►
Description
GitHub Actions gives us the power to use our repositories to speed up the delivery of our software and applications, all from one central point of truth. This workshop will provide you with hands-on experiences with GitHub Actions to leverage GitHub package registry and safely deploy applications to the cloud.
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
A
A
A
A
All
right,
sorry
about
that,
we're
gonna
we're
gonna
start
off
by
setting
some
context
about
what
continuous
delivery
is
and
then
we
will
jump
in
and
do
the
first
workshop.
That's
going
to
be
a
workshop
to
store
a
docket
docker
registry
docker
image
in
in
the
get
a
package
registry
and
we're
going
to
be
using
that
in
a
subsequent
workshop,
when
we
are
building
and
storing
images
in
docker
in
github
package
registry,
and
then
we
will
go
and
do
the
second
workshop,
which
is
going
to
be
continuous
delivery
to
aws.
A
A
A
We
already
seen
that
continuous
delivery
is
not
continuous
deployment,
because
continuous
deployment
has
to
think
about
things
such
as
feature
toggles.
We're
going
to
do
a
b
testing
we're
going
to
move
certain
aspects
such
as
change,
management,
security
and
compliance
left.
In
order
to
do
continuous
deployment,
continuous
delivery
doesn't
have
to
be
a
fully
automated
process.
We
can
have
gating
process
in
continuous
delivery
where
code
is
promoted
from
the
feature
branch
to
production,
and
it's
not
certainly
restricted
to
deploying
to
cloud.
A
A
At
that
point,
there
is
going
to
be
a
manual
process
to
trigger
the
continuous
delivery
pipelines
and
continuous
delivery
workflows,
and
the
continuous
value
workflows
are
going
to
deploy
the
code
to
a
staging
environment
and
they
are
going
to
do
acceptance
testers
and
it
could
be
a
manual
or
automated
acceptance
test.
Once
those
tests
pass,
then
the
code
is
merged
to
the
master
and
the
code
is
deployed
to
production.
A
Now,
if
you're
doing
a
continuous
deployment,
you
could
as
well
do
the
deployment
production
while
you're
on
the
feature
branch
and
then
you
can
use
feature
toggles
to
do
some,
a
b
testing
and
then
there's
a
test
pass
and
you're
happy
that
the
code
is
going
to
work
fine
in
production.
You
can
then
merge
the
code
to
master.
A
With
that
ground
setting,
let's
get
started
on
the
actual
work
workshop
itself
for
the
workshop,
the
prerequisites
are
you
need
a
learning.
Github
account,
you
need
learning
lab,
you
need
to
have
access
to
actions
and
make
sure
that
you
have
access
to
github
package
registry
you
signed
up
for
that.
We
would
also
be
using
a
personal
access
token,
so
we
should
create
that
on
the
aws
side,
you
need
to
create
an
a
you
need,
an
aws
account.
A
And
in
this
workshop,
what
we're
going
to
do
is
we're
going
to
pick
up
from
where
we
left
off
in
the
continuous
integration
and
we're
going
to
modify
the
continuous
integration.
We're
not
going
to
use
the
same
workflow,
but
we're
going
to
take
a
continuous
integration.
Workflow
modify
it
to
create
a
docker
image
and
we're
going
to
store
the
docker
image
to
the
get
a
package
registry.
A
A
A
A
A
A
A
A
A
A
In
this
pull
request,
we
are
going
to
create
a
docker
image
and
docker
images
gives
you
the
consistency
and
immutability
that
guarantees
that
whatever
is
being
tested
is
the
stuff.
That's
gonna
run
in
production.
So
you
can
have
the
confidence
that
your
acceptance
test
is
gonna
test
the
same
code,
that's
to
run
in
production.
A
A
A
A
Another
thing
to
notice
is
it's
using
a
community
action
which
is
built
by
matt
davis
who's
at
github,
but
you
can
also
pick
and
choose
an
action.
That's
going
to
work
for
you
for
this
workshop,
we're
going
to
use
this
action,
but
for
your
work
you
may
pick
and
choose
any
action
that
suits
your
need.
A
A
A
A
A
A
A
A
All
right,
so
I'm
going
to
switch
back
to
here,
so
you
can
actually
look
at
the
links
and
I'm
gonna
man
the
slack
channels.
While
we
watch
for
any
questions,
but
please
go
ahead
and
start
the
workshop
and
then
we
will
stop
at
the
at
this
point
where
your
jobs
are
gonna,
run
and
then
create
the
docker
image.
C
C
C
C
A
A
I
already
have
the
docker
image
running
several
docker
images
there,
but
I
can.
I
can
do
a
docker
login
and
give
the
credentials
my
private
personal
access
token
and
my
username,
and
then
I
can
log
into
my
get
a
packaged
registry
and
I
should
be
able
to
download
this
to
a
docker
poll
and
get
this
image,
so
I
can
copy
it
here.
A
And
then
the
next
step,
what
we're
going
to
do
is
we're
going
to
take
it
another
step
forward
and
then
deploy
the
application
to
aws
in
order
to
deploy
to
aws.
We
are
actually
going
to
deploy
it
as
a
serverless
app.
So
we're
not
going
to
use
the
container
that
we
created
and
we'll
cover
that
later
in
the
workshop.
A
But
what
we're
going
to
do
is
we're
going
to
create
a
workflow
and
then
when
we
want
a
manual
trigger
to
kick
in
the
cd
workflow.
So
what
we're
going
to
do
is
we're
going
to
create
a
trigger
based
on
a
label
and
we're
going
to
create
a
a
label
called
stage
and
we're
going
to
deploy
when
the
label
cost
stages
applied
to
the
workflow.
A
A
A
A
A
A
A
A
A
C
C
C
C
C
C
C
C
C
A
A
A
So
next,
what
we're
going
to
do
is
I'm
sorry
we're
going
to
go
here
and
then
we're
going
to
create
an
aws
account
if
you
are,
if
you
don't
have
one
and
then
we're
going
to
create
an
iam
user.
So
if
you
go
to
adobe,
ps
and
login,
one
of
the
services
would
be
iam.
You
can
go
there
and
create
an
im
user.
A
A
A
A
So
here
what
we're
going
to
do
is
we're
going
to
we
create
we're
going
to
go
to
aws
and
create
an
s3
bucket
and
once
the
s3
bucket
is
created.
Please
make
note
of
the
region
that
you're
using
and
that
has
to
be
consistent
across
all
the
files
and
then
we're
going
to
create
a
an
aws
config
file
and
we're
going
to
update
the
config
file
with
those
entries.
A
A
A
A
A
So
what
we
have
done
in
this
is,
we
have
right
now
created
a
staging
workflow
in
the
staging
workflow.
What
we
have
done
is
we
have
created
a
trigger,
that's
going
to
be
based
on
a
label
labeled
event
and
there's
going
to
be
a
job.
That's
going
to
get
triggered
conditionally,
based
on
a
value
of
the
label,
we're
looking
for
a
label
with
a
value
of
stage
and
that
job
is
going
to
build
the
application
and
then
there's
another
job
that
runs
subsequently
and
that
needs
the
previous
job
to
complete.
A
So
it's
going
to
run
sequentially
and
that
job
is
going
to
check
out
the
artifacts
from
the
previous
job
and
then
deploy
it
to
aws.
In
order
to
deploy
to
aws,
we
have
to
create
a
secret
key.
We
have
to
create
an
iam
account.
We
have
to
create
a
secret
aws
access,
key
and
secret
key,
and
we
have
to
store
that
as
secrets
with
these
values.
A
A
A
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
A
A
A
A
A
A
The
workflow
to
run
so
now,
you're
gonna,
see
the
bot
come
back
and
say
you're
done,
and
you
should.
If
you
go
to
the
actions
tab,
you
will
see
that
there
is
going
to
be
a
the
merge
even
triggered
the
workflow
to
run
and
the
workflow
is
going
to
do
the
build
step.
And
then
it's
going
to
do
the
deploy
step.
A
So
what
we've
done
in
this
course
is.
We
have
created
a
staging
workflow,
the
staging
workflow
we
want
to
trigger,
based
on
a
particular
label
when
the
feature
branch
is
ready
to
be
merged,
that
triggers
the
workflow
to
run,
do
a
build,
run
some
tests
and
then
deploy
the
code
to
a
stage
environment
which
is
in
our
case
on
aws,
and
then
once
the
acceptance
criteria
is
passed
and
and
testing
has
been
completed,
then
we
can
merge
the
code
to
master.
A
There
is
a
mask.
We
created
a
master
workflow
which
is
going
to
get
triggered
when
there's
a
push
to
master
and
the
master
workflow
is
gonna,
I'm
sorry,
the
the
production
workflow
is
gonna
trigger
a
build
and
it's
gonna
deploy
the
code
to
aws,
and
it's
also
gonna
create
a
darker
image
and
then
store
that
in
the
github
packages.
C
C
C
C
C
C
C
C
C
C
A
A
A
A
A
A
A
Something
similar
to
ecs
it's
a
service,
it's
a
it's
a
deployment
methodology
where
you're
running
a
a
container
based
application
in
an
app
in
an
app,
and
in
this
case
what
we
are
doing
is
we
are
using
a
free
tier.
So
when
you,
when
you
sign
up
for
azure,
you
can
run
this
and
keep
this
sku
as
f1,
and
it's
going
to
be
available
for
you.
A
You
can
also
have
another
job
here,
which
is
to
destroy
the
azure
resources,
so
you
don't
have
to
worry
about
like
tearing
it
down.
You
just
apply
this
destroy
environment
and
it's
gonna
tear
down
your
your
azure
infrastructure
for
you
when
we
look
at
the
code
here,
let's
look
at
the
pull
request
and
we
can
walk
through
a
few
things
that
needs
to
be
set
up.
A
So
what
we
need
for
this
deployment
is,
you
need
an
azure
account
and
then,
once
you
have
an
azure
account
and
you
need
azure
cli
and
with
the
cli
you,
you
would
create
a
service
principle.
This
is
similar
to
an
iam
user,
so
you're,
creating
that
with
the
role
of
contributor
and
you
you're,
giving
it
only
a
scope
to
that
particular
subscription
that
you
have
you
can
also.
A
You
can
do
a
more
fine-grained
control
here
and
once
you
create
a
your
service
principle,
you
provide
those
values
as
secrets
in
your
in
your
repo,
so
you're,
giving
the
azure
subscription
id
and
also
you're,
giving
the
azure
credentials.
The
azure
credentials
is
basically
the
json
that
you
get
when
you
run
this
command,
and
you
store
that
once
you
store
have
that
information,
then
all
you
need
to
do
is
then
create
that.
A
A
The
job
itself
has
an
azure
login.
This
is
where
it's
going
to
use
your
azure
credentials,
you're
going
to
log
in
to
your
azure
environment.
Basically,
what's
happening
is
from
your
runner,
you're
logging
into
the
azure
environment
and
then
you're,
creating
a
resource
group
here
and
once
you've
created
a
resource
group
you're,
creating
an
app
service
plan
with
the
particular
skew.
A
A
A
You
build
the
image
and
store
the
image
in
in
get
a
package
registry
and
then
you're
deploying
it.
So
in
this
case,
deployment
is
dependent
on
the
container
being
available
in
the
package
industry,
and
this
is
gonna
with
zero
downtime,
deploy
the
container
and
then
start
the
application.
With
the
new
container
there.
A
A
A
A
A
In
this
case,
I
have
added
a
code
in
this
particular
branch.
So
the
way
you
do
it
is
you
you
have
in
on
aws,
you
create
a
ecs
cluster.
You
create
a
ecs
task
definition.
The
task
definition
could
be
just
a
template.
So
if
you
look
at
this
task
definition,
I
have
a
task
definition
here
and
I
have
specified
an
image
here,
but
I
also
in
order
for
this
image
to
be
pulled
from
github
package
registry.
A
I
need
to
create
a
repository
credential
secret
in
aws,
and
then
I
have
to
store
the
your
personal
access
token,
your
repository
url
and
and
your
username
there.
Once
you
create
a
repository
credentials,
you
take
the
arn
and
right
here
when
you
configure
it,
you
can
trigger
the
you
provide
that
arn
here,
I'm
using
far
gate
here
so
that
I
I'm
not
going
to
be.
A
I'm
in
the
right
place,
yeah
right
here,
so
in
this
case,
what
we're
going
to
do
is
I'm
using
the
aws
action
to
create
a
task
definition.
So
what
this
is
going
to
do
is
it's
going
to
use
the
task
definition
file
that
we
have
and
it's
going
to
replace
it
with
the
container
image
that
we
have,
that
the
the
workflow
is
going
to
build
once
the
task
definition
is
created,
it's
going
to
be
available
for
the
subsequent
tasks
in
the
job.
A
A
A
So
that's
pretty
straightforward.
So
in
order
to
deploy
create
an
action
to
deploy
the
container
to
aws,
all
you
need
is
to
use
the
aws
provided
actions.
That's
there
in
the
marketplace,
create
an
ecs
service,
create
your
ecs
cluster
and
copy
that
task
definition
file
and
put
it
here.
So
that's
pretty
much
all
you
need
to
do
and
then
you
should
be
able
to
deploy
it.
A
A
So
we've
seen
that
actions
is
really
fantastic
for
doing
a
continuous
delivery
because,
first
of
all,
you're
not
managing
and
maintaining
you're,
not
responsible
for
the
care
and
feeding
of
your
cacd
infrastructure
you're.
If
you're
using
get
a
provider
runners,
then
there's
a
low
threshold
and
low
need
for
for
any
kind
of
operational
overload.
A
A
A
A
And
I'll
leave
you
with
the
with
these
links.
I
hope
you
enjoyed
the
learning
lab
workshops
and
I
hope
you
will
continue
practicing
and
experimenting
with
actions,
and
at
this
point
I
will
look
at
the
slack
channel
and
see
if
there
are
any
questions,
but
we
can
also
hand
and
answer
some
questions
that
were
raised
previously.
A
A
A
B
Yeah,
the
only
thing
I
wanted
to
add
was
just
in
a
general
sense.
You
should
treat
it
like
you'd
have
said
as
a
third-party
package,
so
if
it
was
nuget
or
npm
or
anything
else,
that's
open
source
that
you're
consuming.
You
would
want
to
do
whatever
security
reviews
you
need
for
that.
The
great
thing
about
actions
on
the
marketplace,
however,
is
they
are
all
open
source
by
requirement.
A
B
That's
kind
of
what
I
just
answered
with
the
yes
yeah,
sorry
yeah,
specifically
on
that
part
of
the
concern
was
like.
Does
the
action
have
a
security
review
done
before
it's
posted
on
the
marketplace
and
the
answer
to
that
is
no
similar
to
potential
malicious
code
being
in
any
open
source
repository?
A
All
right
thanks,
thomas!
The
next
question
is
other
slides
going
to
be
available.
What
about
the
recording?
Yes,
the
slide!
The
slide
is
there
in
the
repo
and
then
this
this
presentation
is
also
recorded,
so
this
should
be
available
to
you
and
the
recording
would
be
emailed
sometime
after
the
session
next
question.
I'm
looking
for
deploying
to
azure.
Would
this
workshop
cover
that
too?
So
we
did
talk
a
little
bit
about
deploying
to
azure.
I
hope
that
was
helpful.
A
A
A
Time
check,
okay,
all
right!
We
have
we
are
about
on
time
now.
So
maybe
I'll
speed
up
really
fast,
there's
a
labeled
section
in
the
workflow.
Does
that
mean
every
period
needs
a
label
attached?
What
happens
if
no
labels
are
attached?
Basically,
no
every
pr
doesn't
have
every
pr
doesn't
need
a
label
attached.
What
it
means
is
like
when
you,
when
you
want
to
deploy
to
a
particular
pr
to
a
staging
environment.
That's
when
you
apply
a
label.
A
A
All
right
last
question
is
very
important,
so
what
aws
resources
should
be?
Do
we
need
to
remove
in
order
to
avoid
billing,
so
definitely
the
s3
bucket
and
then
the
lambda
function
needs
to
be
removed?
There
is
also
yeah.
You
also
created
the
cloud
formation
stack
by
the
virtue
of
setting
up
the
environment,
so
you
could
also
delete
that
stack
and
you
should
remove
it
and
we
could
definitely
thomas
somebody.