►
From YouTube: Twitch Stream 03 - GitHub and tekton triggers
Description
Adam, Dewan, and JJ play with tekton triggers to get webhooks created and built programmatically.
There were some challenges, but we did get it done!
(We'll add timestamps :soon:)
A
B
A
A
C
B
A
B
D
B
Hi,
I'm
jj,
I'm
also
a
developer
advocate
for
ibm.
Yes,
it
is
just
happenstance
we're
all
ibmers.
This
is
not
an
ibm
stream.
This
is
actually
the
cd
foundation,
but
we
discovered
all
three
of
us
worked
at
ibm,
so
that
was
just
like
happenstance,
but
we
we
we.
This
is
our
third
twitch
stream
that
we've
done.
D
Yes,
it's
going
to
be
all
about
using
webhooks
of
tactile,
so
there's
going
to
be
several
ways
to
do
it.
You
can
just
use
normal
tectonic
triggers
you
can
use
the
webworks
extension
you
could
use
jenkinex.
There's
probably
like
you
know
a
dozen
other
individual
ways
of
doing
it.
I'd
like
to
really
look
at
those
approaches
and
see
which
ones
are
actually
good,
which
ones
work
which
ones
are
difficult
and
hopefully
deploy
something.
Maybe
the
catap.
That's
the
existing
application.
D
B
At
least
for
me,
there's
confusion
right
and,
if
I'm
confused
like
as
they
say
back
in
in
middle
school
or
high
school,
if
you're,
if
you're
confused
in
the
classroom,
raise
your
hand
because
there's
a
very
strong
possibility,
someone
else
is
confused
in
this
room
too.
So
don't
worry
about
it
so
adam.
What
is
a
trigger?
D
Yeah,
so
without
using
the
word
trigger
the
definition
of
triggers,
it's
basically
a
way
to
execute,
as
my
synonym
for
triggers
techton
resources
like
pipeline
runs
are
test
runs
based
on
something
happening,
so
maybe,
for
example,
a
request
gets
sent
to
something
so
like
an
event.
Listener
is
a
techton
crd,
they've
introduced,
and
because
it's
got
this
event
coming
in,
it
will
kick
off
what
you
tell
it
to
which
is
done
by
a
trigger
template.
D
So
basically
the
way
of
saying
when
this
action
occurs
go
ahead
and
kick
off
this
thing
and
a
common
use
case
might
be
github
web
hooks
or
get
lab
web
hooks.
You
may
say
when
you
receive
a
web
hook,
you
get
all
sorts
of
payload
data
and
then
you
want
to
run
a
pipeline
and
some
of
that
data
will
be
used
basically
from
the
github
web
box.
So
maybe
it
have
like
the
branch
in
it
or
the
repo
url
anything
else,
and
it's
a
way
of
basically
kicking
off.
Those
tecton
runs
using
crds.
D
So
tecton
has
a
lot
of
crds
already.
You
have
tecton
task
effect
on
pipeline
a
pipeline
we're
an
attachment.
For
example.
Now
you
have,
I
think,
four
more
there's
trigger
template
trigger
binding
event,
listener,
there's
also
cluster
trigger
binding
as
well.
So
it's
an
extension
to
the
core
group
of
tektron
resources
and
basically
the
main
thing
you
edit
is
the
event
listener.
D
You
have
this
listener
running
somewhere
and
you
say
in
the
spec
you
can
say
trigger
this
template
and
you
know
what
why
don't
we
bring
up
the
main
readme
for
triggers
and
there's
a
really
nice
diagram.
I
want
to
show
people
if
you
bring
up
your
screen,
share
and
head
to
the
docs
folder,
there's
a
nice
little
diagram,
which
shows
the
floor
better
than
I
can
explain.
D
That's
the
one
there
so
an
event
comes
in,
and
the
trigger
binding
basically
allows
you
to
take
things
from
that
event
and
pass
it
through
to
the
trigger
template
and
then
the
trigger
template
might
do
things
like
create
a
pipeline
run
or
create
a
bunch
of
pipeline
resources
that
will
run
in
your
cluster.
That's
really
what
the
flow
is.
Hopefully,
it'll
be
easier.
If
you
actually
show
your
things
working
because
you'll
see
the
disasters
being
created
and
then
the
actual
side
effects
like
a
new
pipe
by
one
gets,
kicked
off.
B
Nice
thanks
okay,
so
we
have
the
cat
app,
which
is
kind
of
our
toy
app.
That
we've
been
using
to
explain
the
this.
This
narrative,
which
is
just
a
simple
node.js
app
that
shows
pictures
of
cats
right,
which
is
great
and
we've
done,
we've
been
able
to
get
it
to
build.
We've
been
able
to
get
it
to
even
give
us
some
slack
messages
too,
if
I'm
correctly,
which
was
pretty
damn
cool
because
in
theory
you
could
have
your
pipeline
inform
you
programmatically
where
it
is
in
the
pipeline
via
slack.
B
So
if
you
have
really
long
pipelines
or
something
like
that,
you
can
actually
just
look
at
the
slack
channel
and
have
the
messages
or,
if
something
breaks
you
can
tell
you
which
is
nice.
The
next
natural
progression
is
the
triggers,
because
ci
cd
is
not
great.
If
you
have
to
go,
kick
it
off
yourself.
Is
that
a
an
accurate
statement,
yeah
100.
D
B
B
There
we
go
perfect,
so
we
got,
we
have
all
three
being
able
to
talk
to
it
and
we
have
okay
get
name
spaces.
We
already
have
tekton
pipelines
already
installed
on
this,
which
is,
if
you
didn't
know
it,
builds
it
inside
of
its
own
namespace,
and
this
is
one
of
the
clusters
that
we've
been
playing
with.
It
is
a
little
bit
dirty
as
you
can
see,
because
you
know
I
do
this.
B
D
B
Perfect,
so
the
best
way
to
do
this,
no
joke,
is
just
to
look
at
the
docks.
That
is
the
way
to
make
it
happen,
so
we
want
to.
As
you
saw
here,
I
went
to
the
docs
section
of
the
getting
started
and
it
sends
us
to
another
link
which
is
install
triggers,
which
is
exactly
what
we
want
to
do
and
what
we
are
going
to
do
as
my
voice
cracks,
because
I'm
a
teenage
boy.
B
And
just
for
friends
at
home
there
I
put
the
link
in
the
chat.
If
you
have
a
desire
to
you
know
do
this.
Also
for
yourself
I
will
say
I
have
gotten
techton
to
be
pretty
happy
running
it
on
a
kind
cluster
which
is
kubernetes
and
docker,
but
obviously
what
we
are
doing
here
with
web
hooks
and
things
actually
out
on
the
real
internet,
and
I
can't
believe
my
voice
is
cracking
again.
This
is
really
embarrassing.
B
The
you
do
need
routing
back,
so
you
will
be
able
to
run
these
things.
These
commands
inside
of
kind,
but
unless
you
have
it
exported
after
exposed
to
the
outside
world,
it's
gonna
you're
gonna
have
some
issues,
but
just
a
friendly
fyi,
all
right,
so
we've
got
that
up.
We've
got
to
point
some
deployments.
B
B
Okay,
I
am
very
lazy,
as
we
will
discover
very
quickly
jet
pods,
all
right.
We
saw
some
deployments
being
created,
so
I'm
assuming
they're
put
inside
the
checktown
pipelines
thing
and
look
look
right.
There
we
looks
like
we
have
a
controller
and
a
web
hook
up
and
running
is.
Is
that
that
is
what
we're
expecting
correct,
adam.
D
A
B
B
D
D
A
D
B
D
D
I
think
the
only
challenge
I
can
see
personally
is
getting
that
publicly
available
endpoint
for
github.com
to
be
able
to
send
its
webhook
too.
That's
the
only
thing
I
can
see
it's
going
to
be
effectively
difficult.
We
know
that
if
he
was
using,
you
know
a
private
network
with
maybe
gitlab
on
the
same
network.
That's
fine,
but
github.com
is
where
most
people's
code
is
I'll,
assume.
B
D
So
I'm
thinking
our
event,
listener
will
sit
and
use
that
sub
domain
and
that
will
receive
the
webhook
event
and
the
template
will
be
the
pipeline
run
for
qatar.
So
basically,
maybe
do
a
push
to
catalog
it'll
route
to
the
event
lessener
through
the
ingress
and
then
kick
off
the
pipeline
run.
So
that's
good.
I'm
glad
you've
got
that.
D
B
D
B
B
D
D
B
D
C
D
D
B
D
Now
one
thing
to
mention:
jj
open
that
file.
Again,
let's
see
what
else
is
in
there,
so
I
think
I
saw
some
more
getting
started
things
if
you
do
a
search
on
getting
started
the
name
of
it
for
the
pipeline
one's
development.
I
don't
want
to
care
for
that.
So
it's
got
there's
a
pipeline
that
it
refers
to
it's
going
to
use
that
as
the
place
to
push
the
image.
D
D
So
I'm
guessing
I'm
going
to
be
using
a
tech
time
task
to
both
create
the
english
and
the
web
book.
Instead,
the
money
we're
doing
it
through
the
ui
for
the
web
part
at
least
that's
to
check
that
song
in
this
name
space.
So
again
you
might
want
to
do
your
take
it
all
minus
then
it
should
bring.
You
reverse
search,
oh
yeah,
but
yeah.
D
There
you
go
all
right,
so
we've
got
triggers
installed.
We
have
anything
else
interesting
here.
Is
there
an
event
listed
yeah?
There
is
el
getting
started
so
you've
got
your
event
listener.
It's
desired1
ready
zero.
So
if
you
do
a
qct
I'll
get
pods,
I
wonder
if
your
event
listeners
failing
to
start
up.
We
don't
need
that
event
listener.
B
D
B
D
Right
so
the
name
space
bit,
we
can
get
rid
of
again.
You
still
get
that
line,
create
certificate,
find
a
secret
phrase.
Eh
external
domain
eat
that
thing
that's
interesting.
Ian
coffee
is
the
one
that
came
up
with
the
first
event
listener
thing
like
over
a
year
ago,
and
it
became
triggers
in
the
end.
So
this
has
got
remains
of
it
being
quite
an
old
original
thing,
I'm
guessing
that's
no
domain,
there's
any
guidance
on
that
should
be.
If
you
go
back
to
the
docks.
D
D
B
B
D
D
D
B
D
B
D
D
If
I
get
older
of
all
bindings,
do
you
have
anything
interesting?
We
don't
okay.
So
what
we
need
to
do
is
either
use
a
cluster.
Well,
it's
got
our
permissions
or
make
a
new
one
and
give
it
the
permissions.
Now
I
think,
last
time
we
did
this.
We
augmented
an
existing
existing
role.
Didn't
we,
I
think
we
were
a
bit
naughty
and
that's
the
role
binding.
I've
got
any
cluster
role
bindings
because
there's
two
different
levels,
so
there's
one
that's
available
for
every
name
space
ever
right
I
want
to
do.
D
A
D
D
D
D
B
A
D
D
B
A
D
D
A
D
B
B
D
D
D
A
D
D
D
B
C
C
B
D
B
I
will
say,
though,
like
if,
if
you
look
at
the
triggers
and
if
I
was
a
first-time
user
of
techton,
that's
great-
to
have
an
example
on
how
to
do
it
like
a
fast
time
or
quick
time
to
dopamine
or
whatever.
The
statement
is
right
but
like
if
you're
looking
at
the
getting
started
right
like
which
would
you
do
examples.
A
D
A
D
If
I
do
get
deployments
again
right
so
there's
and
I
thought,
if
you
get
it
as
yaml,
you
see
the
reason
for
it
not
being
one
of
okay.
There
we
go
forbidden.
Looking
up
sales
captain
triggers
admin.
Now,
if
we
do
a
great
on
tecton
triggers
admin
right,
so
let's
do
k
pi
minus
f.
Does
it
say
to
apply
the
entire
folder
or.
D
Yeah,
you
know
what
I'm
just
gonna
just
do
this
right
and
now
I
wonder
if
my
deployment
needs
a
bit
of
a
kick.
So
if
I
just
get
that
deployment
again,
we'll
see
what's
happening,
nothing
much
so,
let's
just
redo
the
the
run.
Shall
we
so
if
I
do
k
put
it
into
a
k,
delete
minus
f,
ingress,
one
okay,
cool
and
then
do
my
apply
again
and
I
should
get
a
new
deployment.
So
if
I
do
k
get
deployments
just
on
its
own,
I
should
see
against
the
listener,
but
that's
20
minutes.
D
B
A
D
D
D
D
B
B
B
B
A
D
We've
got
the
pipeline,
we
need
to
add
the
secret.
No,
I
don't
believe
that
step.
I
think
that's
outdated.
We've
got
your
docker
secret
already,
which
is
cool.
I'm
not
going
to
show
that
as
yaml.
Don't
worry,
we've
installed
the
trigger
templates.
We've
installed
the
tasks
for
web
poking
ingress
and
let's
actually
create
a
web
book.
Finally,
after
45
minutes
of
prerequisites,
it's
time
to
create
the
wonderful,
the
almighty
web
book
read
a
personal
artist
talking
jj.
D
D
D
D
B
D
D
D
B
D
While
you're
there,
I'm
just
making
sure
this
report
tasks
will
actually
work,
I
don't
think
it
will,
because
nobody
creates
that
service
account
like
ever.
So.
If
I
do
a
grep
on
that
service,
account
name
so
who's,
creating
it
nobody,
unless
this
file
does
did
we
verify
that
this
file
creates
that
service
account,
or
did
I
dream
that.
D
D
C
D
D
C
C
D
D
D
D
D
D
C
D
D
D
Okay,
that
makes
sense.
What
does
that
make
sense?
Is
this
it's
like.
The
task
doesn't
work
either
the
github
repository
isn't
there
and
it
is
or
the
task
doesn't
work.
So
you
get
four
for
not
found
from
github.com.
So
I
don't
look
at
the
task
itself.
If
I
do
cat
wait,
where
is
for
tasks
in
that
big
file?
D
A
D
Secret,
the
external
domain,
the
github
user,
repo
and
org
we've
got
that
that
already
exists.
Github
domain
can
override,
so
you
don't
need
to
use
on
github.com
and
then
it
goes
and
does
create
webhook
using
this.
So
it's
doing
a
curl
to
a
certain
url.
It's
as
though
that
url
doesn't
and
it
doesn't
exist
anymore.
I
bet
it
does
barrier
I'll
be
worried
if
it
doesn't
or
the
repo
exist.
The
repo
does
exist.
C
B
D
D
D
D
A
D
D
A
D
D
D
D
Admin,
let's
check
the
service
account
name
and
the
secret
are
both.
You
know
bound
together,
great
okay,
so
I'm
going
to
edit
and
exit
this
this
file
and
I'm
going
to
do
yeah,
k,
get
say
and
we're
trying
to
use.
Techton
essay
are
we.
I
think
we
are.
Let's
get
the
tasks
from.
Let's
just
prove
that
believe
that.
D
A
D
D
Cool,
so
we'll
don't
do
that
we
gotta
tidy
up
and
delete
that
existing
test
one
and
then
apply
the
red
book
run
and
then
hopefully
see
something
wonderful.
So
let's
do
that
with
a
watch.
I
want
to
do.
Okay
get
the
plus
minus
w
completed,
so
it
is
quite
quick.
It's
just
a
little
curl
command
and
if
I
get
the
logs
for
that,
all
containers
not
found
again
now.
Let's
just
do
some
sanity
checks,
we
do
have
a
secret
called
web
books,
webpack
dash
secret.
I
won't
begin
it
as
yaml.
D
C
D
B
D
B
D
D
D
D
D
B
D
B
B
D
D
B
D
B
D
D
A
D
D
D
B
D
D
D
D
B
D
D
A
D
A
B
D
C
D
D
D
D
D
D
B
D
A
B
B
D
But
what
does
the
ingress
do?
So?
I'm
getting
out
of
that.
If
I
said
your
ingress
is
event
dash,
lesnar
right
and
I
expect
that
ingress
to
have
a
path
to
a
an
event
listener,
so
it
does
world
bouncer
some
ip
and
then
service
name
event
dash
listener.
If
I
do
svc
event
listener,
do
you
exist
unauthorized?
D
D
D
D
We're
going
to
make
that
called
el
getting
started
the
listener,
because
I
don't
have
to
change
the
one
either,
so
you
make
that
yell
getting
started
listener
and
the
same
port.
Hopefully
that's
not
wrong
as
well.
I'm
gonna
save
that
we
got
get
your
existing
ingress
and
that
goes
in
the
bin
goodbye
and
then
we're
going
to
create
that
ingress
run
again,
but
it
should
create
a
new
ingress
which
routes
it
to
the
correct
name,
service.
C
D
C
D
A
D
D
B
D
D
D
D
D
C
B
D
B
This
is
something
like
as
as
three
professional
web
professionals,
I
guess,
is
the
right
way
of
describing
this.
This
should
not
be
this
hard
right,
yeah
this
this
like.
If
we
like,
if
we've
scoped
this
down,
we
have
the
proof
by
this
video.
If
we've
scoped
this
down
and
said
our
ingress
is
really
hard
for
us
to
configure-
or
it's
not
obvious
to
what
we
need
to
do
to
like
make
the
little
line
from
ingress
to
app
actually
work.
That's
that's
a
failure
on
our
part
as
ibm.
D
B
B
D
C
D
A
B
D
A
D
D
B
D
D
That's
all
I
thought
so,
but
you
need
to
do
a
posh
repository.
So
why
don't
you
push
the
repository
and
we'll
see?
If
anything
happens,
you
could
do
a
k
get
all
minus
w.
So
you
can
see
if
anything
gets
created
in
this
name
space
and
then
push
the
cation
and
then
something
might
appear.
It
may
not
build
a
project,
but
it
may
kick
off
things.
B
D
D
B
Yeah,
no,
that
there's
there's
definitely
some
p.
I
don't
know
when,
but
there
are
some
pr's
for
dot
pr's
coming
out
of
this
thing,
because
this
is
this
was
not.
There
were
some
very
obvious
questions
that
we
had
to
dig
for
to
find
the
answer
that
we
should
have
proactively
had
in
the
doc
and
also
we
need
to
rip
out
getting
started
from
this
thing,
because
that
was
just
confusing
like
either
we
have
the
simple
example
or
the
getting
started,
because
that
is
that
is
just
that
is
miscommunication
right.
There
yeah.
D
I
got
a
little
confused
okay,
so
we've
got
something
that
kicks
off
a
is
that
a
pipeline
or
a
task
room.
If
we
do
get
pr,
we
won't
debug
it
now,
but
if
you
can
get
pr,
if
you've
got
a
new
pipeline
run
okay,
if
you
get
it
as
yamo
the
minus
does
it
say
at
the
top
event
id
that
means
it
was
made
by
a
trigger
so
right.
On
top
of
that,
if
you
want
to
arrest
it
or
pipe
into
files
whatever
there
should
be
a
event
id
somewhere
in
here.
D
A
D
Well,
I
think
it'll
be
easy
for
first-time
users
if
we
document
that,
because
it
wasn't
actually,
if
nothing
was
technically
at
fault,
triggers,
wasn't
at
fault
ibm,
causing
that
fault
kate.
On
problems
at
fault,
our
code
wasn't
at
fault,
it
was
just
we
didn't
configure
it
correctly
for
what
to
do
yeah
like
nothing's
broken.
It's
just
made
better
docks
for
our
environment,
at
least.
B
Our
obvious
our
happy
path
like
what
we
need
to
do,
and
maybe
this
is
specific
for
ibm
cloud,
but
we
need
to
create.
We
need
to
revisit
what
the
steps
we
went
through
being
that
we
have
the
video
and
all
that
we
need
to
find
that
happy
path,
create
that
path
to
be
repeatable
and
then
either
upstream
it
to
tecton
or
put
this
on,
like
the
ibm,
developer,
blog
or
whatever,
as
this
is
how
to
get
tecton
triggers
to
work
with
ibm
and
that
I.
D
Think
you
need
format
because
there's
a
lot
of
steps
for
google
cloud
already
and
we
can
just
say
by
the
cloud
make
sure
you
do
your
desk
with
secure
ingress.
Absolutely
we
need
to
encode
those
secret
credentials.
Don't
we
talk
to
them
the
secret
as
angry
shifted
earlier.
Finally,
the
main
reading
period.
B
Absolutely,
and,
and
and
the
beauty
of
it
is
we
have
like
this-
is
something
that
we
people
will
run
into
right
because
they
are
going
to
want
to
get
this
stuff
to
happen
and
have
the
happy
path,
steps
that
we'll
also
need
to
test
right.
The
irony
is
we'll
have
to
create
ci,
cd
or
ci
around
these
steps
to
make
sure
that
they
always
run
to
make
sure
they
run
to
make
sure
they
work.
B
D
We
could
do
the
next
stream
as
getting
those
prs
done
together
and
making
sure
the
actual
trigger
template
and
binding
works
and
deploys
the
thing,
because
ours
is
arabing
for
some
reason,
so
we
can
revisit
triggers,
have
a
happy
path,
vd
vision,
type
objective,
with
a
happy
path,
that'll
take
two
minutes
and
voice
will
say:
oh
that's
how
you
do
it
cool
yeah
yep.
I
like.
B
It
I,
like
it:
okay,
cool
one
last
call
out
to
the
people
on
the
stream
we're
there
there's
about
a
couple
second
delay,
so
we'll
stick
around
for
about
30
odd
seconds
or
so.
If
anyone
has
any
questions
or
thoughts,
please
don't
hesitate
to
throw
them
in
the
chat,
while
we're
still
here
other
than
that,
I
think
we're
pretty
much
done.