►
Description
Presented by Anubhav Mishra, Technical Advisor to the CTO, HashiCorp
Speed is an important success metric for modern development teams. As enterprises start to scale in multi-cloud environments, the Continuous Integration/Continuous Deployment (CI/CD) process is no longer just a developer specific workflow, but a DevOps workflow where both development and operations coordinate their tasks to move forward rapidly. GitHub and HashiCorp are working closely to build an end-to-end DevOps workflow encompassing the entire application development lifecycle. Together we help provide a powerful CI/CD solution that prevents developers and operators from having to wait for their respective tasks to be completed, accelerating delivery. Join this session to learn and see how an updated GitHub Actions and HashiCorp Terraform workflow can simplify collaboration and increase your development team’s velocity through infrastructure automation and applications both delivered as code and on-demand.
GitHub Satellite: A community connected by code
On May 6th, we threw a free virtual event featuring developers working together on the world’s software, announcements from the GitHub team, and inspiring performances by artists who code.
More information: https://githubsatellite.com
Schedule: https://githubsatellite.com/schedule/
A
B
So
our
next
speaker
is
a
DJ,
but
we
didn't
have
him
come
on
to
be
on
the
turntable,
actually
he's
the
technical
advisor
to
the
CTO
of
hashey,
Corp
and
I'm
so
excited
about
this,
because
we
use
Hoshi
court
here
at
github.
So
I'm
really
excited
to
hear
what
he
has
to
say:
yeah,
the
name
of
his
talk
is
collaborative
application
and
infrastructure
workflow
using
github
actions
and
hashey
cork
terraform.
Please
please
help
me
introduce
the
one
and
only
Mishra.
C
C
I
apologize,
my
mic
was
muted,
say
hello,
everyone
say
I'd
really
really
appreciate.
You
know
having
me
at
the
conference.
This
is
one
of
the
confidence
that
look
you're,
gonna
look
forward
to
presenting
at
for
a
while
now.
So
this
is
should
be
an
exciting
talk,
so
today
I'm
going
to
be
presenting
collaborative
application
and
infrastructure
workflow
using
get
up
actions
and
how
she
could
terraform
switch
and
we
talked
about
how
get
up
actions
and
hashcode
perform
work
together,
and
we
will
talk
about
a
few
workflows.
C
So
yeah,
it's
an
exciting,
exciting
gig,
so
in
terms
of
open
source,
I
have
contributed
and
created
a
few
projects.
One
of
them
is
Atlantis
which
helps
you
kind
of
cool,
a
bit
of
infrastructure
using
terraform
and
then
I'm
also
provide
a
mangina
for
virtual
cubelet
and
more
recently,
I
gotten
into
the
new
projects
that
have
come
out
from
Microsoft
open-source.
One
of
them
is
Stafford,
which
allows
you
to
kind
of
distribute
Li
build,
distributed.
C
Applications
I've
contributed
a
few
things
there,
but
the
most
exciting
part
of
my
job,
you
know
in
the
past
few
years
has
been
creating
a
podcast
for
Hoshi
cos.
My
co-host
that
filled
a
bunch
of
folks
at
her.
She
cast
her.
She
called
and
developer
advocacy
team,
ok
a
bit
about
her
she
code,
but
for
those
who
don't
know
about
the
company
they're
a
cloud
infrastructure,
automation,
company,
we
focused
on
four
different
areas:
provisioning
and
securing
applications
connecting
them
and
then
running
them
at
scale.
C
So
we
have
vegan,
which
is
a
developer
environments.
Tool
helps,
you
know,
create,
develop
environments
share
them,
you
have
Packer,
which
is
a
silver
imaging
tool.
It
helps
you
kind
of
create
several
images
on
RAM
cloud
we
have
terraform,
which
is
the
infrastructure
as
code
tool.
You
know
kind
of
you
can
drag
things
in
plain
text
files
and
make
them
happen,
and
this
is
this
is
kind
of
the
focus
of
the
talk,
and
then
we
have
what,
which
is
a
secrets
management
tool,
helps
you
kind
of
manage
secrets
in
one
place.
C
I
can
help
you
with
that
crazy
secret
sprawl
that
we
see
in
our
data
centers,
and
then
we
have
console,
which
is
a
distributed
service
mesh,
a
kind
of
helps
you
connect
application
securely
and
gives
you
the
l7
routing
and
Canary
deployments,
and
things
like
that.
Then
we
have
Nomad,
which
is
a
journal
purpose,
cluster
scheduler,
very
in
line
with
kubernetes,
but
it
can
actually
schedule
containerized
and
non-continuous
applications
and
even
vm.
C
So
if
you
like,
that
was
a
bit
of
a
hardship
so
kind
of
taking
a
step
back
before
you
get
into
github
actions
and
formally
gonna
take
a
look
la
what
is
happening
in
the
industry
right,
like
the
Hashi
core
view
right,
which
is
there's
this
transition
into
the
box
and
the
bit
the
big.
The
big.
The
big
change
that's
happening
is
we're
going
from
this.
Like
traditional
data
center
approach,
in
which
you
had
you
know
hardware
you
have
to
manage,
like
you
know,
networking
cooling,
you
know
black
them
up
figure
out
power
into
things.
C
Things
like
you
know
the
public
cloud
right.
We
have
the
the
modern
data
center
approach
in
which
things
have
you
know
dynamic
in
nature.
They
can
come
up
and
down
and
you
can
just
sign
up
for
a
cloud
provider
account,
and
you
know
you
get
you
get
access
to
amazing
computing.
You
can
do.
We
can
scale
up
applications
fairly
easily.
So
this
is
the
big
change
right
in
terms
of
process.
C
We're
going
from
this
like
very
ticket
driven
approach
in
which
you
had
to
create
tickets
and
then
work
on
them,
and
then
you
know
wait
for
hardware.
Wait
for,
like
you
know
the
teams
to
line
behind
them
into
this
idea
of
like
devops
right,
maybe
use
the
DevOps
philosophy.
Kind
of
you
know
do
things
which
are
which
you
know,
drive
systems
that
are
API
driven
and
give
you
like
that
cell
service.
C
So
why
are
we?
You
know
why
this
chef,
like
what
is
it
there's
many
reasons
for
this,
but
one
of
the
critical
reasons
is
increasing
developer
velocity,
which
is
how
how
fast
can
be
shipped
to
production.
How
fast
can
we
ship
a
feature
to
production,
kind
of
empower
the
developers
and
kind
of
make
an
efficient
use,
and
you
can
make
the
time
to
plan
very
efficient
and
what
helps
you
achieve
this?
C
This
is
the
kind
of
a
DevOps
philosophy
right
you
have
your
core
teams
like
operations,
platform
teams,
security
teams,
kind
of
build
ap-
is
that
the
developers
could
use
all
right.
So
this
is
where
we
kind
of
introduced
era
foam,
and
then
we
talked
about.
If
you
guys
talk
about
making
things
happen,
we
introduced
Hasek
up
there.
Foam
which
is
free
to
use
free
to
download,
and
the
the
primary
goal
of
terraform
is
to
kind
of
provide
you
with
this
unified
workflow
using
infrastructure
as
code.
So
that's
the
core
of
the
of
the
product.
C
The
idea
here
is
to
kind
of
have
you
know
you
write
the
index
files
and
make
them
happen
and
use
the
same
software
engineering
philosophy
that
we've
been
using
for
a
while
in
and
pull
that
into
infrastructure
right
like
and
and
use
it
to
like
version
infrastructure
and
kind
of
iterate
over
it
safely.
The
idea
here
is
to
go
from
one
stage
to
the
other.
C
You
know
kind
of
a
safe
manner,
so
from
nothing
to
something,
and
maybe
they
want
a
to
operations
and
then
the
the
key
thing
here
is:
if
you
want
to
be
provisioning,
anything
anywhere
and
then
I
mean
anything
anywhere.
I
literally
mean
it
like
on
trim
cloud,
DNS
providers,
CDN
providers.
If
you
have
an
API
that
can
do
crud
operations,
you
should
be
able
to
work
on
that
music
there.
C
So
how
does
provisioning
work
whatever
form?
So
you
have?
You
have
guitar
foam
core
and
telephone
provider,
so
as
a
user,
what
you
see
is
you
know
you
just
download
this
binary
and
use
it,
but
in
real
life
there
is
the
turf
form
core
that
is
responsible
for
operations
like
you
know,
parsing
the
config
file.
C
You
know
and
doing
things
like
state
management
and
stuff
like
that,
and
it
doesn't
really
know
about
like
the
implementation
details
of
the
provider
which
is
like
you
know,
how
do
you
manage
a
server
or
so
on,
and
then
the
providers
are
the
ones
that
actually
makes
the
API
calls
to
Amazon,
github
and
kubernetes,
and
so
on.
You
have
200
plus
providers
that
are
open
sourced
and
a
build
on
top
of
the
open
source
model.
C
We
have
1200
plus
contributors
that
actually
maintain
and
help
us
kind
of
drive
that
ecosystem,
which
is
pretty
so
one
of
my
favorite
examples,
is
to
use
github
to
use
terraform
to
manage
github,
which
is
kind
of
meta
in
this
case,
so
even
at
hajikko,
what
we
do
is
we
create
repositories.
We
manage
permissions
all
using
data
form
and
can
provide
this
like
really
simple,
Etsy,
a
language
where
you
can
write
like
what
you
want.
In
this
case
we
are
creating
a
repository
and
maybe
assigning
a
certain
troll
to
a
member
right
fairly
easy.
C
So
a
typical
workflow
for
a
terraform
user
is
like
for
a
single
user.
Maybe
smaller
team
is
as
a
developer
and
operator
you
just
download
the
binary,
and
you
write
some
tear
form.
The
first
thing
you
run
is
something
called
a
tear
form
that
that
helps
you
preview.
What's
gonna
happen
before
it
actually
happens.
So
in
this
case
you
can
see
we
are
adding
one
github
repository
and
you're
able
to
kind
of
preview
things.
C
This
is
just
a
video
only
operation
that
doesn't
really
do
anything,
and
this
is
how
you
can
have
safely
iterate
over
it.
It's
structure.
So
in
this
case
you
see
now
we
can.
Actually,
if
you
diffuse,
this
plan
looks
good
to
you.
You
can
go
ahead
and
apply
it
and
you
can
use
the
plan
as
in
an
input
for
the
apply
and
only
that
plan
will
be
executed.
C
This
is
where
the
fun
kind
of
gives
you
that
guarantee
that
you
know
only
the
change
that
you
want
to
go
out
will
go
out
and
it's
in
real
life.
Something
changed.
For
example,
like
someone
went
to
the
Amazon
console
and
github,
you
I
hated
something
which
you
can
fix
with
this
plan,
then
this
apply
will
safely
fail
and
you
don't
get
your
resource,
which
is
what
we
want
in
reality
right
and
if
everything
works,
we
get
your
Amazon
server.
You
get
your
cube
in
any
spot.
You
get
your
you
know.
It
has
a
tree.
C
None
of
this
happens
using
the
CLI
as
well
on
your
computer.
This
is
for
like
a
1:1,
you
know
one
user
may
be
a
smaller
team,
so
how
do
collaboration
work
with
terraform?
So
this
is
where
I
can
the
whole
time
ops
philosophy
something
to
play.
Let's
say
a
developer
joins
an
organization.
You
can
create
a
bunch
of
HCl
and
say:
hey
I
want
this
role
for
the
sweep
or
something
like
that.
A
quest
for
kit
of
ashes.
C
C
So
that
was
the
workflow
that
you
saw,
which,
which
is
basically
you
know
that
the
base
that
works
for
a
smaller
team-
and
maybe
a
group
of
you
know
a
smaller
team
or,
like
maybe
an
individual
or
two
people
kind
of
just.
You
know
where
you
have
to
coordinate
a
lot
of
changes
and
things
like
that.
So
how
does
you
know
the
larger
organizations
with
many
teams
work
with
cell
phones?
This
is
my
github
can
actually
help
right.
So
we
want
to
use
the
same.
C
You
know
pull
request
really
nice
data
book
so
to
achieve
infrastructure
and
application
changes
as
well
and
you'll
see
how
like
actions
coming
scums
is
to
play
as
well.
So
let's
say
we
have
a
developer
asking
for
up
the
M
or
something
and
they
push
certain
HCl
code
up
to
github.
Then
what
can
happen
is
you
can
actually
take
place
in
github
which
will
be
generated
using
you
know.
C
Maybe
some
CI
system
can
actually
have
a
little
fun
planning
with
you
give
you
that
plan
output,
and
maybe
an
Operations
team,
can
actually
review
the
review
the
plan.
If
that
looks
good
to
them,
they
can
actually,
you
know,
make
that
thing
happen
for
you
and
then
again,
some
system
has
to
run
terraform
apply
right
now.
Ideally,
we
would
want
the
whole
thing
to
happen
in
the
cloud.
So
this
is
an
ideal
situation.
C
So
this
is
where
is
like
the
perfect
integration,
and
most
of
you
know
where
I'm
going
with
this
I'm
really
happy
to
announce.
California
have
actions
1.0
version.
It
is
completely
revamped.
It
uses
the
new
JavaScript
framework
to
make
things
happen
and
I
really
want
to
thank
the
github
engineering
team,
the
Hasek
of
engineering
team
for
working
together,
I'm
just
a
face
to
this.
This
amazing,
you
know
collaboration,
so
here's
a
quick
demo,
I've
recorded
it
for
for
kind
of
time
purposes.
I
don't
have
much
time
here,
so
I
just
recorded
it.
C
I
love
doing
live
demos,
but
unfortunately,
this
one
is
recorded
and
it's
fast-forwarding
portions
to
save
time.
So
here's
what
we're
going
to
work
on
today
in
terms
of
the
demo
right.
This
is
the
overall
picture.
You
have
an
s3
bucket,
that's
just
hosting
a
static
site.
It
has
an
index
page
that
an
error
page.
C
Then
we
have
a
CloudFlare
dns,
that's
pointing
to
this
s3
bucket
and
both
of
these
things
are
managed
using
terraform
and
we're
using
the
new
get
up
and
telephone
action
to
make
things
happen
on
it
upside
and
we're
using
data
from
cloud
to
do
the
kind
of
the
remote
actions
and
things
like
that
so
fairly
straightforward
and
you
see
how
this
all
progresses.
So,
here's
the
user
demo
right
so
this
repo
that
you
see
here,
it's
already
open
source
I'll,
put
a
link
to
the
people
at
the
end
of
the
presentation.
C
You
can
check
it
out
yourself.
It
has
all
the
code,
all
the
things
that
you
need
to
kind
of
make
things
happen
in
this
case
the
HTML
code
and
the
telephone
code.
So
you
can
fork
it
and
try
it
yourself
if
you
like
somebody
cloned
this
repo
I,
can
open
visual
studio
code
and
start
working
on
this.
So
here
I
have
the
s3
bucket
defined
already
then
host
our
website,
which
is
public
and
then
I.
Have
my
I.
C
Have
my
IM
policy
set
up
to
expose
that
bucket
out
to
the
Internet
and
then
I
also
have
a
bucket
object,
which
just
is
uploading
the
index
dot
HTML
file
up
to
after
that
stream
market?
So
here
for
time,
I
provisioned
it
already,
and
this
is
all
managed
using
telephone
and
you
can
see
the
index
dot
HTML
file.
That's
already
present,
all
right!
So
now,
let's
say
you're
an
operator.
You
want
to
create
at
the
NS
record
that
points
to
this
packet
right.
C
So
here
we
have
the
flower
flavor,
for
whatever
way
you
can
manage,
you
know
sow
it
directly
from
directly
using
CLI
here
so
sorry
directly
using
the
HCL
code.
So
what
we'll
do
here
is
we'll
add
another
entry.
So
here
you
can
see
that
dub
dub
dub
entry,
that's
pointing
to
that
long,
s3
URL,
which
cannot
be
seen
and
everything
is
proxy
to
out
there
and
so
what
we
lose.
We'll
add
another
entry
which
is
for
the
subdomain
to
point
to
the
www
entry
as
well.
C
So
what
we
lose
will
I
just
add
another
resource
and
and
basically
use
the
pipe
off
your
app,
which
is
a
domain.
You
can
check
it
out
right
now.
It
should
be
live.
So
in
this
case
we
will
create
the
subdomain
that
points
to
the
dub
dub
dub
dub
right
and
we
are
providing
it
with
cloud
fair.
You
know,
zone
ID
and
email
NPI
keys.
Now,
if
you're
an
operator,
you
have
the
keys
to
the
kingdom,
you
can
pretty
much
like
run
a
telephone
plan.
C
You
have
the
clarity,
I
keys,
you
have
the
AWS,
you
know
API
key,
so
you
pretty
much
run
terraform
plan
and
this
will
work
like
as
a
as
a
sole
operator.
You
know,
have
the
keys
to
the
kingdom,
so
this
is
there
from
there
from
flower.
That's
running
the
plan
here,
so
you
see
we
are
adding
that
one
resource.
But
ideally
you
want
to
share
this
with
your
team,
like
you
want
this
to
happen
in
in
a
collaborative
way.
So
how
would
that
work?
So
that's
where
you
use
the
github
action.
C
D
C
C
Okay,
so
now
what'll
happen
is
we
just
create
a
pull
request
and
I
do
not
type
this
fast
I
wish.
I
was
I
always
like
that
fast
of
a
diaper
but
I'm
not,
and
so
you
switch
back
to
switch
back
to
the
repo
here
and
you
create
the
pull
request.
They
very
straightforward,
workflow
for
an
operator
to
create
the
pull
request
immediately.
This
should
fire
it
up
there
for
mation.
C
So
immediately
you
see
like
we
have
an
expected,
it's
a
check
and
make
sure
you
do
branch
protection
on
these
repos
so
that
you
know
only
approved
for
requests
can
get
merged
in.
If
you
see
our
change,
that's
a
record,
that's
being
added.
If
you
switch
back
and
we
open
up
the
details
for
that
terraform
action,
we
see
this.
This
is
perform
validate
in
it
and
plan
that's
running
at
fast
forwarded.
C
This
again,
you
can
see
we're
adding
that
one
record
as
soon
as
this
thing
finishes:
you'll
notice
that
the
dev
form
of
line
did
not
run
because
this
is
not
the
master
branch
all
right.
So
now
we
have
the
github
bot
that
comments
in
the
plan,
which
is
really
nice.
So
you
can
see
the
telephone
plan
what's
going
to
happen
before
it
actually
happens,
see
the
code
change
and
if
it
looks
good
to
you,
know
I
guess
the
teammate.
C
You
know
they
can
do
a
+1
I'm,
pretty
lonely
here,
because
this
is
just
my
my
repo
with
and
the
only
facility
quite
unfortunate,
but
that's,
ok
and
will
squash
and
merge
this.
This
change
in
and
remember
to
do
the
most
important
thing,
which
is
to
delete
the
branch
right
after
this.
Otherwise
you
have
this
dangling
dead
branches
right.
You
don't
want
that
in
your
repos
I'm
sure
your
teammates
will
appreciate
that.
So
now,
if
you
go
to
the
commits
here-
and
you
see
the
apply
actually
started
running
here-
checking
the
details.
C
We
run
all
the
steps.
Again,
we
don't
an
again.
You
don't
need
to
do
that.
You
can
just
run
apply,
but
in
this
case
I'm
running
plan
and
then
and
then
wait
for
the
fly
to
finish
and
what
we
can
do
is
we
can
actually
see
this
is
getting
triggered
in
inter
form
from
cloud
as
well,
so
you
don't
have
to
use
their
from
cloud
in
this
example
and
music
there
from
cloud
it's
fairly
easy
for
me
to
set
it
up.
This
is
kind
of
my
own
Prem
for
me.
C
So
in
this
case
it's
running
telephone
apply
as
soon
as
this
thing
finishes.
It
takes
about
a
minute
or
so
and
fast
forwarded
this.
For
times
sake,
we
can
go
back
to
go
back
to
our
master
branch
and
make
sure
the
the
checks
are
fast
and
you'll
see.
There's
a
green
tick
mark.
That
means
I
apply
successfully
finished.
C
We
go
back
to
cloud
fair
with
your
fresh
DNS
provider
and
we
see
a
subdomain
entry
pointing
to
the
dub
dub
dub
by
Farid,
our
app
domain,
which
is
again
proxy
through
cloud
flat,
which
is
awesome,
and
if
I
refresh
here,
we
see
that
this
is
working.
I
can
order
more
coffees,
more
and
more
coffees.
Keep
placing
those
orders
I'm
sure
someone's
gonna
like
break
this
I
am
really
bad
in
UI
and
UX
design.
So
this
is
the
best
I
could
come
up
with,
but
I
hope,
I
hope,
I,
hope
this
works
so
here.
C
So
if
you
go
to
any
other
page
for
that
application,
you
see.
There's
a
404
page
right
like
this
is
a
basic
s3
page.
You
don't
really
want
that.
Let's
say
developer
wants
to
change
the
404
page
right,
so
they
can
easily
add
another
object
and
they
don't
have
to
use
the
you
know
clone
the
repo
and
stuff
like
that.
C
You
can
do
this
using
the
github
UI,
but
in
this
case
I'm
actually
I'm,
actually
showing
you
the
full
end-to-end
workflow
here
so
here,
I
just
defined
interface
index
and
use
the
error,
dot,
HTML
file,
fairly
straightforward
and
nothing
complicated
again.
I
create
my
I
check
out
a
new
branch
and
create
a
new
pull
request.
Ok,
so
this
is
how
I
type
this
is,
how
slow
it
is,
and
this
I
do
a
lot
of
mistakes
as
well
as
you
can
see
right
there,
and
this
is
my
normal
speed.
C
Okay,
so
I
can
push
this
branch
up
to
upstream,
go
back
to
my
repo
and
and
we
checked
the
and
we
treated
the
pull
request.
You
know
create
the
pull
request,
hit
the
create
the
pull
request
button
here
and
then
that
will
fire
off
the
get
up
action
again
same
workflow
for
the
developers
as
well.
As
you
know,
this
same
repo
in
this
case
in
have
multiple
repos.
That's
alright,
you
know
you
can
now
you
can
wait
for
like
this
check
to
pass
and
what
can
happen
is
like
a
lot
of
times.
C
Things
don't
go
exactly
right.
Like
things
are,
things
are
you
know
invalid
sometimes,
and
is
the
validator
minefield,
and
we
want
to
see
why
it
fit
like
you
want
to
see.
The
check
has
failed.
We
can
merge
this.
Let's
go
back
and
see
that.
So
when
we
see
the
log,
we
see
there's
a
duplicate
name
for
the
object.
The
s3
bucket
object,
so
you
can
easily
fix
that
developers
don't
have
to
clone
the
repo
again,
it's
all
in
one
context,
you
have
just
the
pull
request.
C
You
create
you
just
edit
this
file
right
here
and
you
fix
the
error,
which
is
a
simple
dipole
of
the
copy
pasting
errors
that
we
all
have
had
in
our
lifetime.
So
we
fix
that
by
just
changing
it,
making
it
unique
I'll,
just
call
it
edit
commit
to
the
same
branch.
It
really
triggers
the
the
github
action
again.
Hopefully
we
scroll
that
yeah
immediately
the
action
goes
and
if
it's
successful
you'll
see
a
valid
plan
in
the
pull
request,
and
this
is
fairly
easy
again,
the
same
workflow
for
developers
and
operators.
C
You
see
right,
there
is
a
plan.
You
can
see
the
plan
right
there.
What's
gonna
happen
before
it
actually
happens.
The
operation
operations
folks
look
at
that.
They
look
at
the
file
change.
They
say
it's
a
fairly
small
change,
but
it's
safe
to
do.
They
can
give
us
a
thumbs
up
right,
but
their
thumbs
up,
and
you
squash
and
merge
it
remember
to
delete
the
branch
go
back
to
the
repo
we'll
wait
for
the
apply
to
finish
here.
I've
fast
forwarded
this.
The
finish
is
quite
quickly
here
and
we
we
scroll
back
there.
C
You
refresh
this
page
and
see
their
ever
face
right
there,
it's
just
perfect,
which
is
what
you
want
to
see.
That
was
a
quick
demo
of
of
using
the
thing
end-to-end
for
developers
and
operators.
Now
that
might
be
maybe
a
bit
too
much.
You
can
actually
immediately
use
our
starter.
Workflow,
that's
available!
You
just
go
to
your
actions.
Tab
you'll,
find
that
the
setup
terraform
CLI
feature
there.
C
C
You
can
go
to
this
URL,
which
is
get
out
there
form
it
have
actions,
and
you
can
find
the
demo
and
my
slides
they're
already
open
sourced,
and
you
can
find
them
at
get
iota
from
get
upset
like
demo,
and
if
you
wanna,
learn
more
about
how
github
and
Hasek
are
working
together
in
terms
pallets
and
and
what
a
vision
is
we
have
a
white
paper,
that's
available,
it's
called
increasing,
develop
of
velocity
in
the
model
and
I
really
really
appreciate.
Having
me.
Thank
you.
C
A
Thank
you
so
much
Vishal
that
fantastic
presentation.
It's
really
really
cool
to
see
how
the
integration
of
github
actions
is
happening
with
so
many
cool
things
now
for
jump
into
questions.
We've
got
a
fault
first
on
the
screen.
Welcome
to
John
one
of
our
own
from
github.
He
is
an
expert
in
github
actions
and
we're
very
excited
to
have
John
here
answering
all
the
amazing
of
questions
are
coming
in
about
github
action.
So
we
have
time
for
a
couple
of
questions.
Jared
you've
been
on
the
discussions
before
you.
B
A
B
B
D
Sure
so
we
provided
a
starter
workflow.
That's
now
available
from
the
actions
tab
of
any
repository
that
you
own
you
had
there
and
you
see
that
the
terraform
sort
of
work,
so
this
just
recommends
a
way
to
get
started
them
and
of
course
you
can
edit
that
and
change
it
as
you
need
to,
but
we
think
that's
a
good
place
to
get
started.
B
C
Yeah,
so
that's
that's.
What
ends
up
happening
in
the
in
the
actual
work
flow?
Actually,
so
you're
able
to
run
the
plan
in
the
pull
request
before
it
actually
merges
in
and
I'll
show
you
like,
you
know
you
can
actually
get
it
commented
in
in
my
demo,
I've
shown
out
how
you're
able
to
just
use
another
action
which
basically
does
a
create
comment.
You
can
take
the
output
from
the
deform
plan
positive
that
action.
That
action
should
be
able
to
create
the
comment
she
able
to
gonna
preview.
C
What's
gonna
happen,
and
then
you
can
add
your
own
actions
that
you
can
do
many
things
for
the
plan,
if
you
like
so
like
Dan
output
in
in
our
case
is
something
that's
human
readable,
but
you
also
have
adjacent
options,
so
you
can
actually
like
dump
it
as
adjacent
file
and
you
can
analyze
and
do
things
with
it
and
you
can
write
anything
anything
you
want
actions
just
allows
you
to
do
pretty
much
anything
you
want.
This
is
pretty
awesome
to
have
this
like
really
like
amazing
flexibility
and
pretty
much
everything
it
runs
in
parallel.
C
A
C
A
My
favorite
things
about
actions.
Is
it
like
if
something
doesn't
exist,
you
know
like
essentially
just
build
an
action
to
like
just
make
it
happen,
and
it's
just
it's
just
so
cool
like
I'm,
so
glad
we
added
in
this
picture.
This
is
amazing.
Just
gives
developers
like
all
the
powers
to
do
all
the
things
yeah
John.
Do
you
have
any
more
questions
coming
through
I.
B
Think
to
try
and
get
as
many
as
we
can.
You
know
have
all
the
time,
but
I
think
you
were
mentioning
it
before
me.
Sure
I
mean
we
actions
has
been
out
for
a
couple
years
now
and
now
it
is
the
most
popular
CI
possible.
So
I
think
it's
just
it's
amazing
what
people
are
able
to
do
with
actions
and
I'm
so
happy
to
see
what
hashey
Corp
has
done
as
well.
Now
enough
about
me,
I
keep
talking
too
much.
C
Cool
this
is
this
is
an
interesting
one.
Yeah,
so
I
have
some
comments
here,
so
I
actually,
prior
to
joining
Hoshi
Corp,
like
I,
was
already
a
Packer
user,
so
backers
the
IQ
that
helps
you
create
server
images,
and
this
was
one
of
the
first
block
books
that
I've
ever
written
online
I
think
a
rich
create
basically
uses
backer
and
and
service
pack,
which
is
another
open
source
product
amazing
project,
and
you
can
use
them
hand-in-hand
and
kind
of
validate
packet
images
as
well.
C
So
you
can
actually
run
write
an
action
that
does
that
for
you
and
you
can
actually
use
terraform
to
spin-off
infrastructure,
that's
kind
of
test
infrastructure
and
basically,
let's
say
you
have
an
image
that
you
just
build
using
Packer
and
it's
like
Ubuntu
1604
for
your
web
server
or
something
like
that.
You
use
the
you
use
a
get
up
action.
You
know
trigger
terraform,
to
create
you
that
infrastructure
I
that
can
actually
run
your
service
pack
tests
and
then
provide
the
output
again.
Based
on
that,
you
can
fail
or
pass
the
full
request.
C
D
B
C
Yeah,
so
actually
most
of
our
products
with
like
pretty
much
all
our
products.
That
Hoshi
called
we're
a
very
cloud
agnostic
company,
so
we
launched
terraform
and
it
was
you
know,
I,
don't
know
how
many
years
ago,
but
it
was
already
you
know
basically
multi
a
cloud
agnostic.
You
could
use,
you
could
use
AWS,
you
can
use
Azure
and
you
can
use
CCP.
You
can
use
Alabama
card,
that's
200,
plus
providers.
So
if
you
go
down
the
list,
there's
pretty
much
all
your
you
know:
you're
happening
cloud
providers.
C
You
have
the
small
cloud
providers
as
well,
I,
just
illusion
things
like
that,
so
you
can
use
pretty
much
any
anything
that
has
a
crowd
API
and
if
there
isn't
one
you
can
just
try
to
which
is
very
easy
to
plug
in
to
terraform
as
well,
very
easy
to
write
a
provider,
there's
a
full
documentation
and
guides
on
on
our
learning
website,
slash
comm.
They
can
follow
and
and
and
kind
of
go
with
that
process.