►
From YouTube: Intro to Azure DevOps for sysadmins (Mike Kanakos)
Description
Azure DevOps is a set of services offered by Microsoft to create and maintain automation in the Azure cloud.
Most people look at Azure DevOps and see a service that is built for dev’s maintaining and release code releases. But if you look closely, you’ll see a set of tools for sysadmins to use for running automation (i.e. scheduled jobs) in the cloud or even on your own corporate on premises servers. This demo shows how you can get started with Azure DevOps and replace your dodgy scheduled tasks in only a few hours’ time.
Speaker Bio:
Our speaker for this presentation is Mike Kanakos, who is an Infrastructure Tools Engineer for InvisAlign and a Microsoft Cloud and Datacenter MVP. Mike specializes in managing directory services through automation.
Check his website https://www.commandline.ninja for more great content!
B
B
A
A
My
focus
is
really
a.d
all
day
directory
services,
azure
certificates,
azure
devops,
enterprise
applications,
oauth
registrations,
and
for
the
past
six
to
eight
months,
I've
been
focusing
on
building
sort
of
tools
to
help
other
team
members
or
other
teams
try
to
eliminate
tasks
that
are
redundant
or
very
manual
driven.
So
I've
been
focusing
on
building
tools
for
our
team
and
other
teams
focused
on
directory
services
stuff
and
I'm
also
a
microsoft
mvp.
B
A
A
So
about
a
year
ago,
when
we
started
doing
more
and
more
automation-
and
I
was
a
little
bit
newer
to
the
team-
I've
only
been
with
with
the
team
for
a
year
and
a
half,
I
was
thinking
powershell
jobs
using
this
task.
Scheduler,
maybe
like
powersheet
universal
whatever,
and
my
manager
is
a
very
forward-looking
guy
and
he
came
from
a
cloud
background
and
he
says
I
don't
want
to
have
really
anything
on
prem
if
we
don't
have
to
I'd
rather
leverage
the
cloud.
A
So
let's
find
a
solution
in
the
azure
cloud,
so
that
was
the
genesis
of
this,
and
my
manager
had
some
experience
with
this.
So
we
were
really
catching
up
to
him
and
since
then,
I
sort
of
myself
and
another
gentleman
we
handled
most
of
the
scheduling
of
automation
in
our
group
and
azure
devops
is
a
tool
that
I'm
in
pretty
much
every
day.
The
difference
here
is,
though,
is
that
I
called
this
talk
for
sysadmins,
because
I
was
going
to
mention
this
in
a
few
minutes.
A
B
And
then
from
that
is
basically
understanding
that
that
hey,
you
can
do
a
lot
more
than
what
we're
going
to
talk
about
at
this
point.
But
at
this
point
it's
really
the
to
to
see
and
to
plant
the
seed
in
everybody's
mind
here
at
this
point
say
that
we
can
go
from
there
and
really
understand
that
it
can
do
one
thing,
but
look
beyond
that
one
thing
and
find
the
other
things.
A
A
Instead,
you
should
be
watching
this
talk
with
here's,
how
we
solve
the
problem
at
all
work,
and
maybe
the
solution
might
work
for
you.
However,
maybe
someone
in
the
audience
tonight
has
a
better
solution
than
what
I'm
proposing
or
offering,
or
maybe
they
can
offer
me
some
tips
that
I
haven't
picked
up
yet
so
this
is
a
knowledge
share.
A
A
A
So
my
environment
is
an
on-prem
active
directory
running
in
my
lab,
the
azure
cloud,
azure
devops
and
a
code
repository
now.
One
big
caveat
that
I
just
want
to
explain
to
people
is
that
you
can
connect
your
on-prem
ad
to
the
azure
cloud
and
sync
those
two
so
that
your
on-prem
accounts
now
live
out
in
the
azure
cloud,
and
then
you
can
use
single
sign-on
and
third-party
services
and
all
the
things
that
azure
offers.
A
But
you
don't
need
to
do
that
for
this
to
work
and
as
a
matter
of
fact,
in
my
demo
here
I
don't
have
any
connection
between
my
on-prem
ad
and
the
azure
cloud.
So
that's
not
a
requirement,
but
just
so
everybody's
clear.
So
when
you
see
that
I'm
logging
into
azure,
you
may
be
assuming
I'll,
be
it's
because
he's
got
an
account
synced
up
now
I
have
a
standalone
azure
environment
that
I've
poked
a
hole
into
my
lab
environment
and
that's
how
I'm
doing
this
work.
So
with
that
being
said,
quick
level
set.
A
We
have
a
box
here
called
dc01,
it's
dc,
as
I'm
sure
you
can
imagine,
and
we
have
an
azure
build
box,
which
I
wasn't
expecting
that
to
be
open.
Okay,
an
azure
build
box
here
and
the
azure
build
box
I'll
explain
in
a
second,
but
in
the
simplest
terms,
azure
is
reaching
out
to
a
server
on-prem
and
that
on-prem
server
is
doing
work
and
I'll
go
more
into
this
in
a
second.
But
what
we're
going
to
do
here
is
I'm
going
to
just
show
you
a
very
quick
demo
here.
A
This
is
the
list
of
my
jobs
here
and
I'm
going
to
I've
set
up
this
pipeline
here
and
I'm
going
to
run
it
and
I'll
show
you
what
it's
going
to
do
here
and
what
that's
going
to
do
if
things
are
going
to
work
correctly,
is
that
should
go
out
and
reach
out
to
this
machine,
create
an
rtp,
sub
demo
folder
and
did
that
just
run
around
8
30?
No,
that's
old!
A
Let
me
delete
that.
I
thought
I
deleted
this
already
all
right.
It
should
create
a
file
in
here.
Okay,
as
you
can
see,
this
job
is
running
here.
If
I
click
this,
I'm
going
to
explain
all
this,
so
I'm
not
going
to
when
my
script
failed.
Yeah,
that's
awesome!
Oh
the
file
already
existed.
So
let's
run
that
again.
A
A
B
So
I
think
at
this
point
it's
a
good
point
to
figure
out
is
that
when
you're
running
these
jobs,
it's
just
as
easy
to
come
in
here
and
because
you
people
have
experience
with
powershell
at
this
point
you
can
very
quickly
hear
and
figure
out
so
granted.
A
powershell
script
is
running
somewhere
in
azure
cloud
somewhere
at
that
point,
and
so
it
can
do
work
and
you
can
just
easily
while
the
jobs
running
is
to
look
at
that
code.
So
that's
really
the
core
at
this
point.
So
is
that
a
good.
A
Yeah
yeah
and
I
was
going
to
cover
that-
and
I
will
be
honest-
I'm
perfectly
fine.
If
demos
fail
tonight
because
that's
part
of
running
pipelines,
jobs
fail,
you
need
to
be
able
to
troubleshoot
why
and
it
makes
it
very
easy
to
troubleshoot,
because
what
happens
is
pointed
out
very
quickly.
I
wanted
to
get
through
one
so
that
someone
can
see
what
a
job
is
that
runs
okay.
So
I
got
a
green
check
here
so
that
ran
and
we
should
have
a
new
folder
rtp
seg
demo,
hello
world.
There
we
go
very
exciting
okay.
A
So
I
made
a
quickie
diagram
here
and
I'm
going
to
explain
this
so
just
to
get
everybody
on
the
same
page,
we're
working
out
in
the
azure
cloud
inside
the
azure
cloud,
this
azure
devops
the
service
and
this
azure
id,
but
I
also
have
an
on-prem
id
in
my
local
network.
So
what
I've
done
is
I've
configured
azure
devops,
to
talk
to
a
server,
that's
on-prem
here
and
that
server
is
going
to
do
work
for
me
now.
A
That's
mostly
what
I'm
going
to
focus
on
tonight,
but
the
concept
between
running
jobs
in
the
cloud
or
running
john's
prem
is
essentially
the
same.
You
configure
a
pipeline
that
pipeline
can
has
a
bunch
of
items
that
you
can
figure.
This
is
the
script
I
want
to
run.
This
is
the
environment
I
want
to
run
in
is
the
credentials.
A
I
want
to
run
it
as
this
is
the
time
I
want
to
run
it,
and
I
want
it
to
repeat
x
amount
of
times
and
then
once
you
set
those
things,
azure
devops
runs
in
the
background
for
you.
The
reason
that
I'm
excited
about
azure
devops
is
because,
if
we
step
back
for
a
second,
we
talk
about
test
scheduling
inside
the
windows
environment.
A
What
options
do
we
have
today?
We
basically
have
one.
We
have
task
scheduler
and
test
scheduler,
it's
fine.
It
works
great.
This
powershell
commandlets
for
scheduled
tests,
there's
powershell
commands
for
scheduled
jobs,
although
scheduled
jobs
didn't
make
it
into
powershell
seven.
And
of
course,
you
have
third
party
solutions
that
you
could
add
on
jenkins,
powershell
universal,
there's
lots
of
choices
out
there,
but
a
lot
of
people
don't
realize
that
azure
devops
is
a
legitimate
option
for
replacing
that
task
scheduler.
A
A
You
just
run
the
jobs
and
it
reaches
out
to
the
agents
that
you
configured
in
this
case.
Obviously
I
have
an
on-prem
agent,
but
the
point
is
that
I
am
now
relying
on
microsoft
and
their
reliability
to
give
me
an
environment
that
works
and
when
I
kick
off
jobs,
particularly
in
the
azure
100
percent
of
non
dependent
on
microsoft.
But
the
cool
thing
is,
I
don't
need
to
provision
anything.
A
Azure
devops
will
spin
up
resources
just
in
time
in
order
to
run
that
job
that
I
need
to
run
in
the
cloud
and
then
when
it's
done
and
throw
away-
and
I
pay
for
just
a
small
sliver
of
time
that
job
ran
in
the
azure
cloud.
On-Prem
is
a
little
different.
I
have
an
agent
that
runs
here,
but
the
azure
devops
service
will
run
that
job.
It
will
check
the
code
out,
pass
it
down
to
the
on-prem
agent
on-prem
agent
and
do
some
work,
but
what
a
lot
of
people
don't
know
about?
A
Yeah,
so
let's
take
a
look
at
that
for
a
second
here
before
I
jump
to
that.
I
should
probably
just
mention
that
to
get
to
the
azure
devops
website,
it's
dev.azure.com
and
when
you
go
there
you'll
be
prompted
to
create
an
account,
whatever
account
that
you
decide
to
make.
That's
your
link,
so
my
azure
devops
environment
is
dev.azure.commcnaccous.
A
A
The
first
thing
that
you
start
with
is
a
project
and
inside
a
project.
It
consists
of
these
five
pieces
here:
boards,
repos
pipelines,
test
plans
and
artifacts.
Now,
let's
take
a
step
back
for
a
second,
this
all
sounds
very
devops
and
this
sounds
maybe
a
little
intimidating
for
assist
admin.
But
if
we
go
back-
and
we
look
at
this
for
a
second-
if
you're
a
sysadmin
and
your
goal
is
to
create
scripts-
that
you
need
to
schedule
and
run
every
day,
I
need
to
produce
reports.
I
need
to
do
a
d.
A
I
need
to
check
status
of
services
for
on-prem
resources,
jobs
that
are
basically
set
once
write
a
script
and
I
just
needed
to
run
over
and
over
again,
but
you
don't
need
agile
planning
and
you
don't
need,
probably
don't
need
test
plans
and
you
probably
don't
need
artifacts,
and
you
see
here
you
can
turn
on
or
off
these
features.
So
now,
let's
take
away
those
three
features
and
what
do
we
have?
We
have
a
pipeline
and
we
have
a
repo
and
now
all
of
a
sudden,
azure
devops
doesn't
sound
so
scary.
A
So
what
I've
done
here
in
my
environment
is
sorry
come
to
windows
here.
If
I
go
into
my
you
see
here
these
two
icons
just
to
show
you
a
quickie,
here's,
the
five
icons,
here's
two
I've
shut
off
those
three
icons.
I
don't
need
boards.
I
don't
need
test
plans,
I
don't
need
artifacts.
So
now,
when
I
come
into
my
project,
I
have
a
repo
and
I
have
the
pipelines
themselves
and
this
interface
here
is
just
the
recent
pipelines
I
go
here.
A
This
is
all
the
pipelines
that
are
in
and
the
interface
is
pretty
easy
to
understand.
Green
is
good.
Red
is
bad
and
I
can
drill-
and
I
can
see
more
so
so
that's
the
interface
and-
and
this
is
the
reason
that
I
wanted
to
talk
about
this,
because
when
you
dumb
this
down,
it's
a
lot
easier
to
understand,
but
at
first
blush,
when
you
look
at
azure
devops
and
you
oh
man,
it's
artifacts
and
test
plans
and
boards-
and
oh
I
don't
that's
not
for
me.
A
The
truth
is
take
away
those
pieces
and
think
about
the
things
that
you
might
need
as
a
sys
admin,
doing
on-prem
jobs
and
now,
all
of
a
sudden,
it's
a
pretty
robust
little
tool
here
and
we're
going
to
talk
about
that
more
in
a
second.
Let
me
just
take
a
quick
look
at
my
notes.
Make
sure.
Oh,
I
wanted
to
talk
about
the
cost,
so
I
do
have
a
link
here.
A
The
link
is
azure.microsoft
enus
practicing
calculator,
I'm
going
to
go
there
right
now,
because
this
is
probably
pretty
surprising
to
most
people
when
you
get
there.
This
is
the
pricing
calculator
for
the
azure
services
on
the
left-hand
side.
You
can
go
down
to
devops
over
here
and
then
you
can
pick
up
way
to
go.
Azure
devops.
A
A
A
When
I
get
to
my
sixth
user,
it's
six
dollars
a
month,
so
how
many
organizations
have
people
who
have
to
schedule
automation
and
need
more
than
five
users
that
need
to
go
into
their
devops
console
if
it's
less
than
five
this
cause,
this
service
costs
nothing!
Okay,
if
you
need
to
do
test
plans,
that's
a
different
story
and
it
gets
very
pricey.
But,
as
I
said,
you
probably
don't
need
that
you
can
rock
on
the
free
plan
and
what
that
will
allow
you
to
get
is
a
number
of
things.
A
So
self-hosted
pipelines
you
are
allowed
to
have
one
self-hosted
pipeline,
which
is
what
I'm
running,
which
means
you
can
install
one
engine
on
a
server
on
your
environment
and
you
can
run
that
and
you
can
run
one
job
at
a
time
and
it
is
free.
So
we
now
have
a
service
that
has
pipelines
and
repos
up
to
five
users
can
log
in
and
work
in
the
terminal
you
can
have
jobs
scheduled
in
there.
A
You
can't
run
more
than
one
at
a
time,
but
all
jobs
you
run
are
absolutely
free,
no
matter
how
much
run
time
there
is
so
now,
when
you
start
to
look
at
this,
you
say
wow.
This
is
a
pretty
cool
solution.
Maybe
I
need
some
more
time.
I
want
to
start
extra
agents,
three
four
five
agents,
sure
maybe
it
gets
a
little
expensive
but
remember
this
is
parallel
jobs.
At
the
same
time,
how
many
times
you're
gonna
have
four
jobs
running
at
the
exact
same
time,
if
you
don't
need
that
bring
this
down.
A
A
Probably
a
service
that
might
not
be
intended
for
our
use
cases
when
you
dig
in
is
actually
a
pretty
good
use
case
for
what
we
want
to
do.
So
let
me
just
jump
back
in
my
notes
here
and
see
what
else
I
want
to
mention.
As
I
mentioned
surprisingly
cheap,
I
talked
about
my
lab
setup.
I
have
an
active
directory
on-prem
domain
azure
id
out
in
the
cloud
azure
devops,
my
automation,
engine
get
is
where
I'm
posting
my
code.
A
A
I
am
not
sure
of
the
answer.
What
I
know
is
that
you
need
to
set
them
per
project
and
what
I
thought
I
would
do
is
just
to
give
everybody
a
quick
glance
here.
I
have
demos
set
up
in
my
project,
but
I
thought
it
would
be
fun
if
we
did
everything
right
and
whenever
I
can't
get
it
to
work,
we'll
just
dump
back
to
what
I
have
pre-built.
A
So
what
I
want
to
do
is:
let's
make
a
project:
let's
set
up
a
repo,
let's
set
up
some
pipelines
and
let's
run
some
jobs
and
if
it
gets
too
tedious
and
cumbersome,
we'll
just
run
the
ones
I've
built
in,
and
everybody
can
then
see
each
part
of
the
puzzle.
So
here
I
am
I'm
at
my
top
level.
I
have
a
project
I
want
to
make
a
new
project.
It's
right
here,
I'm
going
to
call
this
rtp
sub
live
right,
and
do
I
want
to
make
this
project
public
or
private?
A
I
want
to
make
it
private,
so
this
takes
about
30
seconds
and
there
you
go
so
by
the
default
yeah.
I
have
everything
enabled
on
the
side
here.
So
what
is
not
completely
obvious
at
first
is
when
you
bring
this
and
I'm
sorry
my
something's
popping
up
here
but
in
the
bottom
left,
is
a
project
settings
button
and
that
is
easy
to
miss
when
you
look
in
the
interface.
So
when
you
go
to
project
settings,
you'll
now
have
a
bunch
of
things
that
you
can
set
specific
to
the
project.
A
So
what
I'm
going
to
go
here
is-
and
I
forgot
already
where
it
is
settings-
is
how
long
data
is
kept.
Let
me
just
see
if
I
can
remember
where
that
is
make
a
project.
I
don't
know
what
you
need
yeah.
Of
course
I
didn't
make
a
note
for
myself
and
what
was
on
there,
but
let
me
just
go
through
real
quick,
because
it's
one
of
these
project
configuration
no.
No,
no,
I'm
sorry,
I'm
not
going
to
waste
your
time
with
this
look
real
quick
overview.
A
Oh
man,
I
feel
like
a
dummy.
Now,
oh
it
was
off
the
top.
It
was
just
scroll
down.
I'm
sorry!
So
right
when
you
go
into
the
project
overview,
you
just
have
to
scroll
down
and
I
can
turn
off
what
I
don't
want.
So
here
I
am
I'm
going
to
get
rid
of
these
services
test
plans
I
don't
need
and
and
for
what
I'm
looking
to
do,
I
don't
need
artifacts.
A
Now
we
can
talk
sidebar
and
there's
plenty
of
people
on
this
call
that
are
doing
more
devops
oriented
stuff
and
they
would
probably
say
that
you
absolutely
need
test
plans
and
artifacts
and
you
want
to
have
a
little
bit
more
of
a
game
plan,
but
not
for
what
I'm
looking
to
do
so
I
shut
those
off.
I
can
come
in
here
and
I
can
set
some
permissions,
but
for
what
I
need
to
do,
I'm
a
global
admin,
a
lab,
so
I'm
ready
to
go
the
way
I
am
here.
A
I
can
change
notifications.
I
can
get
very
granular
here
and
set
what
comes
up
here
and
I'll
show
you
what
that
looks
like
for
a
second.
If
I
go
here
and
jump
into
my
email,
so
every
time
right
now
with
the
default
settings,
when
I
run
the
jobs,
where
are
my
builds
today,
I
get
a
notification
and
you
probably
don't
want
to
know
about
every
notification
and
you
can
tweak
that.
A
Come
on
wait
for
my
email
to
catch
up
to
me
while
we're
talking,
so
you
can
come
in
here
and
customize
this.
What
you
want
to
be
notified,
what
you
don't
want
to
be
notified
for-
let's
just
go
in
here
for
a
second.
So
if
I
pull
my
mail
for
a
second,
you
can
see
here
see
all
these
azure
devops
emails.
This
is
every
time
I
ran
a
job,
a
notification
that
came
to.
Let
me
know
so
what
we
do
at
my
work
is.
We
have
two
projects
and
we
have
what
we
call
prod
project.
A
We
have
a
test
project
and
the
prod
project
is
where
code
lives,
that's
running
on
a
schedule
and
we
get
notifications
on
what
runs
and
what
fails.
We
have
a
test
project
that
is
essentially
the
same
code
right,
but
you
don't
schedule
jobs
there
and
you
can
test
and
it
doesn't
notify
out
on
every
fail
of
every
job
in
email,
five
people
in
your
group
when
you're
trying
to
change
settings-
and
you
run
a
job
30
times
trying
to
figure
out
the
settings.
A
So
we
have
a
test
pool
and
we
have
a
prod
instance
that
we
use
and
they're
supposed
to
be
the
same
stuff.
That
is
important,
lives
and
pride
and
you
play
and
test.
So
I
have
this
azure
devops
demo
here
right.
Excuse
me
wrong
one.
Let
me
go
to
it
live
all
right.
So
when
I
start
off,
I
get
pipelines
and
it's
blank
and
in
pipelines
there's
a
number
of
different
things
here
which
we'll
talk
about
in
a
second
and
I
have
repos
and
repos.
This
is
a
git
repo.
A
But
now,
if
I
go
back
over
to
my
interface,
the
relay
should
be
no
surprise
here.
If
I
go
into
repo
and
I
go
into
files,
I
have
a
file
there,
so
I'm
now
working
in
the
azure
cloud,
I'm
working
at
my
local
terminal-
and
I
really
didn't-
have
to
do
anything
to
configure
this
there's.
No
there's
really
no
work
to
do
the
hardest.
Part
of
this
whole
setup
is
installing
an
agent
and
configuring,
maybe
some
permissions.
A
B
A
A
You
have
to
create
a
personal
access
token,
if
you're
not
familiar
with
that,
you
would
go
here
and
you
would
drop
down
right
here
and
you'd,
say
personal
access
token
and
that's
going
to
create
a
token
and
a
super
long
key
and
you
copy
out
that
key.
You
paste
that
key
in
here
and
you
hit
enter
now
check
out
what
I
had
to
do
after
that.
It
said,
enter
the
name
of
the
agent
pool
so
just
to
bring
everybody
up
to
speed.
I'm
going
to
show
you
real,
quick
on
my
existing
azure
devops
environment.
A
A
Okay,
so
by
default,
when
you
start
up
a
new
project,
it
creates
some
agent
pools
for
you.
So
this
is
oh
I'll.
Explain
this
one
to
say
new!
The
default
is
azure
pipelines
that
runs
in
the
cloud
and
default
is
just
there,
so
it's
empty.
If
you
go
in
here,
there's
no
agent!
Okay,
if
I
go
into
azure
pipelines-
oh,
I
am
looking.
Oh
it's
pulling
in
the
history
for
my
other
one,
okay,
so
there's
one
agent
right.
So
what
I'm
configuring
is.
Let
me
get
back.
A
There
is
the
agent
for
the
for
this
on-prem
now
to
be
clear.
This
on-prem
pool
didn't
exist.
I
created
it.
I
just
said,
add,
pool-
and
I
said,
give
it
a
name
and
then
what
I'm
doing
when
I'm
configuring
this
agent
back
over
here
in
this
screenshot,
I'm
just
picking
the
name
that
I
assigned
in
azure
devops
and
it's
going
to
add
this
agent
to
that
pool.
And
then,
if
you
look
at
these
settings
here,
I
understand
this
window
is
not
familiar
to
people,
but
if
you
go
through
these
things
enter
agent
name
press
enter.
A
If
it's
the
name
of
your
server
I
hit
enter
successfully,
add
the
agent
you
want
to
run
the
agent
as
a
service
press
y.
Okay,
pick
an
account
if
you
don't
it's
going
to
be
empty
server
network
service.
So
I
essentially
answered
two
questions
and
went
the
equivalent
of
next
and
I
installed
the
agent.
So
it's
not
a
challenging
task.
A
A
So
let
me
show
you
what
I've
done
a
little
bit
and
give
you
a
little
walk
through
the
interface,
because
when
it
comes
to
creating
jobs,
there
is
a
ton
of
choices,
and
this
is
why
I
want
to
spend
a
lot
of
time
on
the
choices
because
running
the
code
is
not
the
difficult
part.
Configuring
the
job
to
run
can
be
a
challenge
in
the
beginning.
A
So
I'm
going
to
jump
back
over
to
my
existing
demo
here
and
I'm
going
to
go
into
pipelines
and
I'm
going
to
these
are
existing
jobs,
I'm
just
going
to
make
a
new
one
just
so
we
can
look
at
it,
okay,
so
the
very
first
thing
it
asks
you
is:
where
is
your
code?
And
I
know
exactly
what
everybody's
going
to
think?
Oh,
it's
an
azure
repo.
A
No,
you
can
do
this,
but
I'm
here
to
tell
you
that
all
these
choices
here
are
more
difficult
in
this
tiny
little
text
down
here
it
says,
use
the
classic
editor,
that's
what
I'm
going
to
use
tonight
because
that's
going
to
create
a
pipeline
without
yaml.
So
what
we
mean
by
without
yavin
is
gooey
interface.
So
when
I
click
the
classic
nml
editor,
this
goes
away
and
I
get
a
new
interface.
A
Am
I
using
the
azure
repos
get?
Yes,
I
am
okay,
so
I
would
say:
continue
it
pre-populated
everything
and
then
I
get
to
this
window,
which
is
just
equally
as
confusing.
These
are
all
the
different
things
that
could
be
run
in
my
pipeline.
However,
I
don't
want
any
of
these.
I
want
to
start
with
an
empty
job
and
there's
a
reason
for
this,
because
it's
much
easier
this
way.
So
when
I
do
that,
I
say
no,
I
don't
want
the
yaml
in
a
face.
A
A
A
A
A
All
I
need
to
do
is
click
this
plus,
and
now
I
get
an
interface
like
we
saw
before
and
there's
all
different
things
here
that
we
can
run
I'm
going
to
stop
here
for
a
second
and
see
if
that
makes
sense,
because
I
don't
want
to
go
too
far
ahead.
If
it's
a
little
confusing
questions,
comments,
concerns.
A
Perfect,
so
I
can
browse
here
very
simple,
but
it's
a
powershell
user
group,
we're
gonna
run
some
powershell,
so
I'm
just
gonna
go
over
here
and
I'm
gonna
type
powershell,
and
when
I
do
that,
it's
gonna
tell
me:
what
do
you
want
to
do?
You
want
a
power
shell.
You
want
powershell
on
remote
machines.
You
want
azure
powershell,
you
want
to
run
functions.
A
Anything
related
to
powershell
is
here
every
job,
I've
ever
configured.
I
just
pick
powershell
and
what
that
does
is
it
adds
this
to
this
pipeline.
Now
that
I
have
this
setting
here
that
says,
I
want
to
run
a
powershell
script.
When
I
click
on
this,
I
can
now
configure
that,
and
here
it's
asking
for
a
script
path.
Do
you
want
to
take
a
guess
where
it's
going
to
pull
from
it's
going
to
pull
from
my
azure
devops
repo
remember
this
is
in
my
pre-populated
project.
So
I
have
all
these
scripts
up
here
already.
A
A
So
there's
an
argument
section
in
here.
You
can
set
up
some
arguments
so
we're
going
to
get
into
that
in
a
second.
But
right
now
I
have
configured
a
job.
If
I
go
save
here-
and
this
is-
if
you
want
to
do
comments-
you
can-
I
usually
don't
okay,
I've
saved
this,
and
now
I
can
just
go
back
up
one
level.
A
A
So
there's
some
things
in
here
that
I
didn't
cover
that
I
want
to
go
back
now
and
dive
in
a
little
deeper
on.
So
the
first
here
is
triggers,
so
triggers
is
where
you
can
set
a
schedule,
so
I
can
say:
when
do
I
want
this
to
run?
I
can
pick
a
time
zone
if
I
have
a
specific
time
zone
and
if
I
want
to
add
more,
I
just
add
as
many
schedules
as
I
want
here
now.
There
is
one
limitation.
A
A
You
do
that
with
yaml
and
the
yaml
scheduling
agent
is
much
more
robust
than
this,
but
it's
a
lot
more
cryptic,
but
so
configuring
jobs
is
a
pretty
simple
task
and
then
this
little
check
box
here
can
say
I
only
wanted
to
run
if
the
build
of
the
source
changed.
I'm
not
interested
in
that.
So
I
uncheck
that
and
if
you
want
to
deal
with
branches,
you
could
deal
with
branches.
A
But
what
we're
basically
saying
here
is
I
have
a
script
and
I
want
to
set
up
that
script
to
run
at
a
certain
time
and
on
certain
days,
and
I
just
wanted
to
run-
I
don't
care
about
if
the
chain
build
changes
or
any
of
that
stuff.
I
just
want
to
automate
a
job,
and
this
is
the
interface
okay.
So
questions.
B
A
And
so
we're
going
to
expand
on
this.
So
what
and
so
the
one
thing
you're
probably
wondering
is
mike
you
mentioned:
you
have
no
rights,
so
how's
this
going
to
run
so
one
we
talk
about
that
for
a
second,
so
to
do
that,
I'm
going
to
jump
into
some
code
all
right.
So
if
we
open
up
here,
let's,
let's
take
a
look
at
what
we
got
for
scripts
here.
A
A
Bear
with
me
one
second:
I
have
a
test
lab
here.
I
should
have
showed
this
earlier
and
I
have
some
test
user
accounts
in
here
and
what
I
want
to
do
is
this
script.
What
it
does
is,
it
goes
and
it
went
and
found
this
user
and
it
populated
a
company
name
for
her
organization.
I
populated
this,
so
it
won't
run
so
I'm
gonna
pick
another
user
here.
Let's
try
s
horton,
okay,.
A
Okay,
if
we
work
from
the
bottom
up,
simple
ad
commandlet,
I
set
I'm
setting
the
company
attribute
on
this
username
s
horton.
However,
I'm
going
to
use
a
credential
to
get
me
to
give
myself
the
rights
to
do
it
because
the
service
that's
going
to
run.
This
is
network
service
and
network
service
is
not
going
to
have
the
ability
to
make
these
changes.
If
I
was
doing
an
ad
query,
get
ad
user,
something
like
that.
A
I
don't
necessarily
have
to
put
permissions
in
unless
I
want
to
do
something
really
crazy,
because
most
people
can
browse
active
directory,
but
when
I
want
to
make
changes,
that's
a
different
story.
So
to
do.
That
is
a
really
simple
way.
We
do
that.
So
what
we've
done
is
we've
built
two
parameters
at
the
top
here
and
we're
going
to
pass
in
a
user
id
and
a
password
okay
and
we're
going
to
take
those
two
credentials
and
we're
going
to
make
a
secure
string
and
then
we're
going
to
make
a
credential
object.
A
Okay,
so
what
we're
doing
is
I'm
going
to
pass
in
some
user
id
we're
going
to
pass
in
some
password
from
azure
devops?
The
script
is
going
to
take
that
password
that
got
passed
in
as
as
a
parameter
and
it's
going
to
create
a
secure
string
and
then
I'm
making
a
second
object
here
called
credential.
It's
just
a
previous
credential
object
and
then
once
I
have
this
credential
object,
I
can
now
run
this
command
as
another.
A
A
A
A
A
What
happens
if
you
want
to
run
powershell
7
right
here,
use
powershell
core.
So
if
I
leave
this
off,
it's
just
going
to
use
the
powershell51
engine
to
run
this
for
what
I'm
doing
with
ad
that
has
no
difference
I'll
leave
it
as
powershell
core.
Now,
let's
talk
about
this
for
a
second
here,
because
this
can
be
a
bit
confusing.
A
What
I'm
doing
is
I'm
saying
in
this
job
there's
two
parameters
in
the
script:
user
id
and
password
okay.
So
here
I'm
saying
I
have
a
parameter
called
user
id
and
a
parameter
called
password
and
for
those
parameters
I'm
going
to
pass
in
this
variable,
which
you
don't
know
about
yet
in
this
variable,
which
you
also
don't
know
about
yet
so
what
I'm
trying
to
get
to
is.
There
is
a
way
to
store
credentials
in
here.
A
You
can
use
azure
key
vault,
but
there's
a
much
simpler
way
inside
a
project
inside
pipelines
you
have
a
library
and
a
library
is
a
way
for
you
to
store
variables.
So
what
I
have
done
is
I
created
a
variable
group.
This
is
not
here.
By
default,
I
created
this
group.
It's
called
credentials
and
in
credentials
I
stored
two
variables:
let's
go
to
variable
called
rtp
sub
service
account
id
and.
B
A
Sorry
about
that
guys,
I
meant
to
do
that
earlier.
Thank
you
for
the
reminder.
Okay!
So
well,
let
me
just
get
back
on
track
here
so
inside
of
our
project
at
the
pipeline
level,
I'm
in
the
library-
it's
not
so
obvious
at
first
but
there's
a
library-
and
I
can
put
any
variables
I
want
in
here-
I
don't
have
to
make
credentials.
Maybe
I
need
to
put
variables
in
for
os
names
or
whatever
I
can
create
all
kinds
of
variables
and
then
set
values
for
those
variables
and
pass
them
into
my
scripts.
A
I
created
the
variable
name,
and
this
is
the
value.
So
this
is
the
accounting
id
that
I
want
to
run
my
script
and
here
is
the
password
and,
as
you
can
see,
I'm
going
to
zoom
this
out
for
one
second,
it's
starred
and
if
I
unlock
this
it
doesn't
it's,
it
doesn't
show.
I'm
gonna
wipe
it
out
here,
because
I
have
a
super
safe
password.
A
A
I
want
to
stay,
and
I
have
to
save
this
because
I
made
a
change
so
go
back
to
the
pipeline
and
I
go
back
to
that
company
perm
here
and
I
go
to
edit
remember.
We
said
the
agent
is
going
to
run
a
powershell
script
when
I
click
on
that
powershell
script,
it's
going
to
run
this
script,
and
now
you
make
it
makes
sense.
What
I'm
doing
here,
I'm
going
to
pull
in
those
variables
from
the
library
pass
them
into
these
parameters.
A
A
So
that
seems
like
a
lot
of
steps,
but
actually
it's
not
because
once
you
do
it
chances
are,
if
you're
doing
using
a
service
account,
you
may
use
the
same
service
account
for
all
your
jobs.
So
then
all
you
need
to
do.
You
literally
can
copy
and
paste
this
line
and
save
it,
and
you
have
to
go
here
to
this
variable
section
and
the
only
catch
here
is
you
have
to
go
to
variable
groups
and
you
have
to
say
link
all
we're
saying
is
hey
for
this
script.
A
I
want
to
link
the
library
to
the
script
when
I
do
that.
It
now
says
these
two
are
available
to
me.
Imagine
if
you
had
lots
of
variables
and
lots
of
groups
you
might
want
to
pick
the
ones
you
want
the
way
I
do
it.
We
just
have
a
subset
of
credentials,
one
I
want
to
use,
and
this
is
how
we
pass
them
in
and
we
essentially
have
the
same
settings
for
every
job.
We
run
with
the
arguments.
A
So
what
seems
a
little
confusing
now
becomes
very
easy
because
it
becomes
repetitive.
So
we
pick
a
different
script.
We
just
use
the
same
credentials
because
as
a
service
account-
and
maybe
we
can
figure
some
other
things
here-
maybe
we
change
some
triggers,
but
it's
essentially
the
same
setting
for
every
job,
we're
just
changing
the
scripts
and
oh
by
the
way.
A
If
I
want
to
run
more
than
one
script,
I
can
put
multiple
scripts
here
and
I
can
add,
as
many
as
I
want,
and
the
agent
will
go
through
and
execute
these
scripts
in
order
and
do
whatever
it
is
that
you
like.
Maybe
you
need
to
run
a
powershell
script
and
then
you
need
to
follow
that
up
by
a
command
interface.
A
And
then,
maybe
after
that,
you
need
to
do
some
bash.
Of
course,
that
doesn't
seem
logical,
but
the
point
is
see
how
easy
this
is
to
do
now,
of
course,
for
each
one
of
these
you'll
have
to
set
up
some
settings
right
and
each
one
of
them
have
logical
settings,
but
once
you
get
used
to
this
interface,
you
get
a
lot
of
options
that
honestly,
you
just
can't
do
in
the
task.
Scheduler
and
remember,
I
picked
an
on-prem
job
here.
B
A
So
I'm
going
to
leave
this
for
a
second
here,
okay
and
I
might
go
back
to
my
pipelines,
so
the
history
is
with
the
job
okay.
So
what
you're
seeing
here
is
the
name
of
the
pipeline,
the
last
run
information
and
how
long
it
ran.
But
if
I
click
on
this,
I
can
go
in
here
and
I
can
see
the
history
now,
let's
jump
in
here,
we
could
see
here
that
this
one,
this
is
my
hello
world.
One
ran
a
couple
were
unsuccessful.
A
When
I
click
on
this,
I
can
go
in
here
and
I
can
go
and
I
can
see
that
agent
job-
and
I
can
exactly
see
here
now.
This
was
the
one
that
ran
now.
What
happens
is
there's
a
process
that
goes
on
here
and
the
agent
and
starts
the
job.
It
checks
out
the
job
which
is
basically
pulling
the
code
from
the
repository.
A
A
A
A
But
what
we
do
is
we
usually
keep
our
logs
on
prem
anyway.
So
we
just
have
part
of
the
script,
create
a
log
and
output
it
to
something
local
on
the
network,
so
these
are
failed
ones
and
when
I
go
into
a
failed
one
here,
it
tells
me
exactly
two
errors
here
and
I
can
dive
in
here
and
I
can
actually
see-
and
it's
these
it's
very
easy
to
see
here.
I
did
not
have
privileges
to
run
this
operation,
and
usually
this
one
is
then
a
cloud
job.
A
A
A
Right,
I
guess
it's
gonna
make
it
look
small.
I
know
this
is
hard
for
you
guys
to
see
when
it's
this
small
so
connecting
to
a
remote
server
failed
with
the
following
errors:
message
access
denied,
so
I
was
trying
to
do
remoting
here.
Okay,
and
this
is
the
line
in
the
script
that
it
failed.
It
puts
it's
hard
to
see
here,
but
normally,
when
it
fails,
it
underlines
the
failed
part
of
the
code.
This
whole
line
failed.
So
it
underlined
the
whole
thing.
B
So
if
you're
doing
this
kind
of
stuff,
you
can
do
your
somebody
brought
up
in
the
questions
saying
the
idea
of
passing
data
from
one
script
to
another
kind
of
stuff,
and
so
you
could
write
to
the
local
disk
and
then
basically
pick
it
up
in
the
next
script
off
the
local
disk.
You
could
do
that
kind
of
stuff
or
have
you
had
any
experience
in
putting
and
passing
data
from
one
part
of
the
job
to
the
next
part
of
the
job.
A
A
B
A
Yeah,
so
you
have
to
work
around
those
issues
and
it's
not
something.
That's
first
thought
of
right
because
we've
moved
the
execution
outside
the
network,
so
you're
almost
always
going
to
hit
some
sort
of
double
hop
issue.
If
you
need
to
do
what
you
guys
just
asked,
I
need
to
pass
and
do
things
if
you're
passing
between
scripts-
maybe
not.
But
if
I
need
to
now
go
from
this
box
to
other
boxes
and
then
do
things
it
can
get
to
be
a
challenge.
A
But
what
you
can
do
is
what
I
did
with
the
credentials,
because
when
I'm
running
the
script
here
and
I'm
using
a
credential
object
and
I'm
authenticating-
I'm
not
being
penalized
for
this
first
hop,
I'm
on
the
network
and
authenticated
and
then
I
can
go
and
do
what
I
got
to
do.
So
you
just
wanted
to
point
that
out
because
it's
often
overlooked.
A
I,
when
we
talk
about
log,
what's
being
logged
right,
because
this
is
just
the
execution
of
the
job,
there's
really
nothing
being
logged
in
here
about
the
job
about
the
scripted
executed,
like,
for
example,
at
work,
one
of
the
jobs
that
we
run
is
we
are
synchronizing
our
workday
data
with
active
directory
because
we're
hybrid-
and
we
can't
do
it
through
the
cloud.
So
we
had
it
design
our
own
sync
script.
A
B
So
in
my
script
I
do
a
lot
of
write
output
just
to
yeah.
You
know-
and
I
guess
part
of
that
is
that's
a
good
point
like
for
me.
It's
part
partially
debugging
in
case
something
didn't
happen.
I
want
to
go
back
and
comb
through
all
the
logs
and
see
on
that
particular
run.
Did
it
pick
up
this
or
whatever
and
yeah?
I
max
out
this?
So
maybe
I
should
come
up
with
a
better
logging
solution
too.
Instead
of
writing
to
the
console
outputs.
A
So
what
we've
done
is
we
in
that
situation?
Is
we
built
a
log
writing
utility
that
just
outputs
what
happened
on
each
line
to
a
text
file
and
that's
happening
now?
The
log
output
is
happening
on
the
local
machine
that
the
script
is
being
executed
on,
which
is
basically
here
rather
than
out
in
the
cloud.
So
the
log
exists
here.
B
A
So
I
have
the
script
that
we're
running,
that
does
that
sync
is
400
lines,
but
it
doesn't
output
a
bunch
of
stuff
to
the
console.
It's
just
doing
the
ad
updates,
so
it's
probably
not
as
buffer
intensive
as
as
you
might
be,
but
I'm
having
no
problems
with
scripts
that
are
at
three
and
four
hundred
lines
right
now.
A
This
may
not
be
the
perfect
solution,
but
I
think
this
is
a
whole
hell
of
a
lot
better
than
the
test
scheduler
and
that's
the
focus.
I
think
we
need
to
keep
here
now,
maybe
if
you're
doing
devops
oriented
stuff-
and
you
want
to
go
down
that
path-
and
maybe
someone
can
talk
about
that
better
than
I
could.
But
if
we're
talking
about
executing
a
job
on
a
windows
server
and
oh
crap,
the
server
went
down
last
night.
None
of
our
jobs
ran
that
kind
of
stuff.
A
A
B
No,
I
think
one
of
the
questions
that
ryan
brought
up
is
just
interesting.
It's
just
the
idea
that
sometimes
he
sets
up
scheduled
to
ask
to
run
at
startup
on
the
server
kind
of
stuff,
but
that's
really
not
what
this
is
looking
for.
He's
thinking
of
trying
to
get
the
pipeline
scheduler
to
hook
into
some
of
those
events,
but
I
don't
think
there's
a
way
to
trigger
that.
Or
do
you
know
if
there's
a
way
to
trigger
a
pipeline
remotely?
A
B
Now,
I
think,
really
what
he
was
looking
is.
He
thought
a
task.
Scheduler
has
a
replaced
scheduler,
but
this
is
almost
like
a
centralized
task
scheduler
in
that
it's
an
overarching,
not
hey,
just
assassin
running
on
some
random
machine,
but
then
also
the
idea
somebody's
brought
up
the
idea
of
using
kicking
the
trigger
off
from
a
web
hook
like
you
could
do
that
you
can
ping
a.
A
Website,
yeah,
I'm
sure
you
can
do
all
that.
I
haven't
personally
done
that
again.
I
know,
there's
a
lot
more,
that
can
be
done.
We
are
basically
running
regular
jobs
at
regular
intervals.
To
do
things
that
we
don't
want
to
have
to
do
manually.
That's
the
intention
for
most
of
the
work
we
do
and
that's
also
a
byproduct
of
the
guys
that
are
on
the
team.
A
A
And
as
corey
meant,
oh
jeremy
mentions
here
if
some
of
the
stuff
could
be
replaced
with
azure
functions,
maybe
that's
a
better
solution
for
some
of
the
stuff
right,
so
we've
gone
through
a
pipeline
and
how
to
configure
one.
I
can
run
some
of
these
pipelines
and
show
you
different
examples
of
what's
going
on.
A
My
concern
is
that
the
code
that
executes
is
not
so
exciting
to
watch
as
it
is
to
build
the
jobs,
so
I
think
maybe
we
would
want
to
go
through
building
some
jobs,
but
it's
really
an
open-ended
forum
here.
It's
what
the
group
prefers
I'd
be
happy
to
do.
I'm
ready
to
run
these
and
show
you
different
things
you
can
do,
but
honestly
you're
going
to
say.
Okay,
I
created
an
ad
user
report.
Big
deal,
it's
the
building
of
the
job.
That's
the
interesting
part
to
me.
B
B
A
Or
start
here,
because
when
you
go
into
this,
oh,
I
might
have
to
ask
tv
for
some
help
here
where
the
hell
is
the
job
view
yaml
and
now
so
now
I
have
the
thing
so
obviously,
with
a
bigger
job,
there'll
be
more
to
see,
but
now
I
can
take
this.
I
can
paste
this
into
a
file
and
I
can
build
upon
this,
but
if
you're
not
comfortable
with
yaml
starting
here
might
be
hard.
A
So
what
I
do
is,
I
usually
start
and
get
through
the
gui
and
then
when
the
job
is
done,
I
copy
out
the
ammo
and
then
build
on
that.
So
what
I
would
do
here
is
if
I
need
to
do
a
more
complex
schedule.
I
would
definitely
want
to
do
it
in
the
ammo,
because
I
can
use
quran
and
the
cron
syntax
is
as
cryptic
as
it
comes,
and
I'm
not
going
to
be
able
to
copy
the
type
that
up
right
off
the
top
of
my
head.
A
A
A
A
B
However,
as
soon
as
you
want
to
do
a
little
bit
more
with
it
like
what
you're
talking
about
when
you
want
to
start
chaining
things
together,
then
you
should
use
the
builds
to
generate
your
content,
and
then
you
use
the
releases
to
actually
do
the
work.
And
so
then
you
can
do
things
like
you
can
build
modules
inside
your
build
and
then
based
on
how
successful
those
are.
B
Okay,
so
at
that
point,
so
some
of
the
things
that
you
disabled
initially
to
kind
of
basically
simplify.
This
is
that,
if
you
want
to
get
more
advanced,
the
advanced
pieces
are
available
to
you,
but
you
just
you'll,
have
to
just
enable
all
those
things,
and
so
that's
what
you
get
when
you
move
to
the
2.0.
A
Yeah-
and
I
think
this
is
essentially
the
classic
interface
problem
that
happens
with
very
granular
tools.
Where
do
you
start
there's
a
lot
of
possibilities
and
a
lot
of
choices,
so
I've
started
at
the
pipeline
level
replicating
what
I
view
as
like
a
scheduled
job
stephen,
who
was
just
talking,
views
this
same
tool
in
a
completely
different
manner
and
looks
at
from
a
release
level
and
artifacts
and
builds
that's
not
my
use
case.
So
I
don't
even
see
that
in
the
tool
as
a
way
of
operating.
A
So
one
thing
I
did
want
to
show
is
invoking,
because
that
kind
of
is
how
I
get
it.
One
way
I
get
around
my
double
hop
thing,
so
maybe
that
would
be
interesting
to
people
just
to
bring
you
up
to
speed
here.
I
have
two
scripts
here
now:
three
jobs
excuse
me
start
and
stop
a
service
right.
That's
a
common
thing
that
you
might
want
to
schedule
or
checking
on
services
on
multiple
machines
and
the
way
that
I'm
doing
that
here
and
just
get
that
script
open
and
have
an
open.
A
So
is
exactly
what
we
said
before
passing
in
a
user
id
and
password
creating
a
credential
object,
and
then
remember
this
is
going
to
execute
on
that
box.
So
from
that
build
box,
I'm
going
to
invoke
to
my
dc
and
in
this
case
start
a
service,
and
I
need
credentials
to
do
that.
So
I
pass
in
credentials
now.
This
list
could
be
extensively
long
and
I
could
easily
invoke
too
many
boxes
and
do
things
same
thing
with
my
stop.
A
A
That's
why
I
said
none
of
these
scripts
are
super
exciting,
it's
the
concepts
behind
them
that
I
want
everybody
to
understand
now.
I
have
not
even
mentioned
running
stuff
in
the
cloud,
but
it's
essentially
the
same
process
now
my
pipeline
for
the
cloud
isn't
working,
but
it
doesn't
really
matter
because
scheduling
the
job
is
what
we're
interested
in.
So
if
I
go-
and
I
look
here
and
I
say,
enable
azure
id
and
when
I
say
it
doesn't
work
because
I
haven't
set
up
a
permissions,
I
can
actually
run
this.
A
It's
just
going
to
fail,
but
when
I
go-
and
I
edit
this
now-
this
is
a
cloud-based
job,
azure
pipelines.
Okay,
it's
not
very
different
now,
because
I'm
in
azure,
I
have
a
different
agent
right.
So
I
have
some
agents
that
I
can
choose
here.
This
is
the
default,
that's
what
I
went
with
and
since
I'm
now
running
in
the
cloud,
I
have
to
add
what
I
want
to
do
with
it.
Instead
of
powershell,
I
added
azure
powershell
and
the
only
thing
it
asked
for
was
after
I
did
that
it
wanted
a
service
principle.
A
A
This
job
will
run
but
it'll
fail
because
the
service
principle
I
have
running
it
doesn't
have
credentials
to
finish,
but
and
it'll
execute
and
it'll
be
the
exact
same
thing.
If
I
go
here
and
I
go
into
the
history
and
you'll
see
here,
it
takes
a
little
longer
to
run
in
the
cloud,
because
what
it's
doing
is
just
spinning
up
an
instance,
real
quick.
It's
setting
up
the
modules
right
and
then
it's
kicking
off
stuff
here.
So
script
takes
a
little
bit
longer
to
run,
but
not
painfully
slow
right.
A
So
I'm
concerned
that
this
may
have
went
over
some
people's
heads
and
I
want
to
make
sure
that
we've
covered
the
basics
before
we
even
attempt
anything
else.
This
could
be
enough
to
consume
for
one
evening
right
here.
It
really
depends
on
how
well
this
came
across
the
people
and
since
I've
been
doing
this
every
day
for
the
last
eight
months,
this
all
seems
second
nature
to
me
and
I
don't
really
have
a
problem
with
it,
but
I
know
that's
not
the
case
for
someone
who's
glancing
at
this.
A
B
A
Microsoft
recently
changed
their
policy
and
when
you
go
into
the
pools
it
says,
there's
a
hosted
agent,
but
somewhere
along
the
lines,
they
pulled
back
the
ability
for
everybody
to
run
free
jobs
in
the
cloud.
So
now
what
happens
is
when
you
go
to
run
it?
It
says:
hey,
you,
don't
have
an
agent
that
can
do
parallelism.
A
Would
you
like
to
apply
for
a
free
agent
and
you
fill
out
a
form
and
it
takes
about
seven
business
days
and
they
will
give
you
a
free
agent,
so
it
is
essentially
free,
there's
1800
minutes
of
run
time
a
month.
That's
running
one
job!
At
a
time,
though,
if
you
need
more
parallel
jobs
to
run,
you
need
more
agents
and
agents
can
get
a
little
pricey.
I
think
they're
15
a
month
per
agent,
but
I'm
not
100
sure
on
that
and
each
purchase
includes
runtime
and
it's
based
on
the
amount
of
runtime
on-prem.
B
A
So
I'm
assuming
that's
the
default,
and
I
don't
know
if
there's
a
programmatic
way
to
get
around
that
or
if
that's
just
built
in,
but
that's
what
I've
seen
in
our
use
case
now
going
back
to
the
cost.
So
there's
a
cost
to
run
things
in
the
cloud,
but
you
can
apply
for
a
free
free
agent
and
then
you
can
run
stuff.
A
A
B
I
think
it's
funny
earlier.
You
talked
about
you
mentioned
how
some
people
might
think
this
is
better
to
do
an
azure
function
or
whatever.
I
completely
disagree,
I'm
trying
to
look
at
this
and
see
that,
in
fact,
I
think
I
saw
where
one
of
your
options
was
to
update
a
function.
I
would
like
to
take
my
function
and
put
it
into
azure
devops,
where
basically
here's
the
code
in
a
repo
make
use
of
test
branches
and
whatnot
and
yeah
when
there's
a
change
to
the
master
branch
go
run
that
job
yep
stuff
like
that
yeah.
A
I
agree,
and
I
will
tell
you
that
when
you
go
into
a
job
here,
the
way
for
this
demo,
I
didn't
schedule
anything
to
fire
on
a
trigger,
but
yeah.
Definitely
that
option
is
absolutely
there.
If
you
want
to
work
on
a
change
of
code,
fire
off
a
job,
that's
a
cool
way
to
go
and
it
does
have
the
ability
to
support
branches
and
different
kinds
of
things.
A
The
problem
is,
I
think,
for
me
is
we
could
probably
spend
a
lot
more
time
talking
about
this,
but
I
think
there's
a
segment
of
the
people
that
are
on
the
call
that
this
is
never
been
seen
before,
and
this
was
a
lot
to
absorb
and
then
there's
a
segment
of
people
on
the
call
that
have
seen
some
of
this
before
and
are
ready
to
go
on
to
201
and
301
stuff.
And
I
don't
want
to
make
the
mistake
of
trying
to
stuff
too
much
into
one
meeting.
A
I
think
we
could
come
back
and
we
can
do
deeper
dives
on
this
stuff.
But
I
think
the
point
that
I
was
trying
to
make
here
is
that
pretty
much
any
admin
that
runs
a
windows
shop
could
probably
get
away
setting
this
up
reasonably
in
an
hour
or
two
fumbling
around
the
settings
or,
if
there's
some
good
documentation
and
be
running
jobs
from
a
cloud-based
automation,
engine
or
orchestration
engine,
as
jason
says,
and
I
think
that
is
probably
the
best
way
to
look
at
it.
A
B
One
of
the
other
options
there,
as
huge
said,
is
you
built
a
windows
system
as
your
build
agent,
but
you
can
certainly
have
a
linux
box,
be
your
build
agent
or
something
else
in
powershell
core
and
so
not
to
even
say
that
anybody
here
who's
just
an
ill
hey.
If
I
just
want
to
use
this
account
here-
and
I
want
to
kick
off
jobs
on
some
linux
box
to
do
some
other
stuff
that
happens
to
be
somewhere
else.
That
is
certainly
possible
for
you.
A
A
B
A
No,
the
only
other
thing
I
was
going
to
talk
about
was
that
invoke.
If
anybody
wants
to
see
that
we
can
run
that,
but
I
was
wondering
myself
if
maybe
we're
in
a
natural
place,
to
wrap
up
here,
but
the
idea
here
is
that,
oh
so
there's
one
thing
I
didn't
cover
on
that
build
box.
Okay,
this
build
box
is
essentially
running
the
jobs
that
you
want
to
schedule
out.
So
I
only
talked
about
installing
the
agent,
but
don't
forget
this
box
also
needs
the
tools
that
you
need
to
run.
A
So
I
didn't
show
it
tonight,
but
I
installed
the
rsat
tools
on
this
box,
so
I
can
run
the
ad
commands.
If
you
need
other
modules,
you
need
to
install
them
on
your
build
boxes.
Now
you
could
schedule
that
to
a
job,
to
install
those
agents,
install
those
modules
and
things
like
that,
but
I
didn't
mention
it
is
worth
mentioning.
The
build
box
runs
an
agent,
but
it
also
has
the
tools
on
there
to
do
the
things
that
you're
going
to
schedule
on
it.
A
The
reason
it
teams
here
is,
I
pulled
down
office
last
night
on
this
box
because
I
ran
a
user
report
and
I
wanted
to
just
open
it.
We
didn't
even
get
to
that
and
then
that's
okay,
if
we
don't
but
keep
in
mind,
your
build
boxes
have
to
have
the
tools
and
able
to
run
the
commandlets
that
you're
going
to
push
down
to
it
to
execute.
A
A
This
is
probably
a
good
place
to
break.
Let
me
say
thank
you
for
watching
tonight,
if
you're
watching
at
home,
there's
more
content
on
our
youtube
channel
where
you're
seeing
this
so
take
a
look
at
our
other
videos.
We
have
content
on
most
of
the
topics
that
relate
to
powershell
we'd
love.
To
have
you
stick
around
and
see
some
more
and
if
you
get
the
chance
to
come
and
join
us
for
a
live
meeting,
you
can
see
us
on
meetup.com
we
meet
on
the
first
and
third
wednesday
of
the
month.