►
From YouTube: GitLab Speed Run: Add GKE cluster and run Auto DevOps
Description
Daniel (Orchestration PM) runs through adding a GKE cluster and running an Auto DevOps pipeline.
A
Hi,
this
is
Daniel
guru,
so
I'm
a
product
manager
on
the
orchestration
group
at
the
configure
stage
here
at
gate
lab
and
today,
I'll
be
doing
a
speedrun
of
two
features,
mainly
I'm,
going
to
add
a
proven
ettus
cluster
using
gke
to
my
project
and
then
I'll
have
it
run
through
Auto
dev
ops.
So
let's
go
ahead
and
get
started
so
here
I
have
a
sample
project.
A
A
A
Okay,
now
that
helm
is
installed
in
the
cluster,
I
will
go
ahead
and
install
ingress
and
I
will
wait
for
the
ingress
IP
address
to
come
back,
so
I
can
go
ahead
and
use
it
on
my
base
domain
that
it
will
basically
allow
me
to
route
traffic
to
my
app
ok.
Now
that
I
have
the
ingress
endpoint
I
am
going
to
go
ahead.
I
see
that
it
lab
automatically
populates
the
suggestion
using
nib.
That
I
owe
so
I'm
going
to
use
that
just
as
well.
A
You
could
use
custom
domain
that
you
own
and
just
have
to
enter
a
DNS
record.
Let's
make
it
easy
I'm
just
going
to
go
ahead
and
use
this
ok.
So
the
next
thing
that
I
want
to
do
is
run
Auto,
dev
ops,
so
this
project
happens
to
be
part
of
a
group
that
may
not
have
auto
dev
ops,
enabled
at
the
group
level,
so
I'm
going
to
go
to
settings,
see
ICD,
auto
dev
ops
and
make
sure
that
I
enable
it
here
once
I
enable
auto
dev
ops.
A
Okay,
so
I
see
that
my
pipeline
has
finished
and
I
see
a
couple
of
failures
here
that
had
caused
the
pipeline
to
failed.
So
let's
take
a
look
at
that
right
off
the
bat
I
see
that
the
test
job
failed
and
due
to
a
script
failure,
and
so
let's
go
ahead
and
take
a
look
at
that
and
let's
figure
out
how
we
can
solve
that.
So
this
lab
the
selected
build
pack
for
Python
does
not
support
the
test
of
feature.
So
that
means
a
couple
of
things.
A
We
can
either
add
something
to
manually
test
it
or
for
the
sake
of
testing.
We
could
simply
disable
the
test
feature.
So
let's
go
ahead
and
do
that
so
Auto
DevOps
allows
me
to
control
what
jobs
run
as
part
of
my
pipeline
through
the
use
of
environment
variables.
So
if
I
go
to
my
project
settings
and
then
I
move
on
to
see,
ICD
I
can
expand
the
variables
section
and
I'm
going
to
add
a
variable
here,
the
specific
one
for
disabling
test.
A
This
test
underscore
disabled,
so
I'm
going
to
get
that
a
value
of
one
or
said
it's
true
and
then
save
the
variables
so
once
I
do
that.
What
will
happen
is
that
the
test
job
will
no
longer
run
the
test.
Job
in
Auto
DevOps
is
not
allowed
to
fail
and
since
that
bill
pack
doesn't
support
the
test
feature.
The
pipeline
will
stop
there.
So
now
that
I've
disabled
tests,
let's
go
ahead
and
run
an
auto
devops
pipeline
once
more
and
let's
take
a
look
at
the
outcome.
A
Okay,
so
I
see
that
my
pipeline
has
finished
and
as
expected,
it's
in
a
blocked
State,
because
I
have
a
manual
trigger
into
production.
So
as
soon
as
I'm
ready,
I
can
just
execute
the
production
job.
But
first,
let's
go
ahead
and
check
out
the
staging
environment,
so
I'm
gonna
go
to
operations
and
environments
and
I
should
see
here
the
deployment
to
the
staging
environment,
which
I
do
that's
great.
So
let's
go
ahead
and
open
the
live
environment.
Verify
that
everything
is
to
our
satisfaction.
I,
don't
have
an
HTTPS
certificate
here.
A
A
Okay,
so
I
see
that
my
pipeline
has
finally
finished
I,
see
that
my
rollout
to
production
was
successful.
I
see
the
rollout
100%
I,
see
that
my
browser
performance
testing
was
also
successful,
and
so
with
that
the
pipeline
has
finished
now.
I
did
finish
with
warnings
here,
because
my
dependency
scanning
job
failed,
but
that
it's
a
job
that
is
allowed
to
fail,
and
that
has
basically
to
do
with
the
build
pack
that
I'm
using.
So
let's
head
over
to
operations
and
environments.
A
Now
we
should
see
the
deployments
for
both
staging
and
production
on
the
board
and
I
see
that
that's
the
case.
I
see
each
instance.
Has
one
pot
running
so
let's
go
ahead
and
take
a
look
at
the
live
production,
environment
and
okay.
I
see
that
it
looks
great
so,
as
you
could
see
in
this
speedrun,
while
not
every
build
pack
supports
every
feature.
Other
DevOps
is
very
easily
customizable
to
support
the
stages
that
you
may
or
may
not
want
you,
and
so
I
hope
that
this
has
been
helpful
for
you.