►
Description
Note: https://www.youtube.com/watch?v=kTNfi5z6Uvk has the most recent workshop from Feb 2021 :)
Agenda with timestamps:
* Workshop intro 6:04
* Let's get started: CI/CD Introduction 9:30
* First steps with CI/CD 11:16
* Build Go Project 16:39
* Unit Tests 23:50
* CI pipeline stages: Deploy & run 27:29
* CI/CD Variables & Templates 29:53
* Monitoring 32:08
* Container Images and Packages 33:45
* Code Coverage with Go 35:09
* Project Badges 35:46
* Round-up 36:32
We hosted a virtual meetup featuring Michael Friedrich, Developer Evangelist at GitLab, on 2020/03/24.
Title: Make your life as a developer easier with GitLab CI and the Web IDE
Description: Ever wanted to get CI quickly up and running? We will be exploring the capabilities of GitLab CI from simple scripts to building code and running unit tests as we code inside the Web IDE. Everything which makes your journey with CI a breeze in group practice.
If you’re interested in learning how to build a CI pipeline in GitLab this talk is for you. If you’re a more experienced GitLab user, we’d still love for you to attend and lend your expertise to our breakout groups.
Slides: https://docs.google.com/presentation/d/1scYkmV4Xdfj-8iwwpEiLCe0vBfpAdrL5pyA2w1Fgnf0/edit?usp=sharing
Repository: https://gitlab.com/dnsmichi/meetup-2020-cw13
---
We would love for you to contribute to our virtual meetups.
You can find upcoming virtual meetups and resources to plan your own here: https://about.gitlab.com/community/virtual-meetups/
We want to hear your ideas for topics. Remote? Security? Contributing to OS? Email suggestions to evangelists@gitlab.com.
We also love when community members present at GitLab meetups. Send an email to evangelists@gitlab.com if you’re interested in being the featured speaker at a future remote meetup.
A
My
name
is
John
Coghlan
I'm
on
the
community
relations
team
at
gitlab,
thanks
to
everyone,
who's
joining
us
today,
big
thanks
to
those
folks
who
are
brave
enough
to
turn
on
their
cameras
and
let
us
all
see
their
you
know
smiling
faces
I.
You
know
I
really
appreciate
that
this
is
a
kind
of
new
experiment
for
us
doing
remote,
meetups
and
so
we're
just
trying
to
figure
out.
You
know
what
works,
what
doesn't
work,
and
so
thank
you
for
kind
of
being
part
of
this
experiment
and
I.
A
Think
many
of
you
probably
understand
the
reason
why
you
know
we're
starting
to
do
our
meetups
remotely,
but,
given
you
know,
the
concerns
around
Kovan
19
we've
decided
to
put
our
in-person
meetups
on
hold
globally
until
the
kind
of
pandemic
slows
down,
but
we
really
want
to
keep
our
community
connected.
I.
A
Think
one
of
the
things
that
you
know
many
folks
are
realizing
right
now,
as
a
result
of
the
crisis
is
just
the
importance
of
community
in
their
lives
and
the
importance
of
human
connection,
and
so
you
know
I'm
excited
to
be
here
today
and
to
see
many
of
you
I
know.
I'm
met
many
of
you
before,
and
some
of
the
folks
here
are
kind
of
new,
but
I'm
hoping
we'll
become
friends
as
we
go
through.
A
A
Some
other
cool
things
about
get
lab
were
the
largest
all
remote
company
in
the
world.
So
if,
if
any
of
you
were
kind
of
working
remotely,
maybe
for
the
first
time
and
kind
of
rushed
into
that,
we
have
a
lot
of
great
information
available
to
help
to
help
folks
who
or
who
are
learning
to
work
remotely
or
starting
to
work
remotely
more
frequently,
including
an
e-book
that
just
came
out.
A
So
if
you're
curious
about
learning
more
about
gitlab,
not
just
the
product,
but
also
how
we,
you
know,
operate
as
a
team
and
what
our
values
are
and
kind
of,
where
we're
going
as
a
company,
that's
all
available
in
our
handbook,
so
I
encourage
you
to
check
that
out
after
today,
yeah.
So
as
I
as
I
mentioned.
A
You
know,
plan
remote
meetups
for
their
communities,
because
I
realized
that
you
know
because
of
time
zones
and
language
barriers
and
things,
even
though
you
know
remote,
meetups
offer
lots
of
kind
of
scale
and
opportunity,
they're
still
not
totally
inclusive,
and
we
want
to
make
sure
that
everyone
in
our
community
can
still
participate
and
get
leb
events
and
so
we'll
be
sharing
those
best
practices.
Out
with
you
know
our
meetup
organizers
and,
and
hopefully
they'll
be
able
to
organize
a
remote
meetups
for
their
communities
in
the
in
the
very
near
future.
A
We're
not
have
any
specific
topics,
but
we
encourage
folks
to
share
where
they're
from
you
know
what
they
do,
what
they
like
to
do
in
their
free
time
as
a
way
to
get
the
conversation
started,
and
you
know
some
other
topics
that
might
be
of
interest
could
be
talking
about
Michael's
presentation,
talking
about
remote
work
or
just
kind
of
talking
about
how
you're
managing
the
situation
right
now.
I
know.
A
A
lot
of
us
have
been
throwing
situations
that
are
kind
of
outside
of
our
norm,
and
you
know
we
encourage
folks
to
talk
about
whatever
is
on
top
of
mine,
and
so,
if
that's
top
of
mind
for
you,
don't
feel
like
it's
taboo
and
most
of
all
just
have
fun
and
connect
with
other
folks.
You
know,
as
I
mentioned.
This
is
our
first
time
doing
this
so
expect
some
bugs,
like
my
internet,
crapping
out
at
the
top
of
the
hour
and
me
disappearing.
A
B
B
A
Be
the
perfect
time
for
you
Michael,
though
jump
it
yeah
just
one
last
thing
you
know
we
love
for
folks
to
participate
in
the
future.
We're
gonna
be
doing
these
more
frequently,
so,
if
you're
interested
in
speaking,
organizing
and
meetup
in
your
local
community,
whether
that's
remote
or
sometime
in
the
future
in
person
or
just
have
ideas
for
topics
you
want
to
hear
about,
definitely
reach
out
to
evangelists,
get
calm
and
I
guess
if
this
one
I'll
throw
it
over
to
Michael
before
my
internet
drops
again,
if.
B
Thinks
I'm
hoping
good
light
and
it
still
works
I'm.
Just
one
thing:
I'm
sharing
the
presentation
with
you
in
the
chat,
so
everyone
can
just
like
watch
my
screen
and
you
can
also
like
copy/paste
from
us
from
the
slides.
The
thing
is
before
I
get
started
with
everything.
This
is
some
sort
of
hands-on.
So
if
you
want
to
follow
my
exercises
as
I
continue.
B
Hi
and
with
that
I
just
want
to
like
give
you
a
little
overview
of
what
I
think
could
make
your
life
easier
with
collapse,
the
ICD
and
the
web
IDE,
which
is
something
I
prepared
for
today.
Let's
see
if
this
works,
yes,
it
works.
I
just
want
to
take
the
opportunity
for
the
next
like
25
to
30
minutes.
We
start
with
a
short
introduction
of
myself.
Some
see
ICD
overview.
B
The
first
steps
with
the
web
IDE
to
get
the
CI
configuration
started,
then
building
some
code
or
actually
analyzing
some
code
and
figuring
out
whether
it's
built
or
how
does
it
work
then
back
to
unit
tests
and
exploring
the
stages
in
the
pipeline
and,
at
last,
a
short
conclusion
with
that
short
introduction
of
myself.
So
my
name
is
Marty
Friedrich,
which
is
totally
funny
toes
to
pronounce
in
English
I'm
a
developer
evangelist
at
github,
it's
my
fourth
week
so
I'm
pretty
new
to
everyone
to
the
culture
to
the
company
to
the
community.
B
I
think
I'm
used
using
gitlab
for
around
about
three
to
four
years
now
explain:
production
before
training
at
lab
I
was
doing
in
the
backend
development
on
one
Trent
project
for
the
past
11
years,
and
before
that
I
was
working
at
the
University
of
Vienna
in
the
domain
administration
department,
which
is
kind
of
my
nickname.
It's
innocent,
Mickey
Mickey,
it's
the
lovely
version
of
Marty
and
German.
B
So
that's
me
on
Twitter
and
get
left
at
home
key
table
wherever
you
want
to
stalk
me
I'm,
a
true
believer,
so
everything
I
do
I'm,
trying
to
make
it
open
source
to
communicate,
to
share
to
work
on
the
greatest
success
for
everyone
and
when
I'm
not
doing
anything
anything
open.
Suse
I
am
building
Lego
models,
so
I
really
left
you
like
shut
down
my
head
and
focus
on
the
things
to
build
and
to
see
them.
B
It's
it's
really
funny
and
and
entertaining
for
me
yeah,
and
you
can
also
like
follow
me
on
Twitter
and
get
a
lot
of
Lego
in
your
field
for
the
technically
evangelism
team.
I've
added
a
link
over
here.
So
you
can
just
check
out
what
what
else
is
going
on
what
we
are
planning
for
this
year
and
if
you
want
to
like
try
my
adventure
or
follow
my
adventure.
B
What
I'm
doing
I've
started
blogging
about
in
what
I'm
doing
at
github,
which
is
kind
of
a
lot
in
the
past
three
weeks
already,
but
with
that
I
want
to
just
get
things
started:
yeah
we're
not
actually
building
Lego
is
just
like
a
model
at
home,
but
sort
of
look
into.
What's
coming
so
I,
don't
know
if
you're
familiar
with
continuous
integration
and
continuous
deployment
or
continuously
live
early.
Sometimes,
though
I've
just
stolen
it,
the
graphics,
from
documentation,
basically
as
a
developer,
when
I
start
working
on
a
new
feature,
create
a
git
branch.
B
That's
basically
the
continuous
integration
part
and
for
the
deployment
at
a
later
point.
Somehow
it
gets
automated
and
put
in
production.
You
know
before
it
was
in
staging
and
being
tested.
Some
environments
require
to
halt
and
stop
so
it's
not
really
automated
and
when
someone
needs
to
like
schedule,
a
maintenance
window
and
click
Play.
My
mission
is
to
like
remove
the
play
button
and
make
everyone
deploy
into
production
and
that's
the
general
idea
behind
it
to
get
started
a
little
easier.
B
Was
things
like
how
to
black
work
with
that
so
before
we
get
actually
started
with
trying
some
things
up
things
out,
you
can
actually
like
go
to
that
repository.
You
can
fork
it
and
you
can
do
the
exercises
as
I
go
along.
You
can
also
just
watch
me
doing
that
and
do
it
later
on,
so
the
slides
are
self-contained
and
the
source
code
is
also
over
there.
I
will
be
just
moving
over
here
because
I
have
it
prepared.
B
B
The
configuration
for
a
key
clip
CI
is
located
in,
don't
get
laid,
see,
I,
don't
so
I
can
just
select
it
over
here
and
the
same
most
simplest
playground
or
the
most
simplest
way
to
explore
whether
the
CRM
is
working.
Everything
else
in
departments
is
working,
I
started
out,
saying:
okay,
I
want
to
run
it
in
a
docker
image.
Let's
just
use
CentOS
7
as
a
tag
and
I
define
a
simple
test.
B
Job
basically
freely
definable
without
the
keyboard
and
I
want
to
run
a
specific
command
which
needs
a
top,
and
we
say
script
and
top
again
echo
hello
from
our
lovely
fit.
Let
meet
up
and
just
to
have
an
exit
code.
I'd
say
exit
1,
and
this
is
basically
it
then
I,
say:
ok
I
want
to
commit.
It
basically
feature
meet
up
as
a
new
branch.
If
I
can
talk
quickly
at
first
see,
I
can't
fake
craft.
So
this
is
the
thing
and
and
there's
a
card
I
don't
need
a
new
merge
request
and
with
death.
B
I
can
just
wait
for
everything
apart.
There's
a
nifty
thing
within
the
web.
Ide
is
that
I
have
two
rocket
symbol
over
here,
which
brings
me
directly
into
the
CI
pipeline,
for
you
basically
on
the
side
and
I
can
just
continue
working
if,
when
I
want
you,
but
in
this
example
in
this
exercise,
I
just
want
to
see
what
the
pipeline
is
doing
and
I'm
changing
basically
into
the
CR
ICD
pipeline
view.
B
Some
set
of
stuff
is
going
on
and
then
our
script
is
executed,
so
we're
actually
being
greeted,
and
just
because
the
exit
code
was
run,
it
simulates
failure
and
everything
is
broken
right
now,
but
still
we
are
within
the
web
ID.
We
can
fix
things
and
just
to
see
that
it
basically
works.
I
can
change
the
exit
code
to
zero
and
fix
the
CI
CI
config.
It's
kind
of
like
that
in
this
regard,
still
not
a
much
request
and
I
can
continue
working.
B
This
is
something
which
I've
been
teaching
a
lot
of
people
in
my
past
trainings,
because
it's
really
hard
to
get
started.
In
my
experience
just
to
know
your
building
being
overwhelmed
with
sharing
specific,
are
you
trying
to
build
your
code,
and
this
is
just
one
for
the
ahead
and
the
light
in
between
just
to
get
things
going
and
see?
B
Oh,
this
is
how
it
works,
and
it
also
helps
to
troubleshoot
if,
for
example,
the
CI
runners
are
not
working
for
some
reason,
it's
the
most
easiest
in
the
most
convenient
way
to
just
fire
up
a
shop
and
test
whether
the
the
CI
runners
are
actually
working.
So
still,
this
takes
some
time.
So
it's
actually
it's
finished.
Sometimes
I
just
need
to
talk
over
the
time,
really
quick,
because
the
grandness
going
on
with
that
I'd
like
to
switch
to
some
real-world
examples
onto
some
more
practical
example.
So
I've
prepared
some
go
code
within
the
repository.
B
If
we
switch
back
over
here,
you
can
see
there's
basically
main
function
which
is
executed
as
the
entry
point
when
the
binaries
is
run,
and
it
just
attempts
to
do
the
same
as
before.
I'm,
just
printing
something-
and
if
you
now
imagine
that
this
cold
or
this
lipid
repository
was
given
to
you
and
you
need
to
continue
working
on
it,
you
actually
want
to
know.
Does
it
really
does
it
build?
Does
it
work?
Should
I
can
and
I'll
continue
with
coding?
B
What's
best
practice
in
that
regard,
and
for
that,
since
it's
go
line
or
go,
I
want
to
actually
configure
this
configure.
The
continuous
integration
and
continuous
deployment
thing
and
I
have
two
possibility,
because
you
can
look
it
up
on
the
internet.
You
can
google
for
it,
however
other
standard,
but
we're
in
the
lucky
position.
That's
for
specific
languages.
There
are
CI
CD
templates
already.
B
So
when
I
move
over
to
the
github
CI
Yama
for
the
five
templates,
I
can
just
search
for
go
and
have
my
configuration
overwritten
at
some
point
and
I
get
a
nice
template.
I
do
need
to
make
some
slight
changes
to
it,
so
it
tells
me,
please
add
it
to
your
did
project,
which
is
basically
for
you.
It's
your
namespace
and
then
the
project
name
I'm
just
copying
over
here,
because
I'm
lazy,
I'm,
always
lazy-
and
this
is
the
one
I
need
to
edit
and
I-
also
want
I,
don't
want
to
call
it.
B
My
binary
I
want
to
call
it
oops
Lukie,
which
is
basically
the
binary
which
should
be
built
by
the
stages
and
the
chops
for
the
for
the
template
itself.
It's
basically
using
the
goaline
docker
image
and
doing
all
the
setup
stuff
for
coding
define
some
stages
and
within
the
stages
we
have
a
former
job
and
a
compiled
job
which
basically
checks.
Whatever
the
goal
line.
Format
is
correct
and
then
actually
built
a
binary
and
also
makes
it
an
artifact,
so
we
can
actually
download
it
later
on.
B
For
now,
we
just
want
to
use
that
and
say:
okay
I
will
be
adding
as
the
icon
fake
for
our
goal
code
and
with
that
I'm
committing
the
changes
and
I
don't
know.
This
doesn't
really
does
that
refresh
in
in
the
way.
I
want
okay,
I'm,
not
late,
and
not
even
lazy,
I'm
super
nervous
that
everything
is
working
out
when
I
never
get
into
the
pipeline
code
them
and
encode
in
the
pipeline
overview.
I
can
see
that
there's
the
test
and
the
build
stage,
and
actually
the
first
job
is
running
right
now.
B
B
It's
also
nice
to
see
that
there
are
more
colors
the
build
chops.
The
runtime
of
the
build
chops
is
now
shown
over
here.
I
think
that
it
was
coming
in
the
in
the
in
their
recent
most
recent
versions.
So
it's
really
new
to
me
enters
the
thing.
The
code
which
we
actually
got
in
our
repository
is
broken
because
we're
using
I
am
using
a
package
or
I'm
using
a
function
from
a
package
which
has
not
been
imported.
B
So
this
is
what
the
Chop
is
telling
me,
and
it's
also
telling
me
from
the
pipeline
and
it's
just
refresh.
Ok,
the
Chop
was
broken,
so
the
other
stage
is
not
even
triggered
because
the
job
in
the
line
is
needs
to
be
fixed.
Now,
how
can
I
fix
that?
Well,
I'm,
navigating
into
my
code
and
I've,
prepared
it
already?
We
need
to
in
order
to
call
the
print
line
function.
We
need
to
import
the
FMT
package,
and
this
is
totally
simple.
B
Let's
just
import
that
and
commit
the
change
once
again
fix
the
code
and
it
records
could
commit
it
missing.
Package,
import
and
fix
the
build
is
a
bit
commit
message:
I'm,
always
teaching
people
that,
with
that
I'm,
hoping
that
this
updates
ok,
it
always
takes
a
little
while
I'm
way
too
fast
for
everything,
ok,
and
with
that
it
should
actually
complete
the
build.
Hopefully,
let's
just
check
over
there
again,
it's
basically
when,
when
using
containers,
you
get
the
repute
reproducible
built
environment.
So
it's
always
the
same.
B
It's
using
the
same
base
image
and
then
you
can
just
install
run
specific
things
and
afterwards
it's
it
is
destroyed.
Only
artifacts
and
and
caches
are
shared
between
the
stages
and
chops
between
the
stations
yeah
and
still
sometimes
it
takes
a
while
and
if
you
want
to
like
scale
it
in
a
specific
way.
You
are
advised
to
use
your
own
hardware,
your
own
virtual
environment,
while
you
can
just
install
separate
grammars
and
speed
things
up.
B
Let's
do
a
quick
check
on
the
slides,
I
think
I
skipped
some
yeah
I'm
fixing
the
build,
and
the
next
step
will
be
adding
unit
tests.
Let's
just
check
what
it's
doing.
So
everything
is
fine,
it
invokes
go
build
and
we
are
actually
getting
an
artifact,
it's
totally
nice
and,
as
you
can
see,
on
the
right
hand
side
you
can
download
the
artifact.
B
We
want
to
call
a
function
but
in
the
same
moment,
with
adding
a
function,
we
also
want
to
write
a
unit
test
because
testing
if
the
function
really
works
is
a
good
thing.
There
are
specific
unit
test
frameworks
around
for
for
different
languages,
so
for
Python
or
Perl
for
C++.
Whatever
there
were
external
frameworks
go
line.
Has
that
basically
built
in
so,
if
you're
going
for
a
specific
format
with
like
you
have
them,
you
have
a
code
file
and
then
you
just
say:
okay,
it's
tanuki,
taco
and
tanuki
underscore
test.
B
Co
Co
implements
actually
the
test
for
this
code
and
you
follow
specific
test
interface
and
then,
basically,
everything
is
built-in
and
you
get
even
coverage
reports
for
that.
The
thing
is
I
just
want
to
try
it
out
so
I've
prepared
some
source
code
over
here.
You
need
to
ignore
all
that.
Basically,
the
preparations
here
over
above
this
is
just
a
function
which
is
called
get
a
new
key
and
it
returns
a
string
and
for
testing
this
function.
B
It's
basically
defining
a
table
and
checking
for
specific
strings
which
are
which
should
be
in
in
the
returns
a
string
from
that
function.
This
is
a
pretty
cool
thing
because
it
can
just
add
like
20
tests
or
20
checks
inside
and
any
time
it
fails.
You
get
a
feedback
immediately
and
executed
exactly
during
the
CI
pins
right
now
the
code
is
not
really
executed,
so
we
just
want
to
do
that
and
I'm
again
lazy
I'm,
just
including
the
source
code
and.
B
B
B
We
actually
want
to
like
deploy
something
or
we
want
to
do
a
ramp
time
check
or
something
else,
so
not
only
the
binary
we
are
currently
like
building.
Maybe
we
want
to
create
a
package
out
of
it,
want
to
install
the
package
and
then
later
on,
install
the
package,
shoot,
production
or
we'll
just
create
a
binary
and
want
to
like
also
get
some
wire
use,
ansible
tasks,
telephone
whatever,
which
is
basically
done
within
the
deployment
stages.
B
This
only
completes
when,
when
the
stages
before
succeeded,
basically-
and
it's
pretty
easy
to
do
that
or
to
add
that
and
for
that
I'm
just
switching
again
into
the
kidnap
CRM
and
as
you
can
see
over
here,
we
have
inside
stages,
there's
already
deploy
so
I.
Don't
need
to
really
edit
and
I
can
add
another
job
which
I
just
call
run.
B
I
assign
it
to
the
stage
deploy
super
super
nice
with
autocomplete
and
I
want
to
execute
a
script
which
basically
execute
a
new
key
binary
I've
created
in
the
previous
job,
and
with
that
I
think
everything
should
be
ready.
So,
in
the
next
step
we
are
grounded
mouki
inside
the
ICD
and
I'm.
Always
watching
out
for
good
commit
messages,
because
in
a
year
when
you
just
read
fix
the
pill
boy,
it's
something
you
don't
really
know
why!
B
So
with
that,
the
pattern
again
has
been
started
and
the
new
thing
is
the
deployment
stage
got
in
your
job.
So,
basically,
right
now
everything
is
everything
is
running
through
coming
back
to
my
slides.
Let's
just
see
what
else
I
got
to
tell
you
up
until
the
pipelines
are
running
yeah.
Another
best
practice
are
variables
and
templates.
So
when
you
have
some
credentials
or
some
sensitive
data
which
you
need
within
your
jobs,
so
basically
a
MySQL
database
base,
password
or
kuben
is
credentials
or
something
else.
B
You
can
actually
use
variables
which
are
passed
as
environment
rather
to
the
runners,
and
you
can
hide
them
in
your
project
settings.
So
you
don't
need
to
actually
write
them
in
your
CI,
sed
configuration
files
and
with
conditions
granted
only
to
for
example,
although
the
project
maintainer
developers
won't
see
those
credentials,
you
can
also
instrument
specific
jobs,
for
example,
to
specify
a
build
type
and
then
a
script
just
builds
released,
binary
packages
or
debug
packages
which
contain
debug
symbols.
B
For
binaries,
and
and
even
more
for
that-
and
you
can
also
like
use
that
in
specific
scopes-
that
you
don't
need
to
write
all
the
time,
my
binary
or
something
else,
but
keep
in
mind
that
not
all
scopes
work
for
replacing
variables.
So
you
actually
need
to
check
the
documentation
to
figure
out.
What's
what's
working
and
what's
not,
if
you're
planning
to
create
jobs
which
always
use
the
same
script
and
stages
and
specific
things
like
if
chops
for
at
at
seven
rotate
and
Center
is
basically
only
use.
B
B
B
Okay,
when
it
comes
to
monitoring
for
the
CICE
pipelines,
one
thing
which
I
recommend
doing
is
using
to
monitor
the
CI
runners
which,
when
you
are
hosting
it
on
yourself
on
your
own
on
your
own
infrastructure,
you
could
use
the
rest
api.
There
are
specific
check
plug-ins
for
different
monitoring
tools
out
there
and
it
let
itself
also
has
self
monitoring,
of
course,
and
CI
CD
analytics,
where
I
can
just
also
do
reporting
on
which
chops
failed
in
the
past.
B
B
I
couldn't
make
it
happen
to
make
it
click
ever
because
it's
a
shell
output,
but
still
that's
an
that's
an
older
demo
website
I
created
but
I,
totally
love
it
and
I
just
wanted
to
share
it
with
you,
and
maybe
someone
now
feels
old,
because
he
knows
he
or
she
knows
KP,
it's
it's
life,
so
you
can
just
open
it
and
stress
it
yeah
and
what
else?
Well,
I
wanted
to
share
with
good
lab
CICE,
it's
also
possible
to
maintain
your
own
container
images.
B
For
example,
if
you're
planning
to
heavily
use
your
own
custom,
docker
images,
you
can
do
that
you
can
even
install
specific,
build
tools
inside
because
downloading
them
all
the
time
is
pretty
time-consuming.
You
can
spin
it
to
specifics
of
the
versions,
for
example,
if
you
installed
it
with
pip,
install
or
something
else,
and
once
you
you
are
sure
you
want
to
really
rebuilt
the
container
images
you
can
just
do
that,
I
think
with
12.8
or
12.7.
The
container
registry
was
enabled
by
default,
so
can
actually
make
use
of
get
labs
container
registry
and
yeah.
B
You
can
also
use
these
token
images,
for
example,
to
create
your
own
build,
build
architecture
for
creating
binary
packages,
so
real
good
packages.
Fpm
is
not
really
good.
In
my
opinion,
I
am
planning
to
do
some
webinars
later
this
year,
because
I
really
think
that
best
practices
should
be
shared
if
you're
interested
in
that,
please
keep
in
touch
or
get
in
touch
with
me
or
we'll
just
talk
about
it
later.
On
with
that,
there
are
other
things
where
we're
or
you
can
try
it
out.
B
One
thing
is
called
coverage
so
within,
for
example,
golang
you
get
some
test
results
which
are
possible
yrx,
and
you
can
specify
that
within
project
settings
and
cried
out
I'm
leaving
that
as
an
exercise
to
you.
So
if
you're
trying
it
out
and
if
you
like,
try
it
out
and
it
works
tweet
me
then
share
it
on
Twitter,
a
blog
about
it,
share
your
findings.
Also,
if
you
want
to
highlight
specific
patches,
which
means
build,
passing
was
something
else
or
build
is
broken,
which
is
not
that
good.
B
You
can
also
generate
this
patch
and
embed
them
into
your
week
me
and
the
last
run
is
yeah.
If
you're
used
to
current
code
you've,
your
start
coding,
you
break
something
because
breaking
things
is
always
fun
and
you
can't
just
try
it
out
in
practice.
Practice
working
with
the
web
IDE
with
CI
conflict
with
code
and
don't
give
up
I
think
it
took
me
20
or
30
comets
until
I
was
totally
up
and
running
with
get
lip
CI,
so
replays
and
squash
is
totally
fine.
B
If
you
know
all
that
from
Kate
just
to
round
up
and
I
think
somehow
good
in
time,
hopefully
you
can
use
tulip
idea,
basically
for
setting
up
the
ICD,
so
I
prefer
that
way.
I
know
that
there
are
different
ways
like
to
see
at
I
automation
or
this
departmental
web
interface,
but
still
the
web
IDs
I
think
on
a
good
boy.
You
like
replace
existing
IDs,
and
you
can
also
use
it
for
simple
coding
tasks.
B
B
That's
of
course,
a
good
idea,
unit
tests,
test
coverage
and
all
sort
of
pointing
to
the
credentials,
keeping
safe
container
images
and
yeah.
That's
basically
it
and
last
but
not
least
before
I
say
thanks.
I
want
to
highlight
one
of
our
gains
from
the
technically.
When
a
evangelism
team,
because
you
know
next
week,
q
con
would
have
taken
place.
It's
not,
but
we
decided
to
make
it
a
cloud
native
summit
which
happens
as
a
worst
event.