back
and
generate
the
graph
locally,
and
you
might
also
notice
this
supports
an
array,
so
I
can
actually
give
it
a
list
of
computer
names
in
one
graph
see
you
know
what
these
five
servers.
What
are
they
talking
to?
A
So
I
got
a
little
bit
more
polish
on
these.
My
get
graph
as
well.
I
don't
have
the
links
in
this
one,
but
as
I
run,
this
one
actually
come
back
to
that
one
first,
let
me
do
by
so
by
defaults.
I
show
like
just
the
commit
IDs
and
the
branches,
but
I've
added
some
parameters
that
lets
you
put
the
show
to
get
commit
the
get
commit
message
and
the
direction
and
I've
used
this
quite
a
bit
to
troubleshoot
boots.
A
You
know
some
scenarios
that
I've
gotten
into
so
here's
a
scenario
where,
where
I'm
doing
a
dotted
line
to
a
commit
to
reference
like
this,
you
just
like
more
of
a
comment
about
that.
I'm
keeping
these
on
the
same
level
by
using
a
rank
here
like
this
message,
and
this
node
are
ranked
the
same:
to
keep
them
side
by
side.
Otherwise,
they'd
be
floating
off
in
different
different
directions
and
wouldn't
keep
that
relationship
as
tight
as
this
is.
B
A
With
that
and
I
have
a
simple
one
here
that
shows
the
ast
command
graph
like
if
I
have
a
module
and
I
want
to
see
what
commands
it
calls
this
guy
parses
ast
and
shows
me
what
calls
was
so
you
know
my
graph
command
inside
of
it
we'll
call
these
convert
to
or
get
indents
and
I
can
just
see.
I
got
a
clustering
here
of
my
grep
commands
and
then
my
helper
functions
that
kind
of
support
the
graphs
export
PS
graph.
A
Some
argument,
completion,
type
stuff
and
and
I
saw
have
actually
I
have
a
short
command
called
show
PS
graph,
which
is
the
same
as
the
export
PS
graph,
with
the
show
wrap
parameters.
I
should
retrofit
these
examples
to
use
that,
but
they're
effectively
this
one
effectively
wrapper
for
at
this
other
one
but
yeah
I,
think
I've
also
mentioned
you
know
it
is
for,
like
your
AWS
or.
A
A
B
A
B
A
So
if
you
want
run
these
I
have
a
demo.
Prep
that
is
forewarn
is
destructive
right,
like
it's,
creating
what
I
was
doing
but
I
think
I'm,
creating
groups
creating
users
create
demo
users,
so
I'm
doing
a
lot
of
putting
a
lot
of
garbage
in
there
in
order
to
query
stuff
like
creating
groups
with
Oh
use,
but
the
demo
itself.
A
So
this
80
groups,
I
list
the
80
users
and
I,
give
the
label
of
the
name
the
title
in
the
department,
then
for
each
group
we
get
the
members
and
as
long
as
there
are
member
as
I
draw
edges
from
the
members
distinguished
name
to
the
group,
distinguished
name,
I
use
a
distinguished
name,
because
that
was
like
my
unique
identifier
that
was
on
all
the
objects
that
I
was
querying
and
then
we're
shown
the
graph.
So
I
love
this,
because
this
is
the
one
graph
that
everybody
wants
to
generate.
A
A
problem
is
once
you
actually
do
generate
this
guy.
He
is
a
piece
of
spaghetti
monster
like
all
having
you
know,
a
thousand
some
nodes
on
the
graph
without
reference
back
to
each
other,
so
you
will
find
it
hard
to
see
what's
related
or
not,
but
I
do
take
it
further
and
I
said:
there's
nothing
wrong
with
generating
for
every
group
a
graph
showing
the
first
and
second
level
relationship
and
that
one's
actually
I
think
one's
really
impressive.
To
that
I.
A
Don't
have
my
my
my
VMs
set
up
on
this
box
to
do
it,
but
so
I'm,
basically
two
for
each
on
the
groups
and
I'm
graphing,
every
group
and
croute
so
and
then
to
tie
together
I'm,
adding
a
URL
to
an
HTML
file
that
matches
the
name
of
the
group
and
I.
Do
that
same
hack,
where
I'm
creating
an
image
on
the
fly
and
what
it
would
do
is
create
a
series
of
website
pages
that
each
one
shows
one
group
with
their
the
resident
leg.
A
Ships
but
they're
all
clickable
to
click
on
to
the
next
group
to
show
is
stained,
brac,
I,
suppose
I've
kind
of
replace
this
one
with
the
get
commit
lookups
since
I
didn't
have
a
my
VM
environment
up
and
going,
but
so,
if
you
do
have
an
active
directory
you
can
play
with
the
this
query.
Actually
is
pretty
generic
like,
like
my
80
groups
and
users,
is
indicates
filtering
on
just
a
no
you
in
my
demo,
so
the
query
of
the
data
would
be
safe
in
your
live
environment.
A
B
B
A
So
here's
a
way
that
I
can
walk
my
load
balancing
configuration
all
the
way
from
my
v
from
my
nodes
to
my
pools
to
my
GTM
to
the
endpoints
in
a
diagram,
the
green
of
stuff
that
was
online
when
I
ran
so
now
dependencies
at
this
component.
Dependence
is
depended
on
by
these
things,
and
it
depends
on
these
other
things
and
then
I
got
like
network
firewall
rules
that
are
showing
how
everything
relates
to
everything
else
and
the
access
available
to
those
items.
A
So,
there's
like
three
real-world
scenarios
where
we
dynamically
generate
diagrams
and
I,
don't
have
an
example
handy,
but
I
have
a
simpler
one.
When
we
right
now,
we
I
made
a
lot
of
stuff,
but
security
has
to
prove
all
like
external
firewall
changes
right,
which
makes
sense,
so
we
actually
generate
a
diagram
for
them,
showing
what
changes
we're
putting
in
place
like
these
servers
need
to
talk
to
these
servers
in
these
zones,
and
here
is
the
diagram
of
those
connections
in
the
port's
that
we
want
open.
So
it's.
B
A
Then,
instead
of
looking
at
an
excel
sheet
of
names
and
IP
addresses
the
might
at
a
glance,
be
hard
to
process,
because
you
gotta
really
think
through
what
you're,
seeing
if
I
can
get
a
diagram
showing?
Oh
hey,
it's!
Actually,
these
six
servers
in
this
zone
try
and
talk
to
this
one
server
in
this
zone
and
ice
their
diagram.
It
makes
their
job
easier
and
they
can
turn
on
those
requests
so
much
faster.
B
A
So
the
scale
of
our
environment,
like
we
do
a
lot
of
automation
and
we're
DevOps.
So
we
have
we
have
few
servers,
but
we
deploy
lots
of
components
to
things
and
we
have
280
deployable
things
like
somewhere
a
dev
to
make
a
change
to
any
one
of
these
280
things.
And
then
we
kick
off
deployments
to
like
one
of
you
know,
QA
dev
stage
production
and
spoke.
A
We
perhaps
because
we
have
to
actually
configure
the
exact
same
thing
in
every
environment
and
deploy
our
pipelines
to
work
and
we
actually
maintain
firewall
rules
between
each
component
individually.
So
of
that
280
some
components
there
deployables
I've
got
firewall
rules
for
each
one.
Tell
me
what
other
component
it
talks
to,
and
you
know
when
security
said
hey,
we
want.
You
know
individual
firewall
rules
for
every
little
thing
to
do.
We
said
sure
no
problem
and
we
generated
them
a
thousand
some
rules
to
process
and.
A
B
All
right,
gentlemen,
the
that
was
going
crazy
before
and
now
everybody's
gotten
super
quiet
sweetness.
If
you
have
a
question
for
Kevin,
let's,
let's
take
advantage
of
the
time,
I
mean
I.
Think
all
of
us
is
kind
of
blown
away
by
the
presentation
that
was
just
awesome.
Yeah
I,
just
you
know
a
lot
of
people.
They're
just
basically
said
that
was
really
cool
and
that
you
know
thank
you
very
much.
There's
a
lot
of
people.
B
A
B
A
Definitely
I'll
definitely
reuse
the
best
pieces
from
this.
One
I
will
probably
have
the
Active
Directory
demo
fleshed
out
and
working
and
I
think
the
other
one
I
have
I
think
is
a
nice
strong,
hitter
I
guess
would
be
V.
I
I
have
some
code
that
grasps
the
PowerShell
flow.
I
guess
like
the
ASC,
actually
mapped
the
ASC
out.
So
an
if-statement
would
show
the
branches
of
codes
and
functions
and
I
have
Advanced
Mode
that
explodes
that
out
and
does
the
ast
annotations
for
every
single
item
in
the
script.
B
Well,
I'll
just
reiterate
what
Phil
said
earlier.
First
of
all,
thanks
for
the
time
tonight
super
appreciative
of
the
presentation
on
trying
to
imagine
the
amount
of
time
that
went
into
making
a
tool
like
this.
Just
the
the
interesting
thing
about
the
tool
is,
there's
a
lot
of
options
there,
but
there's
nobody
that
sees
this.
That
doesn't
come
away
with
a
bunch
of
different
possibilities.
It's
not
it's
not
a
fool
suit.
That's
just
limited
to
like
doing
one
thing:
it's
really
an
idea
starter
for
lots
of
other
things
and
that's
the
cool
part,
absolutely.
A
That's
what
my
favorite
things
like
I
unlock.
You
know
your
trait
your
creative
potential
and
and,
like
I,
said
early
on
that
the
trick
is
finding
data
that
actually
has
both
your
primary
key
and
your
your
reference
lookup,
because
when
I
did
the
org
chart
right?
If
you
can
find
data
that
your
node
name
yeah,
you
can
find
out
where
your
node
name
is
the
ID.
That
is
also
a
property
li.
A
B
A
A
Absolutely
like
it's
your
right
so
yeah
we,
we
have,
you
know
a
caught
a
swimmer,
but
a
collection
of
nightly
tasks,
the
execute
and
that's
where
I
do
something
like
this
right.
If
you're
doing
an
org
chart,
this
is,
depending
on
you,
set
up
a
schedule
task
that
basically
fires
it
off
every
night
and
it
will
keep
upping
up
the
date
as
they
make
their
changes
throughout
the
day.
B
So,
do
you
have
automation
downstream?
That
says,
like
hey,
I
produce
these
files
and
you
have
something
that
grabs
those
files
and
puts
them
where
they
need
to
be
for
the
other
pieces
of
the
business
to
see
it
like
did
you
do
all
this
work
and
then
you
sat
there
and
drag
stuff
in
the
confluence.
Automatic
I
mean
manually
I,
don't
I,
don't
imagine
that's
the
case.
No.
A
A
We
basically
Jerry
the
grid,
the
diagrams
and
use
a
module
to
post
and
upload
them
in
that
cable
and
then,
like
our
firewall
rule,
rule
rules
request
like
talking
about
that's,
also
scripted
automated
to
create
a
service
now
requests.
So
we
different
the
image
we
use
their
API
to
create
the
request
for
ServiceNow
and
attach
the
image
automatically.
So
we
don't
have
to
think
about
it.
You
know
my
team
is
definitely
an
automation.
A
First
team,
like
as
we
step
into
new
projects
and
tasks
we
almost
automated
the
first
time
before
we
deploy
the
first
version
of
it
because,
like
I
said,
are
our
snowflakes.
We
just
have
so
many
of
them
because
they
are
in
every
environment
that
they
truly
aren't
snowflakes.
As
far
as
we're
concerned.
B
A
And
that's
that
the
DevOps
team
I
know
on
the
in
pressure
side
like
so
like
it's
the
economy
of
scale,
right
I
got
these
two
hundred
some
eighty
components,
but
I
can
probably
group
and
group
them
into
five
patterns.
You
know
most
of
them
are
API,
some
are
static
web
pages
and
it
becomes
a
cookie
cutter.
Cookie
cutter
pattern
there,
but
as
we
have
to
support
multiple
environments,
so
I
mentioned
like
a
dev
to
a
stage
and
production,
but
I
think
there's
like
often
four
or
five
others.
A
So
we
just
know
if
we're
doing
at
once.
We
are
doing
this
exact
same
thing.
At
least
you
know
six
or
more
times
I
mean
actually
in
production.
We
have
fail
overs
right,
so
it
balloons
pretty
fast.
So
if
we
intend
it
should
have
been
assembly,
it's
worth
it
to
build
out
the
DSC
or
the
script
or
the
tooling
to
automate
it,
because
it
won't
be
the
last
time
now
on
the
actual
infrastructure
side.
You
know
when
they're
saying
setting
up
shares
or
server
specific
purposes.
B
A
To
them
is
almost
too
unique
to
automate,
so
they
will
just
go
ahead
and
set
them
up
and
they'll
just
triage
it
later
if
they
need
to
remediate
problems
or
things
of
that
nature
like
we're
trying
to
like
we
kind
of
meet
in
the
middle.
Like
anything,
we
can
automate,
we
take
off
their
hands
and
kind
of
take
ownership
over,
but
we
at
the
drops
mean
we
still
depend
on
infrastructure.
Quite
a
bit.
You
know
like
exchange,
we
will
never
take
ownership
of
exchange
or
Active
Directory.
B
All
right,
so
with
that
guys,
what
I
think
I'd
like
to
do
is
we've
been
doing
with
these
virtual
taps?
Is
we've
left
the
meeting
open
for
20
minutes
or
so
30
minutes
after
was
just
for
people
to
conversate,
but
at
some
point
we
need
to
end
our
recording.
So
let's
give
it
another
we'll
give
everybody
here
another
two,
maybe
five
minutes
to
get
some
questions
in
and
then
we'll
call
it
a
day
on
the
presentation
and
you
guys
want
to
just
conversate
about
anything.
That's
fine
too.
So
last
call
for
questions
for
Kevin.
A
Alright,
so
we've
got
two
things
happening
here:
I
think
my
prompt,
so
I
think
I
have
a
the
PS
get
module,
that's
injecting
this
for
me,
so
I
mean
I'm
going
to
get
branch.
So
it's
adding
that
to
my
prompts,
but
what
I
also
tend
to
do
is
I
only
show
the
current
folder
that
I'm
in
actually
I
have
a
lot
of
goodies
in
here.
Actually
I.
B
B
B
A
Year
does
the
edits
for
for
for
gate
over
I?
Do
a
fancy
thing
doing
final:
come
whatever
the
console
of
automatic
eating
a
history
if
I
keep
my
own
so
every
time
I
run
a
command
I'm,
actually
logging
to
a
file,
the
can't
the
command
in
the
execution
time
with
Iran.
So
I've
got
history,
going
back
like
three
years
of
every
PowerShell
command
I've
executed
and
then
my
prompt
itself,
I,
basically
split
the
directory
to
just
show
the
folder.
A
The
leaf
of
the
location
and
I
set
that
as
my
prompt,
so
this
here
says:
I'm
in
the
PS
graph,
folder
somewhere
in
the
structure
of
things
and
I
will
set
the
title
least
from
the
normal
shell
to
the
full
location.
So
over
here,
I
can
see
I'm
in
see
users
k
mark
wet,
although
I'm
only
showing
my
using
folder
here
so
by
feeding
into
my
workspace
folder
see
DPS
graph.
What
I
found
was
that
I,
don't
like
my
my
prompt
getting
longer
and
longer
and
eventually
wrapping
all
my
commands.
A
I
felt
like
it
is
sufficient
to
know
what
folder
aren't
currently
in
and
if
I
ever
need
it
right.
There's
always
the
you
know,
get
location
that
can
yes
location
to
tell
me
where
I
actually
am,
if
I
ever
need
it
and
then
well
I
think
I
would
use
that
title
bar
I,
really
don't
I
could
probably
drop
that
from
my
prompt
function.
So
with
that
they
answer
your
question.
It.