►
Description
In this session, DJ will demonstrate how to deploy GitLab in OpenShift using GitLab’s official Docker image and help you get familiar with the Gitlab web interface and CLI tools that will help you achieve your application development goals.
For more background on Getting Started with GitLab on OpenShift, check out this blog post : https://about.gitlab.com/2016/06/28/get-started-with-openshift-origin-3-and-gitlab/
For the latest information on OpenShift 3.2 and available briefings, please visit http://commons.openshift.org or subscribe to the OpenShift Blog (https://blog.openshift.com).
The OpenShift Commons exists to provide a platform for customers, partners, developers and other open source technology initiatives to collaborate, share and accelerate the pace of innovation and adoption of OpenShift globally.
The OpenShift Commons represents a new open collaborative community model designed to facilitate communication and sharing of best practices, feedback and development across the many open source initiatives that integrate with OpenShift. The best way to get involved is to join the conversation today at http://commons.openshift.org
A
A
Good
morning,
everybody
and
welcome
to
another
openshift
Commons
briefing
this
week.
We
are
really
happy
to
have
DJ
Mookie
here,
a
fellow
Canadian
from
Vancouver
who's,
the
build
engineer
over
at
git
lab,
which
I,
don't
think,
isn't
in
Victoria.
It's
not
a
Canadian
company,
it's
based
in
the
States,
but
we're
really
happy
to
have
other
Canadians
with
us
and
he's
going
to
give
us
how
to
code
test
and
deploy
with
git
lab
on
OpenShift
top.
Today.
A
The
way
this
this
briefing
works
is
we
give
him
the
floor
for
20
30
minutes
to
do
a
talk
and
the
demo
and
walk
through
the
tutorial
or
whatever
it
is
that
he's
got
in
store
for
us,
and
you
can
ask
questions
via
the
chat,
we'll
try
and
answer
them.
The
good
ones
will
read
out
and
have
DJ
talk
to
dirt
for
the
sake
of
the
recording
and
then
we'll
open
up
the
mics
as
well,
and
let
people
ask
questions
and
have
a
dialogue
at
the
end.
A
B
Thank
you
so
here
at
git,
lab
we've
been
working
together
with
some
of
you
had
open
shift
on
creating
a
templated
deploygate
lab
in
openshift
for
the
past
couple
months.
Our
target
for
this.
This
initial
template,
which
is
actually
releasing
today
with
our
release
of
Gil
AB
812,
is
the
the
actual
target.
Is
the
OpenShift
all-in-one
VM
as
a
developer
environment?
B
So
a
little
bit
about
me.
My
name
is
DJ
mountainy
I'm,
a
bilge
engineer
at
gate
lab
and
what
build
engineers
they
get
LabVIEW
is
we're
in
charge
of
packaging
and
configuration
for
the
gate
lab
product
so
underneath
there
I
have
a
little
description
of
the
pipeline
of
what
we
do
and
I
think.
That's
particularly
interesting
to
this
talk
as
it'll
make
it
clearer
it'll
make
it
clear
what
it
is
we're
dealing
with
in
what
what
went
into
it
for
the
template.
B
And
then
one
of
the
things
we're
going
to
do
is
distribute
those
Debden
rpms
to
customers
for
install
the
other
thing
we
do.
Is
we
put
them
into
cloud
images
like
Amazon
ami
images
and
put
those
in
the
the
Amazon
store?
We
also
build
from
those
same
Debs
and
rpms
or
specifically
from
the
Deb.
We
built
a
docker
image
and
push
that
to
docker
hub
and
then
from
that
official
docker
image.
We
are
creating
container
deployment
templates
today
we're
talking
about
the
openshift
one.
B
So
that's
that's
kind
of
the
pipeline
of
what
we
do
and,
as
you
can
see
as
we
go
through
we're
talking
about
this
openshift
template.
It's
built
on
top
of
this
docker
image,
which
is
built
on
top
of
our
omnibus,
bundled
Deb,
which
is
built
on
top
of
which
includes
the
source
code
and
the
other
dependencies
just
really
quickly
a
little
bit
about
gate
level.
I'm
not
actually
going
to
talk
that
much
about
gitlab
the
product
other
than
what
you
see
and
I'm
not
actually
going
to
talk
that
much
about
git
lab
the
company.
B
If
you
have
questions
at
the
end
about
either
one
be
sure
to
ask
me
for
sure,
but
my
my
demo
is
actually
because
it's
the
only
one
vm
I'm
running
it
on
my
laptop
and
it's
not
going
to
be
the
fastest
demo
in
the
world,
and
so
it's
going
to
actually
take
up
most
of
the
time.
So
most
of
the
slides
that
I
have
will
just
be
there
for
going
over,
while
we're
waiting
for
some
builds
to
run
so
about
get
lab.
B
It's
an
on-premises
development
collaboration
tool
based
on
git,
still
that's
kind
of
its
its
selling
feature.
We
do
have
free
private
repos
in
the
cloud
on
get
lab
comm,
and
we
also
do
a
service
offering
on
fiat
host,
do
where
you
can
pay
for
your
own
private
instances
that
we
host
and
manage
for
you,
but
primarily
what
we
are
is
this
on-premise
installation
of
source
code
of
a
source,
control
solution
and
collaboration
around
your
first
control
and
your
software
development.
B
So
it's
we're
heavily
based
on
gift
and
then
we
we
add
in
all
the
tools
around
it,
to
help
you
test
and
deploy
and
iterate.
It
comes
in
two
editions:
community,
the
free
Community
Edition
and
the
edge
addition
which
is
paid
licensing.
The
Community
Edition
is
open-source
and
licensed
under
MIT,
and
we
release
every
month
on
the
22nd,
which
is
today.
B
So
a
little
bit
about
the
template
we're
using
today
so
we're
using
the
opal
ship
template
for
gitlab,
or
maybe
it's
the
gitlab
template
for
OpenShift
I'm,
not
sure,
but
it's
essentially
it's
a
OpenShift
template
that
defines
yet
labs
deployment
configs
and
it's
located
here
at
this
URL,
and
it
actually
is
there
right
now,
but
it's
pointing
at
an
older
version
which
includes
more
steps.
But
after
this
meeting
we're
going
I'm
going
to
be
updating
the
template
at
that
location
to
be
what
we're
using
here
today,
what
I
was
waiting
on
is
about
half
hour.
B
Before
this
meeting
we
actually
started
pushing
our
tags,
live
for
8:12,
so
I
couldn't
really
update
and
pointed
at
docker
images
that
didn't
exist
yet
an
easier
way
to
find
the
template
is
to
go
to
docker
hub,
finding
our
gait
lab
official
image
and,
in
the
description,
there's
a
link
to
the
docker
file
and
this
file.
Oh,
this
OpenShift
template
is
a
peer
to
our
docker
file,
so
I'm
actually
going
to
go
there
in
a
minute
here,
but
just
just
a
my
last
slide
before
actually
jumping
into
my
screen
sharing.
B
What
does
this
template?
What
is
this
template
intended
for
and
why
am
I
saying
that
we're
releasing
the
template
today?
Well,
it's
because
it
works
with
them
yet
lab
812
we've
been
working
for
the
last
few
months
on
workaround
issues
we
had
with
the
persistent
volumes
and
our
requirements
for
for
things
running
his
route,
and
so
as
of
today
in
the
eighth
dot
12.
We
have
it
working
out
of
the
box
in
the
all-in-one
dev
VM.
B
So
that's
what
we're
saying
it's
released
today
and
it's
dependent
also
on
the
open
shift:
origin,
1.3,
that'll
release,
which
was
released
a
few
days
ago
or
any
alpha
or
RC
of
that
it
works
fine
on
today,
we'll
be
using
our
c1
in
my
own
one
divyam,
even
though
there
has
been
a
release
of
the
anyone
already
so
I'll
talk
a
little
bit
once
once
we're
into
the
demo,
but
I'll
talk
about.
What's
the
differences
between
what
we're
doing
and
what
we
have
to
do
to
set
it
up,
if
you're
not
using
the
all-in-one
BM.
B
B
B
B
B
B
So
every
fanta
gated,
a
system
user
to
to
the
all-in-one
VM,
has
access
to
the
use
any
UID
when,
when
deployed
containers
in
the
case
of
the
gate
lab
container,
because
we
do
rely
on
omnibus,
which
is
a
chef
solo
server,
and
it
does
expect
to
be
able
to
do
things
like
shown
and
shown
as
a
root
operation,
we
do
have
to
make
sure
that
the
gitlab
user
is
able
to
run
as
any
UID
and
the
only
one
VM.
This
is
already
set
up.
B
Because
of
this
line
here
in
the
open,
stripped
origin
outside
of
the
all-in-one
VM
you'd
have
to
manually
add
a
service
account
SEC
role
to
the
n,
a
UUID
we're
going
to
actually,
even
though
I'm
in
the
all
one
VM
image,
I'm
just
gonna
go
ahead
and
delete
that
line
just
to
make
it
and
what
more,
like
a
regular
install.
You
know
if
you
were
just
installing
this
in
the
all-in-one
image.
B
B
B
So
those
two
changes
I
made
or
the
changes
to
the
security
context
I
made
you
wouldn't
have
to
normally
do
in
the
the
on
image.
What
you
do
have
to
do
right
now.
You
actually
have
to
add
the
template
in
to
OpenShift,
so
we're
going
to
go
ahead
and
add
our
template
so
I've.
This
is
a
check
out
of
that
that
omnibus
repo,
that
I
was
showing
you
and
I'm
going
to
add
the
OpenShift
template.
B
Go
into
the
back
in
the
UI
I'll
go
into
the
increase
font
here
and
we'll
go
into
the
gitlab
project
will
add
the
project
and
you
can
see
we
have
the
gate.
Lab
seee
application
here
so
like
I
was
mentioning
before
we're
targeting
the
all-in-one
VM
for
this
first
release,
and
this
is
kind
of
the
first
release
we're
considering
supported
and
as
a
result,
it's
still
missing
a
bunch
of
links.
One
of
the
things
that's
missing
is
support
for
a
gate
lab
ee
image
at
the
moment.
So
that's
currently
not
here.
B
So
one
of
the
other
things
is,
as
you
can
see,
there's
a
bit
of
config
options
here
in
the
initial
go,
the
initial
release,
but
there's
if
you're
familiar
with
git
lab-
it's
certainly
not
all
of
them.
Nor
nor
close
to
all
of
them.
So
that's
another
thing
we
want
to
enhance
going
forward,
but
this
is
kind
of
the
state
of
it
at
the
moment.
B
Postgres
Redis,
it's
going
to
put
them
up,
I
primed
some
of
the
images
for
for
picket
lab
C
for
when
it
is
pulling
it
in
so
it
won't
take
as
long
as
it
would
normally
normally
it's
pulling
from
from
docker
hub,
so,
depending
on
your
connection
and
depending
on
how
much
load
is
on
dock
or
how
they
can
take
quite
a
bit
of
time.
Yes,
which
is
why
I
kind
of
pre-primed
it
for
jump
into
monitoring
we'd
see
already
some
health
probes
failed,
that'll.
Be
fine,
though
they'll.
B
I'll
just
walk
through
some
of
the
settings
in
the
UI
or
some
of
the
the
objects
in
the
UI
that
we've
added
in
our
template.
So
under
storage,
we're
using
persistent
volume
claims,
so
we
have
one
for
Redis
one
for
Postgres.
We
have
one
for
your
get
lab
data,
so
that's
your
git
repositories,
your
build
artifacts
from
CI
and
your
user
uploads
for
images
and
stuff
like
that
on
comments
and
stuff
like
that,
so
that's
all
store
it
there
and
then
we
have
the
gate
lab
Cee
TC
directory.
B
So
this
is
where
the
config
file
live,
so
that
we
have
a
couple
different
ways.
You
can
configure
this.
So
as
you
can
see,
you
can
configure
a
few
of
the
options
on
the
first
click.
The
install
screen.
Another
place
you
can
configure
is
using
our
file
system
configuration
file
which
is
located
on
this
drive.
B
So
if
you
have
access
to
wherever
this
volume
claim
is,
is
persisted,
you
could
go
in
and
change
the
etc'
get
lab,
get
lab
RB
file
to
to
change
your
configuration
and
just
do
a
read,
apply
or
you
can
go
in
and
of
course,
change.
It
manually
in
your
deployment
using
environment
variables,
so
we
have
this
omnibus,
config
environment
variable
that
includes,
basically,
what's
in
what
you
can
put
in
that
RB
file
at
the
moment.
B
What's
in
here
at
the
moment,
is
actually
just
the
config
necessary
to
move
the
storage
directories
to
those
persistent
volumes
and
make
use
of
the
Redis
and
Postgres
the
only
other
thing
other
than
that
is
this
password
that
we
know
that
ignored
to
put
in
and
your
actual
external
URL.
Those
are
the
only
two
things
separate
from
kind
of
just
moving
directories
around
for
this
template.
B
For
is
images
we
have
the
gate,
lab
seee
image
which
is
pulling
using
our
release
candidate
later
today,
I'll
switch
it
over
to
using
our
actual
release
and
also
a
Redis
image,
which
is
using
the
official
Redis
image
on
github,
and
then
the
Postgres
image
is
actually
coming
from
open
ships
itself.
It's
using
open
shifts,
Postgres
image.
B
And
there's
not
too
much
more
to
show
in
terms
of
objects
created
with
the
template.
We
have
our
Redis
and
our
Postgres
already
up
and
running
the
gitlab
see
the
image
should
have
been
pulled
and
installed,
and
now
what
it's
doing
is
properly
populating
the
database.
If
we're
going
to
logs
this
kind
of
running
our
reconfiguring
configuring,
the
first
time
set
up
subsequent
set,
ups
and
installs
will
be
faster
than
this.
B
B
B
It's
up
and
running
should
be
available
at
this
URL
for
me,
and
it
is
before
we
do
too
much
I'm
going
to
go
ahead
and
scale
it
to
two
just
to
show
that
it
can
be
scaled
to
I'm,
not
going
to
really
wait
for
it,
though
I'm
going
to
go
ahead
and
show
off
you
lab
a
little
bit,
probably
most
some
of
you
at
least
have
already
seen
it.
So
this
will
be
probably
boring
but
we're
doing
a
CI
CD
demo.
So
that's
what
you
get
because
I
didn't
specify
root
password
in
then
click.
B
B
B
Settings
are
located
that
are
stored
in
the
database,
that
you
don't
have
to
change
on
the
file
system
or
in
environment
variables.
So
there's
a
bunch
of
additional
settings
in
here
that
I
won't
really
talk
too
much
about
just
for
fun,
we'll
turn
off,
but
we'll
leave
sign
in
and
sign
up
disabled.
So
no
one
else
can
sign
up
for
now,
not
that
it
matters
who
is
just
on
my
local
machine
bunch
of
different
settings
that
don't
require
because
the
store
in
the
database
don't
require
you
to
change
anything
in
your
file
system.
B
I'm
going
to
make
it
public
for
the
sake
of
the
deploy
step
later
on
we're
going
to
keep
this
public,
you
can
set
it
up
to
to
check
out
using
username
and
password
as
well,
even
in
the
deployment,
but
for
the
demo
we're
going
to
keep
it
simple
and
plus
this
website
is
already
public.
So
it's
going
to
go
ahead
and
import.
The
this
repository
is
actually
like
half
a
almost
half
a
gig
in
size.
B
So
in
retrospect,
when
I
was
doing
said
running
through
this
demo
a
few
times
I
realized,
maybe
it
wasn't
the
best
one
to
show,
because
it's
going
to
take
a
while,
like
everything's
going
to
take
a
while,
unfortunately,
but
that's
what
I
ended
up
doing
so
once
this
import
is
done,
we'll
see
a
few
more
options,
but
we
already
see
some
features
of
get
lab.
We
see
the
activity
we
see,
we
have
an
activity
feed,
we
have
pipelines,
which
is
the
CI,
our
integrated
CI
solution.
B
B
B
You
would
first,
you
would
install
Jenkins,
so
it's
actually,
as
you
guys
all
know,
probably
it's
very
easy
to
install
here
in
openshift,
but
we
also
need
for
gate
lab
integration.
We
need
the
get
lab
plug-in,
which
is
located
here
for
Jenkins,
so
you'd
have
to
make
sure
that
that
plug-in
is
enabled,
and
also
dr.
dentals
plugin
comes
in
pretty
handy
as
well.
So
the
kind
of
the
general
steps
that
we're
not
actually
going
to
do
if
you
would
have
to
do
for
Jenkins,
is
you
need
to
install
Jenkins?
B
Of
course
you
need
to
create
a
new
gate,
lab
user
for
Jenkins
in
get
lab
because
you're
going
to
use
their.
If
you
want
to
use
private
repositories
and
said
the
public
once
you
want
to
use
their
username
and
password
in
the
credentials
plug-in
and
also
their
private
token,
for
integrating
with
the
plugin,
you
would
then
to
get
lab
server,
hostname
and
the
jaqen
users
private
token.
In
the
Jenkins
management
page
in
the
Jenkins
management
page
there's
an
after
you
have
the
get
lab
plug-in,
there's
a
section
for
get
lab.
B
You
would
then,
when
you're
building
your
Jenkins
job,
you
would
configure
the
get
lab
plug-in.
It's
now
a
build
option
in
the
Jenkins
job,
and
you
would
basically
you
point
it
at
your
your
repo
using
the
get
URL
and
you
use
a
credential
plug-in
to
provide
it
with
logging,
if
it's
not
a
public
repo
and
while
you're
setting
up
that
plug-in
it'll,
actually
give
you
a
callback
URL
to
use
in
the
get
lab
web
books
section
of
the
product,
so
we're.
B
B
So
using
git
lab
CI,
so
gillip
CI
what
does
get
loud,
CI
its
its
CI,
that's
tightly
integrated
into
get
lab.
It
does,
in
addition,
so
it's
the
the
API
side
of
is
already
in
that
get
lab
docker
image
that
we
loaded
up.
It
was
under
that
pipelines
in
the
UI
was
under
the
pipeline's
tab
in
your
project.
It
does
require
that
you
register
these
runners
to
execute
CI
scripts.
These
runners
are
what
we're
going
to
load
up
in
openshift
our
runners
and
the
runner.
B
B
B
It's
it's,
it's
really
yeah
it.
It
kind
of
shows.
Definitely
what
I
think
everyone
wants
that
have
happened
with
with
the
github
CI
or
if
we
want
Caleb,
CIN,
OpenShift
that'd
be
really
awesome
and
with
the
kubernetes
support,
we're
we're
going
to
get
it
there.
It
just
today
isn't
today
specifically
isn't
the
day.
Unfortunately,
then,.
B
Jump
back
to
my
web
site
and
we're
going
to
set
up
CI
so
because
I
did
a
clone
of
this
was
a
clone
from
gitlab,
and
this
is
our
actual
website.
It
actually
already
has
a
CI
configuration,
so
we're
just
going
to
click
on
that
link,
and
this
is
our
get
lab.
Ci
yeah
mol
file-
we're
not
going
to
use
all
this
today,
because
we're
going
to
leave
the
build
and
the
deploy
up
to
the
source
to
image
stuff.
So
we're
really
only
going
to
do
the
linker
in
terms
of
CI.
B
Our
CI
animals
include
a
docker
image,
so
this
looks
like
a
big
scary.
Docker
image
I
actually
have
it
open
here.
This
is
docker
for
that
image.
It's
actually
just
a
ruby,
2/3
slim
that
is
installing
a
couple
packages
and
then
cleaning
up
after
them
and
then
setting
the
encoding
environment
in
terms
of
UTF
encoding
and
the
reason
we
have
this
in
an
image
of
course,
is
to
so.
B
So
I'm
going
to
close
that-
and
this
is
our
official
website-
because
that-
and
this
is
my
deploy-
okay,
so
I'm
going
to
head
and
edit
this
file
run
master
right
now,
I'm
just
going
to
drop
out
the
deploy
step
back
on
when
I
drop,
the
build
step
when
I
drop
this
tag,
because
we're
only
going
to
set
up
one
CI
runner.
This
tag
lets
you
determine
which
runners
is
going
to
run
on.
B
B
If
we
over
to
runners,
we
can
see
that
it
tells
us
how
to
set
up
a
new
runner,
but
we
don't
actually
have
any
runners,
apparently
so
we're
going
to
add,
add
some
others,
we're
going
to
add
them
to
open
shifts
so
jump
back
over
to
our
command
line,
and
so
we're
going
to
use
my
hacky
template.
My
hacky
template
is
using
doctor
and
doctor
and,
as
a
result,
I'm
going
to
have
to
throw
from
the
container
in
privilege
mode
and
just
to
get
it
easily
set
up.
B
B
It's
about
that
only
okay,
so
we've
added
the
service
account
to
where
it
needs
to
be
to
run
a
privilege
container.
So,
like
I
said,
like
I,
said
the
the
part
of
the
demo,
where
we
actually
showed
what
we're
releasing
today
and
support
is
done.
We
should
get
lab
now
we're
showing
the
hacky
stuff
that
I,
just
I
didn't
want
to
set
up
Jenkins,
so
I
instead
set
up
a
lab
CI,
so
we're
going
to
create
a
from
new
file
and
I.
Have
it
called
my
CI
template.
B
So
if
we
go
back
into
our
kit
lab
project,
we're
going
to
add
it
to
the
project,
I
now
have
a
get
lab
Runner.
It
needs
access
to
the
gait
lab
instance.
I
we
have
ours
using
that
the
Public
DNS
name
server.
I,
don't
want
to
over
load
that
name
server
with
requests
from
my
machine
so
I'm,
actually
being
that
this
is
deployed
in
the
cluster
I'm,
going
to
go
ahead
and
use.
B
So
it
was
in
a
good
lab
project
service,
cluster,
local
cluster,
oh
and
it's
/div
/div.
If
we
go
over
to
the
miners
page,
we
can
see
it
was
flash
CI
for
communicating
and
then
we
need
a
CI
token
registration
token,
and
that's
this
guy
right
here
and
copy
him
over
and
then
I
have
volume
sizes
for
build
volumes
and
cash
volumes
go
ahead
and
create
that.
B
And
it'll
create
a
new
runner
and
boot
it
up
and
because
I
had
this
was
another
image.
I
had
pre
seeded
I
bring
it
up
pretty
quick,
and
if
we
look
in
it
it
already
registered
the
runner
which
we
can.
If
we
go
back
to
our
runners
page
and
we
refresh,
we
can
see
it's
registered
the
runner
and
it's
already
grabbed
that
first
build
that
we
ran.
So
we
can
go
over
to
our
our
pipelines.
B
We
can
see
our
build,
is
now
running,
we
click
in
it.
We
can
go
into
the
specific
lint
build
and
we
can
see
it's
current
status,
so
this
first
run
is
going
to
be
it's
going
to
take
a
little
while
because
it's
going
to
have
to
pull
it
in
this
registry
image
which
isn't
going
to
actually
be
the
bulk
of
the
time
it's
going
to
spend
both
the
time
doing
the
bundle
install,
but
this
is
going
to
cache
those,
and
so
our
next
builds
are
going
to
be
a
lot
faster
in
terms
of
CI.
B
So
while
that's
running
this
is
this
is
kind
of
so
far
we
kind
of
followed
a
nice
nice
flow,
but
now
things
are
getting
a
little
bit
slow,
we're
gonna,
start,
jumping
all
over
the
place,
so
hopefully,
after
I
finished,
jumping
all
over
the
place
in
the
demo.
I
can
quickly
run
through
the
slides
to
kind
of
put
it
back
in
order
for
you
and
and
I
apologize.
B
And
our
website
is
using
Ruby,
so
we're
going
to
use
the
Ruby
2
3
builder.
That's
already
comes
with
OpenShift
I'm,
just
going
to
call
it
website
at
once
again,
because
this
is
in
the
cluster,
we're
not
going
to
quite
use
the
URL
that
it
gave
us
because
I'm
using
the
public
domain
name
server
provided
by
a
pimp.
B
Forget
lab:
we
need
to
use
the
generic
trigger,
but
the
only
differences
are
these.
Last
two
sections,
this
secret
and
the
last
path
program
go
ahead
and
copy
it.
We're
gonna
go
back
to
to
get
lab,
which
is
still
letting
that
build
we're
going
to
we're
in
the
project
and
we're
going
to
set
up
web
books
for
this
project
we're
going
to
enter
the
URL,
and
we
need
to
basically
replace
this
secret
and
this
last
last
path
here.
So
one
way
I've
found
to
find
it
from
the
UI.
B
B
And
we
see
it
does
have
a
generic
web
book
set
up
here
and
it
had
does
have
a
different
secret.
We're
going
to
go
ahead
and
copy
that
get
in
here
and
we're
only
going
to
trigger
this
on
build
events.
We
don't
have
to
specify
secret
token,
because
it's
already
there's
already
one
included
in
the
URL
and
we're
going
to
disable
SSSs
L
of
verification,
because
I'm
using
a
self-signed
certificate.
B
Installing
the
bundle
inspected
out,
it's
think
that
it's
going
to
spend
some
time
installing
it's
going
to
take
a
while.
The
one
thing
we
do
have
to
change
specifically
for
our
website.
The
deployment
that
got
automatically
set
up
here
isn't
quite
enough
to
get
ours
our
our
particular
website
working.
It's
very
close,
though
the
only
thing
it's
missing
is
some
environment
variables
to
get
it
into
English
utf-8
encoding,
I
just
have
a
copy
of
environment
variables.
Here,
I'll
get
it
going
to
before.
B
A
B
I
can
answer
right
now,
so
the
persistent
storage,
when
you
scale
up
so
it
uses
the
same
persistence
or
so
we
specifically
separated
the
storage
out
so
that
essentially
the
instance
based
config
files
in
terms
of
that's
in
terms
of
what
happens
when
it
first
when
an
instance
first
boots
up
in
configures
its
nginx
and
it's
unicorn
workers
and
the
rest
of
the
things
to
run
the
application.
Those
are
not
persisted,
but
things
like
the
git
repos
were
persisted
and
things
like
the
your
build
artifacts
and
your
uploads
and
stuff
like
that
were
persisted.
B
So
when
you
increase
the
the
instances
it
ends
up,
pointing
at
just
the
same
file:
storage
for
your
repos
and
your
builds
but
uses
other
than
that
it
doesn't
persist.
The
rest
of
the
instances
running
like
kids
or
anything
like
that.
So
it
basically
just
ends
up
bringing
up
multiple
of
the
rails,
application
all
pointing
at
the
same
NFS,
yet
repositories
or,
however,
you
setup
your
persistent
storage.
B
And
so
you
can.
You
can
scale
those
application
servers
up.
We
we
kind
of
do
the
same
thing
except
outside
of
openshift
on
git
lab
comm.
We
have
kind
of
like
20
of
these
get
lab
application,
servers
that
are
running
that
are
just
all
attached
to
the
same
storage
or
or
the
same
storage
provider,
and
then
we
have
our
database
and
our
Redis
separate
as
well.
So
it's
setup
it's
designed
to
work
the
same
way
in
openshift,
even
though,
even
though
right
now
the
target
is
the
all-in-one
VM.
B
B
B
So
in
our
CI
instance,
we're
still
checking
out
and
installing
gems
as
well
luckily
for
the
CI
build
all
these
guys
will
be
cached,
so
our
next
build
won't
have
to
use
them
or
I
won't
have
to
re
build
them.
Unfortunately,
I
didn't
spend
too
much
time
mucking
with
the
sourced
image
stuff,
so
the
V
build
on
this
side
will
have
to
you.
B
B
B
B
A
While
you're
you're
waiting
for
that
I've
been
asked
another
question,
so
one
of
the
folks
on
the
call
is
just
asked
them.
Why
do
you
keep
saying
this
is
targeting
the
all-in-one
as
he
understands
you,
it's
NFS
storage
or
any
other
concurrent
readwrite
file
system
provider
is
available.
What
would
prevent
it
from
running
on
any
origin
or
OpenShift
enterprise
cluster?
On
freman,
four,
on
five
yeah.
B
So
it's
more
just
an
excuse
to
say
why
some
of
the
config
is
missing
so
certainly
like
this
is
this
is
an
actual
install
of
gitlab
using
our
official
image.
So
if
you
do
go
into
the
file
system
and
you
edit,
the
config
or
you
go
in-
and
you
want
to
edit
your
deployment
configure,
you
want
to
add
things
build
on
top
of
our
template.
It's
it's
totally
all
there
and
ready,
and
that's
fine,
it's
more
just
like
explaining.
What's
missing
from
the
template,
so
the
build
itself,
the
way
it
runs.
That's
that's
fine!
B
B
A
Alright,
we'll
get
we'll
get
it
up
there,
so
it's
all.
It's
also
might
be
worth
sharing
what
the
process
was,
that
you
went
through
for
working
with
Red
Hat,
to
get
this
working
and
and
who
you
work
with
and
how
that
was
so
I
I
know:
you've
had
a
lot
of
interactions
with
Red
Hat's,
and
so
you
know
other
commoners.
What
was
your?
What
was
the
process
like
you
get
all
this
yeah.
B
So
essentially,
we
started
with
the
OL
in
one
beam
as
well.
It
seemed
like
that
was
that
was
kind
of
the
initial
desire
was,
let's
see,
working
in
the
all-in-one
VM
and
then
move
on
to
the
other
stuff.
So
we
worked
with
Steve
and
I'm
gonna
mispronounce
his
name.
It's
either
Cori
or
George
I'm,
not
sure.
B
So
we've
been
working
with
them,
Hori
essentially
created
the
initial
template
for
us
and
then
we've
been
modifying
it
and
then
he
great
initial
basic
template
based
on
our
image
and
then
we've
been
modifying
and
getting
him
to
review
it
and
provide
feedback
on
it.
So
it's
been
a
back
and
forth
in
terms
of
that,
not
that's,
basically
how
it's
been
so
far.
B
Don't
go
ahead
and
accept
this
merge,
merge
request,
it
will
merge
into
master
it'll,
kick
off
additional
builds,
and
hopefully,
at
least
maybe
even
the
first
build
will
have
finished
in
OpenShift
we
go
look
at
this
was
the
first
build.
Oh,
it
didn't
even
finish
it.
That's
okay!
Well,
just
what
we'll
do
is
we'll
only
show
the
end
output
of
the
first
build,
and
you
can
just
imagine
the
other
ones.
B
So
we
did
we're
doing
the
deploy
in
this
demo
using
the
source
image,
so
we
created
the
OpenShift
at
we
created
a
ruby
app
for
for
our
website
and
pulled
from
a
github
repository.
We
enabled
the
generic
get
trigger
and
we
we
plugged
that
trigger
in
to
get
left,
webhooks
so
kind
of
like
what's
next
for
the
integration
we
need
Docs,
so
definitely
to
get
the
docs
up
there.
One
thing
also
is
get
ssh
support.
Right
now
you
would
have
saw
me
only
cloning
over
HTTP.
We
need
to
expose
the
SSH
port
externally.
B
I
understand,
there's
a
change
to
the
public
router
that
went
into
two
OpenShift.
That
will
allow
us
to
do
that.
We
just
haven't
made
use
of
it.
Yet
that's
a
funny
way
to
spell
exposed,
but
expose
more
of
the
configure
options.
Specifically
email
is
one
that's
missing
from
the
from
the
one-click
installer
adding
smtp.
We
want
to
add
the
gulabi
images.
B
We
have
documentation
already
on
using
kubernetes
and
setting
up
our
runners
to
use
it
and
setting
up
as
a
privileged
runner,
but
they
don't
currently
work
out
of
like
this
template
itself
doesn't
work
out
of
the
box.
If
you
just
plug
it
into
openshift,
we
need
to
do
some
investigation
on
getting
that
working
would
be
fun
to
do
deployments
from
the
Runner
and
a
clearly-defined
backup
restore
procedure
right
now.
B
I
have
a
question
mark
after
this
one,
because
it's
likely
not
needed
because
often
you're
going
to
have
a
container
registry
already
in
openshift,
but
just
threw
it
in
there
because
I'm
not
sure,
then
maybe
there's
a
use
case
for
it
in
openshift,
with
Gil
AB
or
maybe
there's
not
so
just
to
jump
back
to
our
push
was
successful.
Oh
jump
back
to
overview.
Our
website
is
up.
It's
running,
more
builds
basically
for
each
push.
I
did
I
mean
in
the
merge
itself.
We're
gonna
ignore
those
builds
and
just
look
at
the
websites.
B
A
B
A
Put
up
your
last
QA
slide
and
or
your
laughs
your
final
slide
and
we'll
do
that.
I
am
actually
not
seeing
any
questions.
Jorge
has
joined
us
the
gentleman
from
Red
Hat,
who
gave
you
a
hand
on
some
of
that
I've
unmuted
him
and
to
see
if
he
has
any
other
things
to
add
in,
but
if
you
could
also
tell
people
how
to
get
a
hold
of
you
DJ
or
where
the
right
place
is
to
reach
out
at
gitlab.
If
people
want
more
information,
that
would
be
a
great
thing.
Sir
yeah.
B
So
my
email
is
DJ
at
gate,
lab
calm
and
the
the
repository
that
I
monitor
is
the
omnibus
repository.
So
if
you
have
any
issues
with
the
template,
we
do
have
a
issue
tracker
on
that
omnibus
repository,
which
was
a
link
to
at
the
beginning
of
the
slides
and
link
to
from
our
docker
image
on
docker
hub.
So
any
issues
you
have
feel
free
to
put
there.
A
The
recording
of
this
and
I
will
make
DJ
send
me
his
slides
as
well,
so
that
you
don't
have
to
type
in
everything
that
he
was
typing
in
and
can
steal
from
that,
and
that
should
be
up
in
a
day
or
two
on
the
blog
openshift
comm
site
and
on
the
OpenShift
youtube
channel,
and
please
also,
if
you
have
questions
post
them
on
the
open
ship,
Commons
mailing
list
and
we'll
get
DJ
to
answer
them
there.
So
everybody
can
see
them
and
we
will
be
back
again
next
week
with
another
open
ship.
A
Commons
briefing
and
again
I
will
do
a
slight
pitch
for
if
you're
up
for
doing
the
Commons
briefings
in
person,
November
7th
will
all
be
gathering
in
Seattle
at
the
open
ship,
Commons
gathering
you
can
register
today.
It
is
a
prequel
to
pucon
and
cloud
datacom
co-located
at
the
Seattle
Sheraton
and
the
address
to
find
that
is
Commons
dot,
OpenShift
org,
slash
gatherings
and
like
what
you've
seen
in
the
briefings.
A
You'll
love,
seeing
everybody
in
person
and
getting
to
drink
a
beer
or
have
lunch
with
them
and
pepper
them
with
questions
too
so
I'm
hoping
DJ
and
some
of
the
gitlab
folks
will
be
there
and
get
your
questions
answered
there
in
person.
So
once
again,
thank
you
for
joining
us
at
this
open
ship,
Commons
briefing
and
thanks
DJ
for
doing
such
an
awesome
job,
showing
off
open
shift
and
working
with
us
on
getting
this
integration
up
and
running.