►
From YouTube: TGI Kubernetes 111: CloudEvents and Argo Events
Description
Show notes: https://github.com/vmware-tanzu/tgik/blob/master/episodes/111/README.md
- 00:00:00 - Welcome to TGIK!
- 8:00 Sysdig Kubernetes 1.18 blog post
- 10:51 CNCF CNCF Webinar kubernetes-1,18 from the release team!
- 13:00 Digitial Ocean Kubernetes tutorials in diffferent languages
- 18:35 The start of the CloudEvents and Argo Events discussion
Come hang out with Joe Beda as he does a bit of hands on hacking of Kubernetes and related topics. Some of this will be Joe talking about the things he knows. Some of this will be Joe exploring something new with the audience. Come join the fun, ask questions, comment, and participate in the live chat!
This week we'll be looking at CloudEvents (https://cloudevents.io/) and Argo Events (https://argoproj.github.io/projects/argo-events). These projects help to shape/transform/deliver/launch stuff based on events. Let's learn together!
A
Hello,
hello,
hello,
welcome
to
another
episode
of
tea,
GI
kubernetes,
I'm,
Joby
I'll
be
your
host
for
today
and
for
those
not
in
the
know,
TGI
kubernetes
is
a
weekly
livestream
that
we
do
where
we
cover
all
things.
Kubernetes
I
started
it
when
we
were
a
small
baby,
startup
called
hefty,
oh
and
and
we've
joined
VMware,
and
here
we
are
and
yeah
so
every
week,
I
try
and
go
online,
learn
something
new
or
me
or
some
other
folks.
Talk
about.
A
You
know,
what's
happening
in
the
ecosystem,
what's
happening
with
kubernetes
and
then
and
then
dive
into
some
technology.
So
yeah
and
I
haven't
been
able
to
do
as
many
recently
as
I
really
want
to,
and
so
Duffy's
been
helping
to
pick
up
the
slack
and
so
just
a
huge.
Thank
you
to
Duffy
for
really
doing
some
awesome.
Tgi,
K's
well
well,
I've
been
so
busy
doing
so
many
other
things
yeah.
So,
ok,
so,
first
of
all,
what
I
love
to
do
is
I
love
to
say
hi
to
everybody
from
around
the
world.
A
That's
joining
us
I
know
that
this
is
the
worst
time
ever
because
you
know
you
all
are
done
for
the
week.
But
hopefully
you
know
you
can
either
catch
this
live
or
catch
this.
After
the
fact,
we
have
well
lead
from
saying
hello
and
telling
us
we're
talking
about
kubernetes
vs
cube
vs.
kate's.
When
we
write
stuff,
it's
the
the
k8s
actually
came
later,
I'm,
not
sure
who
introduced
that
we
started
talking
cube
early
on
when
before
we
went
public.
A
But
then
the
k8s
nomenclature
came
later
Martin
from
the
Netherlands
pratik
from
India
Rory
from
Scotland
gladsome
good
to
see
you
Alex,
who
eats
kittens,
please
don't
eat
kittens
rodolfo
Vivian
from
Munich
Duffy
is
actually
sort
of
watching
on
the
on
the
other
end
here,
helping
out
l'm,
addi
good
to
see
ya
Phil
from
New
York,
Alex
yeah.
So
I'll
tell
you
about
my
quarantine
experience.
You
know
where
I'm
getting
a
little
getting
a
little
crazy
but
staying
home
staying
safe.
A
The
kids
are
in
the
house
and
they're
they're,
getting
a
little
stir-crazy
we're
cooking
a
lot
zoltes
from
Finland
Jason,
except
yo
person
with
us
here
at
VMware
yeah.
You
see
my
hefty
o
posters
that
we
had
in
the
background
we
had
these
commissioned
right
before
we
sold
the
VMware
and
they're
just
so
awesome,
I'm
kind
of
bummed
that
we're
not
using
them
more
so
I've
been
I,
had
them
printed
up
on
metal
and
they're
sitting
behind
me
here.
A
Let's
see,
yeah
George
is
off
today,
so
he's
doing
some
family,
stuff,
I,
think
and
yeah.
So
he's
not
he's
not
helping
out
Mo's
upper
good
to
see
you
Tim
from
Frankfort
Thomas
Sadiq
from
Toronto
slim
and
trim
yeah.
You
know
a
lot
of
stress
with
the
quarantine
and
the
virus.
I
think
I'm
losing
weight
because
of
it
all
right.
You
know,
let's
see
so
we
have,
and
then
Scott
is
in
Seattle
he's
one
of
the
folks.
A
That's
doing
some
of
the
cloud
event
stuff
that
also
a
member
of
the
K
native
team.
So
he
can
definitely
give
us
some
feedback
on
the
cloud
event
stuff
and
we
have
a
vibe
hob
III.
Tell
me
how
to
pronounce
your
name.
If
you
can
do
it,
phonetically
who's
joining
us
also,
who
is
I,
believe
one
of
the
folks
behind
the
Argo
event
stuff
that
we're
gonna
be
taking
a
look
at
so
yeah,
so
we
so
yeah.
So
we
got
stuff
going
oh
and
Tania's
asking
why
the
astronauts
on
the
posters
at
the
mo
team?
A
A
Want
these
things
to
sort
of
be
about
sort
of
exploration,
trying
new
things,
because
you
know
that's
a
lot
of
what
we're
doing
with
kubernetes
with
infrastructure
and
stuff,
and
so
my
inspiration
there
was
and
I
can
take
the
camera
off
and
we
can
take
a
look
at
it
were
these
posters
that
are
actually
made
by
NASA
that
are
sort
of
Mars
exploration
posters.
So
those
things
and
I
don't
know
the
artist's
name
off
the
top
of
my
head.
A
But
I
mentioned
this
to
Scott
our
marketing
person
and
he
actually
went
off
and
found
that
artist
and
commissioned
him
to
actually
do
the
the
hep
do
posters
also,
and
so
I
really
think
these
are
cool.
So
we
have
like
a
sort
of
maritime
Explorer
with
a
8-spoke
wheel,
I'm
not
going
to
complain.
We
have
like
a
jungle
explorer
like
an
astronaut
and
then
somebody
who's
like
going
through
the
desert
or
maybe
they're
on
Mars
I'm,
not
sure
so.
Yeah
really
really
excited
about
those
I
love
them.
Okay,
yeah!
A
So
today,
I
welcome
to
everybody,
I'm,
just
gonna
jump
in
because
otherwise
we'll
never
get
going.
Let
me
switch
to
my
screen
here
so
Geordi.
Yes,
this
does
get
recorded.
If
you
go
to
TGI
Kyo,
you
can
see
the
playlist
of
all
the
TGI
Ches
and-
and
that
is
the
we
record
all
of
these
things,
because
it
is
tough
to
catch.
These
things
live
for
sure,
Frederico
good,
to
see
ya
all
right.
A
So
today
we're
going
to
be
covering
cloud
events
and
Argo
of
ents
and
sort
of
I'm
gonna
be
digging
into
sort
of
the
venting
landscape
in
general.
Talking
with
Scott
I
think
it
might
be
interesting
to
look
at
how
events
are
used
in
ke
native
and
a
future
episode,
but
we're
not
going
to
be
looking
at
that
today.
I
think
I
wanted
to
do
something.
A
That
I
figured
I'd
dig
into
another
one
of
the
the
Argo
projects
as
part
of
this,
but
there's
a
lot
of
exciting
stuff
going
on
this
week
and
a
huge
thank
you
to
to
Duffy
for
pulling
together
all
the
links
here.
So
the
the
kubernetes
1.8
team
released
I
believe
yesterday
and
so
a
huge
thank
you
to
all
the
folks
that
actually
made
that
happen.
Doing
a
kubernetes
release
is
is
not
easy.
There's
a
lot
of
moving
parts,
there's
a
lot
of
cats
to
herd,
and
so
it's
it's.
It's
an
amazing
accomplishment.
A
Every
time
we
actually
make
this
stuff
happen,
and,
and
so
huge
thank
you
to
the
folks
drivin
that
so
there's
a
blog
post
here
that
to
a
lot
of
the
details
talking
about
all
the
stuff,
that's
actually
making
its
way
to
stable
a
lot
of
really
good
stuff
happening
here.
There's
some
other
changes
that
are
happening
in
terms
of
some
of
the
earlier
stuff.
That's
in
the
beta
phase.
A
Right
now,
and
just
as
a
reminder
when
we
mark
api's
and
in
kubernetes
as
being
alpha
or
beta,
that's
really
about
the
sort
of
ongoing
commitment
to
that
API.
Once
something
hits
stable,
we
have
a
deprecation
policy
where
we
won't
change
or
deprecated
that
API
for
I
believe
three
or
four
versions,
whereas
when
we're
in
beta,
we
want
the
implementation
to
be
solid,
but
we
also
know
that
we're
still
learning
about
the
best
form
of
the
API
and
so
a
lot
of
the
stuff.
A
A
It's
a
Large,
Hadron
Collider,
inspired
logo
for
kubernetes,
1.8
and
I
love
that
you
know
the
folks
who
put
together
these
releases
sort
of
have
that
sort
of
esprit
de
corps
around
you
know
in
it
shipping
it
and
then
you
know
doing
some
custom,
sort
of
logos
or
swag
or
something
like
that.
They
can
only
get
if
you're
really
helping
out
and
and
helping
to
get
the
release
out.
A
I
think
that
might
be
fun
also.
But
if
you
go
to
TGI
K
dot,
io
/
notes,
you'll
actually
get
to
you'll
actually
get
to
this,
and
you
can
add
some
your
own
notes.
You
know,
have
some
some
of
the
sort
of
time
things
and
I'm
going
to
copy
and
paste
this
into
the
into
the
repo
as
we
go.
So
cystic
did
a
really
nice
blog
post
of
what's
new
and
1.8,
also
and
I
think
they
really
called
out
some
of
the
some
of
the
most
exciting
things
that
are
going
on
here.
A
So
the
you
know,
there's
some
really
good
security
stuff
here,
there's
been
a
bunch
of
stuff.
That's
been
ongoing
around
being
able
to
use
service
account
tokens
for
more
than
just
talking
back
to
the
kubernetes
api
server.
So
that's
really
good
to
see.
There's
a
API,
refactoring
and
change
with
the
the
signing.
The
certificate
signing
API
cube
control,
debug
I
haven't
seen
this
yet
so
this
creates
debug
containers
and
redeploy
pods
with
the
different
configuration
I
mean
I
would
love
to
play
with
this
sometime.
A
So
if
you
all
have
you
know,
changes
that
you
think
are
a
big
features
in
1.18
that
you
think
we
should
cover
at
some
point,
definitely
drop
it
in
the
comments
or
file
an
issue
in
the
TGI
K
repo
and
we'll
definitely
be
looking
at
that
as
something
that
we
can
dig
into
and
future
episodes
of
TGI
K,
all
right.
So
I
want
to
keep
moving
here.
A
A
What's
new,
so
that's
awesome,
so
Jeremy
and
I
know
I
saw
on
Twitter
that
Jeremy
has
some
his
his
father
is
sick,
and
so,
let's
send
good
positive
thoughts
out
to
Jeremy
right
now
and
so
they're
going
to
be
going
over
the
details
of
what
happened
in
1.18.
So
that's
that
looks
good,
alright.
So,
on
the
tansu
side,
so
on
the
VMware
side,
I
just
want
to
say,
like
we
did
a
big
announcement
on
March
10th
really
exciting
here
talking
about
VMware's
investments
in
the
kubernetes
space.
A
Both
are
you
know,
as
we
look
to
build
a
business
around
kubernetes
to
complement
a
lot
of
the
work
that
we're
doing
upstream,
and
so,
if
you
want
to
check
out
what's
going
on
there,
this
is
a
great
blog
post,
and
then
we
did
this
like
big
sort
of
video
thing.
That
I
believe
is
probably
linked
off
of
here
also.
A
So
this
is
the
details
of
what
we're
doing
with
tansu
commercially,
so
quick
commercial
plug
there
of
check
it
out,
and
then
the
kubernetes
podcast
from
google
has
an
interview
with
jangly
from
who
works
on
etsy
d,
and
so
that's
a
that
looks
really
really
good
too,
and
so
definitely
check
that
out.
I
haven't
had
a
chance
to
listen
to
it
yet,
but
it's
definitely
on
my
list
yeah.
Let's
see
so
I'm
reading
some
of
the
comments
here
so
once
it
said
our
Thomas
says
that
the
CNC
F
is
doing
a
virtual
conference
next
week.
A
So
TGI
K
TG
t
kg
deployment,
yeah
I,
think
that
would
be
I
think
you
know,
as
I
would
love
to
do
an
episode
where
we
actually
go
through
the
ins
and
outs
of
TKG
and
what
that
experience
looks
like
it's
really
really
built
on
top
of
cluster
API.
So
a
lot
of
the
TG
IKS
that
Duffy's
been
doing
around
cluster
API
are
very
relevant
to
TKG.
Let's
see
so,
here's
a
blog
post
on
migrating
from
helm,
2
to
3
definitely
helm.
A
3
is
a
huge
step
forward
in
terms
of
security,
getting
Reddit
rid
of
tiller
rid
of
tiller.
My
understanding
is
that
there
there
might
be
some
gotchas
as
you
go,
and
so
this
looks
like
a
pretty
comprehensive
blog
post
going
over
sort
of
what
it
takes
to
migrate.
So
thank
you
for
that
and
then,
and
then
this
is
really
cool.
Digitalocean
has
some
of
the
best
tutorials
and
content
out
there.
That
is
really
about
just
sort
of
like
hey
I
want
to
get
some
stuff
done.
How
do
I
make
it
happen?
A
And-
and
this
is
really
fascinating
because
they
have
a
lot
of
these
things
in
different
languages,
which
is
really
cool,
so
13
in
Spanish
20
in
Portuguese
and
14
in
Russian,
and
so
I
love
to
see
that
that
you
know
sort
of
worldwide
focus
as
we
bring
folks
who
don't
necessarily
speak
English
well
make
this
this
technology
accessible
to
them.
So
Lee's
asking
about
sort
of
like
why
was
tiller
a
security
risk,
so
so
tiller,
really,
you
know
uh-oh
Duffy
did
you.
A
We
were
you
doing
a
Gus's
article,
I,
don't
think
the
I
don't
think
the
the
link
came
through,
but
tiller
what
it
did
it
is.
It
ran
generally
with
sort
of
very
wide
privileges
on
the
cluster
and
it
kind
of
existed
before
we
really
had
our
back
in
kubernetes
and
it
had
its
own
API.
That
generally,
you
know,
at
least
with
a
lot
of
default.
A
So
you
could
say
you
know
I
want
to
have
a
average
time
based
on
the
HTTP
verb,
and
so
then
you
would
have
a
metric
based
on
put
and
get
in
head
and
that
type
of
things
there
was
a
problem
in
kubernetes,
where
I
believe
that
the
metrics
were
then
split
out
by
if
I'm
not
mistaken.
Here
were
split
out
by
the
client
ID
on
when
users.
A
Actually,
maybe
this
was
like
the
user
agent
ID
when
calling
the
server,
and
so
that
meant
that
a
malicious
user
could
make
a
bunch
of
calls
to
the
kubernetes
server
with
slightly
different
and
long
client
IDs
and
then
in
doing
so,
actually
cause
the
cluster
to
the
API
server
to
run
out
of
memory,
and
so
I
think.
The
lesson
from
here,
which
I
think
is
really
fascinating,
is
that
you
don't
want
to
you.
A
Have
user
input
drive
the
metrics
that
you're
actually
exposing,
because
if
you
do
that,
then
that
becomes
a
vector
for
essentially
out
of
memory
denial
of
service
attack
on
you,
so
yeah
really
interesting
kind
of
subtle
bug
there,
because
it's
a
it's
a
it's
not
obvious
you're
like
oh
I,
just
want
to
check
the
different
clients
and
how
many
clients
are
contacting
me
and
well.
It
turns
out
that
if
you
let,
if
you
let
the
client
specify
that
ID,
that
becomes
a
problem
alright.
So
that
is
the
news
of
the
week.
A
Clearly,
the
big
news
is
1.18
being
out
there.
We're
all
excited
for
that,
and
and
in
doing
so
I
and
Duffy
helped
me
figure
this
out.
I
upgraded
all
the
stuff
running
on
my
machine
so
that
I'm
running
kubernetes
1.18.
If
I
go
here
and
if
I
do
cube
control
version,
you
can
see
I'm
running
1.18
for
both
the
client
and
the
server
I'm
doing
this
with
kind
on
my
local
machine
and
I
dropped
sort
of,
like
you
know,
on
a
Mac.
A
This
is
how
easy
it
is
to
actually
get
a
local
1.18
server
up
and
running
so
yeah.
So
if
you
want
to
play
around
with
that
there,
you
go.
That's
the
easiest
way,
all
right,
so
there's
two
things
that
I'm
going
to
be
covering
today
and
they're
very
much
related.
So
the
first
one
is
this
project
that
started
as
a
working
group.
It
was
an
output
of
the
server
list,
working
group
in
the
CNC
F,
that's
called
cloud
events,
and
this
is
really
more
of
a
spec
plus
a
bunch
of
tools.
A
It's
not
really
a
system
in
terms
of
like
a
lot
of
the
other
things
that
we
think
about,
and
then
I'm
going
to
be.
Looking
at
Argo
Argo
eventing,
let's
see
so
this
is
Argo
events
as
a
as
something
that
actually
uses
the
cloud
event
stuff
and
we'll
be
talking
about
the
relationship
between
these
things
and
I
think
the
relationship
doing
cloud
events
in
Argo
is
similar
in
some
ways
between,
say
the
relationship
between
events
and
something
like
k,
native
and
so
I
think.
A
It's
really
interesting
that
we
have
we're
finding
ways
to
create
standards
that
create
value
and
ways
to
be
able
to
reuse
things
across
multiple
different
environments,
which
is
really
interesting
and
so
and
so
definitely
looking
for
finding
other
places
where
cloud
events
actually
come
into
play,
whether
it
be
with
K
native
or
beyond
that,
so
there's
a
website,
a
cloud
event,
CO
website
that
looks
really
interesting
and
really
talks
about.
What's
going
on
here
and
fundamentally,
what
cloud
events
is:
is
it's
a
way
to
take
an
event
and
then
express
it
using
a
standard
schema?
A
A
Think
one
of
the
interesting
things
as
I
was
doing
just
a
little
bit
of
cursory
resource
search
here
is
that
there
is
a
slight
difference
between
an
event
and
a
message.
An
event
is
just
a
statement
that
something
happened
at
a
certain
time,
whereas
a
message
generally
has
extra
metadata
around
routing
like
who
are
you
sending
the
message
to?
How
do
you
respond
to
that
message?
A
There's
a
whole
bunch
of
sort
of
you
know,
identity
and
forwarding
going
on
that
doesn't
apply
to
events
that
does
apply
to
messages,
but
I
think
that
there's
there's
a
bunch
of
you
know.
Messaging
systems
can
also
be
used
for
events,
so
it's
very,
very
common
for
folks
to
take
something
like
cloud
events
and
then
throw
that
into
Kafka
or
Nats,
or
you
know,
transmitted
over
something
like
HTTP
or
MQTT.
A
A
I've
bought
these
off
of
Amazon.
You
can
get
these
like
little
smart
plugs
and
these
things
are
like
Wi-Fi
enabled
plugs,
and
you
can
like
flash
this
custom
firmware
on
it,
and
I
got
this
from
my
coffee
machine
and
and
for
this
one
to
do
the
custom
firmware
you
have
to
like
open
it
up
and
solder
a
jumper
or
solder
a
controller
on
there.
But
the
idea
is
that,
like
we
see
these
echoes
between
sort
of
like
okay,
how
do
we
have
events
and
communication
in
sort
of
like
an
IOT
context?
A
How
does
that
relate
to
sort
of
like
I
check
something
into
github?
It's
really
fascinating
to
me
at
least,
to
see
these
sort
of
worlds
colliding
through
protocols
like
MQTT,
so
some
cool
stuff
there
so
yeah.
So
the
idea
here
is
that
cloud
events
is
essentially
a
standard
schema
for
what
goes
into
an
event
with
a
bunch
of
sort
of
metadata
about
that
vent
event
along
with
sort
of
essentially
a
data.
A
A
So
that
would
be
an
example
of
how
these
things
play
together
and
or
not
actually
know
the
source
might
be
yeah,
maybe
I'm
wrong.
Okay,
the
source
would
be
the
bucket,
for
example,
and
then
the
subject
would
be
a
the
name
of
the
object
and
then
the
type
could
be
something
like
a
new
object,
created
right
and
so
standardized
sort
of
how
we
encode.
That
stuff
is
really
what
we're
getting
into
here
and
then
time,
and
so
it's
a
very,
very
sort
of
thin
set
of
capabilities.
A
But
the
idea
is
that
it
ends
up
showing
up
something
like
this
and
with
the
idea
that
there
is,
you
know,
here's
sort
of
Microsoft
event
grid
right
and
we
have
Google
cloud
functions
and
we
have
AWS
cloud
watch
events
an
IBM
open.
So
all
of
these
things,
you
know
are
events
with
their
own
way
of
expressing
it
in
cloud.
A
Let's
see
he
sent
me
some
interesting
stuff,
that's
actually
coming
on
here.
So
there's
like
here's
an
example
of
essentially
creating
a
small
little
program
that
listens
on
HTTP
and
then
hoist
that
event
onto
a
nap
skew
right.
So
that's
that's
sort
of
like
the
ideas
that
it's
super
easy
to
deal
with.
Events
in
a
sort
of
abstract
way
like
that
and
we're
gonna
see
that
when
we
start
looking
at
agro
events,
my
understandings
are
go
immense,
doesn't
use
the
go
SDK,
but
I
think
they
want
to
get
there
so
they're
not
there.
A
Quite
yet,
but
that's
okay,
because
it's
a
binary
specification,
or
at
least
it's
a
wire
specification.
So
that
means
it's
totally
reasonable
to
have
multiple
implementations
and
then
something
else
that
Scott
pointed
me
to
is
that
there's
this
cool
tool
that
cloud
events,
conformance
testing,
which
is
actually
a
good
command-line
tool
for
sending
events?
And
so
I
think
this
might
be
a
fun
thing
to
you
know.
As
we
look
to
send
events
to
our
go
offense,
we
can
actually
see
some
level
of
interoperability
using
tools
like
this
and
I
think
did
I
do
I.
A
B
A
A
B
C
B
A
There
we
go
ok,
so
now,
I
have
the
cloud
events
tool,
ok,
yeah
I
at
some
point:
I
need
to
actually
learn:
Z
shell
or
figure
it
out.
Yeah,
ok,
so
we're
all
set
here
and
then
I
have
my
I
believe
that
I
still
have
my
my
kind
cluster
up
and
running:
ok
cool,
so
yeah,
so
we're
gonna
play
with
that.
So
this
actually
looks
like
a
super
useful
tool
when
working
with
cloud
events,
because
you
can
both
send
and
listen
on
these
things-
and
it
looks
like
there's
a
there's-
a
to-do
list
there.
A
If
you
all
want
to
get
involved.
Ok,
Lee's
asking
oh
yeah
before
I,
go
onto
our
go
events,
talk
about
sort
of
what
our
goal
is
and
sort
of
how
our
go.
Events
actually
fits
into
it.
That's
a
really
good!
That's
a
really
good
prompt!
That's
definitely
something
I
should
do
let's
see
and
then
the
other
thing
that
Scott
pointed
me
to
is
he
created
this
tool
called
Sakai,
which
looks
really
interesting,
and
this
is
something
that
that
you
can
run.
A
Maybe
we
can
figure
this
out
as,
like
you
know,
I
want
to
understand
sort
of
if
Argo
events
can
actually
sort
of
act
as
a
router
and
dispatch
to
a
running
pod
or
something
that
might
be
an
interesting
thing.
Also,
okay,
so
that's
what
cloud
events
is,
and
it's
one
of
those
things
where
it's
hard
to
do
a
demo
of
cloud
events,
because
it's
just
a
specification.
It's
like
do
me
a
demo
of
HTTP.
A
Well,
it's
hard
to
do
that
with
or
without
talking
and
having
a
browser
or
a
web
server
or
whatever,
and
so
so
I
think
it's.
It's
really
useful
for
us
to
look
at
this
in
the
context
of
argot
events,
and
so
that
does
beg
the
question
and
thanks
for
Lee
for
the
yeah
Scott
says:
please
help
with
his
stuff.
A
A
A
A
lot
of
uptake
in
the
machine
learning
community
around
that
cube
flow
uses
that
I
believe
so
that's
one
of
the
places
where
it
actually
showed
up,
but
then,
along
the
way,
Argo
really
turned
into
essentially
a
toolkit
for
doing
a
bunch
of
automation,
type
of
stuff,
and
so
what
we
see
here
is
that
we
have
our
go
workflows.
We
have
our
go
continuous
deployment
or
go
rollouts,
which
is
essentially
a
way
of
doing
rolling
out
and
enroll
Bax
and
then
events
is
is
another
part
of
the
Argo
project.
A
And
so
our
go
flux,
and
so
the
you
know
the
the
idea
is
to
bring
these
things
together
and
that's
happening
through
this
project
called
get
ops
engine
as
part
of
our
NGO,
and
so
you
may
look
at
this
and
see
that
hey
nothing's
happened
in
months,
but
it
turns
out
that
I
believe,
like
the
a
lot
of
I,
don't
know
that
I
think
there's
stuff
happening
on
a
branch
I'm,
not
sure,
because
this
stuff
is
happening.
I
know
there
is
work
ongoing
here,
but.
A
Yeah
I
don't
know,
I
was
talking
to
Alexis
who's.
The
the
CEO
of
weave
and
I
know
that
there
is
some
work
ongoing
here,
so
yeah,
so
we're
actually
seeing
sort
of
some
consolidation.
As
we
see
you
know
more
folks,
look
at
the
agro
tool,
kit
and
figure
out.
You
know
how
they
can
start
using
these
bits
and
pieces
for
different
things.
A
So
that
is
like
generally
like
the
work
flow
pipeline.
Stuff
is,
but
then
that
really
begs
the
question
of
like
well:
when
do
you
run
a
workflow
and
that's
where
the
event
stuff
comes?
In
it's
really
about
you
know,
hey,
there's,
some
sort
of
event
happens
someplace.
How
do
we
get
that
into
the
system
and
then
decide?
How
do
we
want
to
react
to
that
event?
And
so
that
is
the
the
purpose
of
the
Argo
events
project
and
vibe
hub
and
again
I
apologize.
A
If
I'm
messing
your
name
up,
you
said:
I
did
okay,
but
I.
Don't
believe
you
gave
me
a
bunch
of
links
to
start
with
in
terms
of
how
our
NGO
events
works,
and
so
we're
gonna
install
this
and
actually
make
it
work.
But
but
basically
it
sounds
like
the
the
key
things
that
we're
talking
about
here.
Is
that
there's
a
gateway
which
is
essentially
a
place
that
accepts
incoming
events
via
a
whole
bunch
of
different
mechanisms?
A
If
I'm
wrong
folks,
the
trigger
is
actually
causing
something
to
happen
in
a
system,
and
so
there
are,
let's
see
if
we're
looking
at
triggers
a
trigger,
is
the
resource
executed
by
sensor
once
the
event
dependencies
are
resolved,
and
so
that's
some
stuff
that
we're
gonna
actually
sort
of
start
digging
into
here,
and
so
it
looks
like
there's
a
whole
bunch
of
different
things
that
can
play
in
in
terms
of
event
sources
and
then
and
in
terms
of
triggers.
There's,
let's
see
a
trigger
source
is
this
source
of
a
trigger
resource?
A
It
can
either
be
I,
don't
know
so
we're
gonna
dig
into
this,
because
I
don't
fully
understand
everything
that's
going
on
here,
but
hopefully,
as
we
install
it
and
start
playing
with
it,
I'll
get
a
better
idea
of
how
how
to
get
going.
So,
let's
go
ahead
and
get
started.
We're
going
to
start
with
the
our
go
events,
installation
and,
let's
see
so
we're
going
to
we're
gonna,
create
an
Argo
of
ents
namespace.
A
So
Alexander
says
we're
releasing
final
release
before
merge.
This
2
PRS
is
that
for
the
the
get
ops
engine
Alex
Alexander
okay,
so
we
have
that
up
and
running,
and
then
they
have
sort
of
a
big
ol
installation,
yeah
Mille.
So
what
I'm
gonna
do
is
I'm
gonna,
yeah,
okay,
so
cool
so
like
stuff
is
happening
with
the
the
get
ops
engine.
It's
just
folks
are
working
to
bring
it
together.
Okay,
so
so
this
is
all
the
different
things.
Let
me.
A
That
goes
in
here:
I'm,
not
gonna,
go
through
all
of
this.
Let's
see
this,
so
this
looks
like
it's
really
built
to
run
inside
of
a
namespace
with
a
so
there's
an
Argo
events,
namespace
okay,
we
already
have
that
up
and
running.
So
one
of
the
things
that
we
can
do
is
I'm
going
to
do.
Cap
and
I
just
I
always
forget
the
I
know:
I
have
to
give
it
a
name,
deploy
I.
B
A
C
A
All
right
so
here
this
is
this-
is
why
I
love
cap
is
that
it
actually
shows
you
what
it's
about
to
do
so
we're
gonna,
install
a
custom
resource
definition
for
event,
sources,
gateways
and
sensors
and
then
inside
the
Argo
events,
namespace
we're
gonna,
have
a
service
account
of
deployment,
a
config
map
for
the
gateway
controller,
a
sensor
controller
and
then
we're
gonna
have
an
Argo
events,
roll
and
roll
binding.
What
I
like
about
this,
which
I
think
is
really
really
cool,
is
that
this
actually
has
a
mode
where
it
installs
into
a
single
namespace.
A
A
Doo-Doo-Doo-Doo,
it's
probably
downloading
images.
Actually,
a
pretty
fast
internet
here
at
home
I
got
like
symmetric
gigabit
up
and
down.
So
you
know
with
this
whole
work-from-home
thing
I've
been
sitting
pretty
but
I
know
not.
Everybody
is,
is
in
the
same
situation.
I.
Definitely
like
you
know
as
we
go
through
this
whole
quarantine
thing.
I've
definitely
had
a
bunch
of
you
know
emotional
ups
and
downs.
A
A
A
A
A
A
A
B
A
A
What
did
I
do
there?
Okay
cap,
so
this
is
actually
one
of
the
great
things
about
cap
help
is
that
you
can
actually
go
through
and
it
has
a
bunch
of
different
ways
to
actually
help.
You
say
you
know
which
namespace
or
you
only
want
it
to
work
in
annex
Ingle
namespace.
So
we
can
go
through
and
do
and
and
are
go.
A
A
A
Yeah
so
so
Kappas
is
pretty
cool
with
stuff
like
that,
which
is
one
of
them.
That's
one
of
the
reasons
I
really
like
cap
is
that
it
essentially
provides.
You
know
all
these
different
tools
for
you
to
essentially
ring-fence
the
yeah
mole
so
that
you
can
say
well.
I
only
want
to
allow
it
to
work
with
in
this
particular
namespace
that
type
of
thing
really
cool
stuff
there.
Okay,
so
we
are
where
we
act.
We're
here,
we've
done
that
we
should
have
the
thing
installed
and
then
the
next
thing
is
there's
essentially
a
tutorial
introduction.
A
A
B
A
I
think
it
sounds
good
with
the
new
mic.
Is
there
a
kubernetes
events
importer
into
Argo?
Oh,
that's
a
good
question
right.
Can
you
actually
use
kubernetes
events
to
be
able
to
trigger
stuff?
You
know
I,
don't
know
like
if
kubernetes
events
are
as
useful
as
they
should
be
there.
That's
actually
a
really
good
question.
A
All
right,
so,
okay,
so
we
have
the
web
hook
event
source.
Okay.
So
let
me
go
back
here
and
do
code
here.
We'll
take
a
look
at
this
okay,
so
we
have
event
sources,
oh
and
there's
all
sorts
of
stuff,
so
we're
gonna
do
the
web
hook,
one
which
is
probably
the
most
generic
one
and
then
there's
also
stuff
like
github.
A
All
right
I
need
to
dig
into
this
and
actually
see
how
these
things
work.
Okay,
so
this
one
so
we're
creating
an
event
source
of
type
web
hook,
gateway
can
run
multiple
HTTP
servers
simply
define
a
unique
port
to
start
a
new
HTTP
server,
and
so
where
HTTP,
port
12,000
and
then
the
example,
and
it's
a
post.
Okay
and
this
one
is
unsecure,
but
you
can
also
do
TLS
for
this
stuff
too.
Okay,
so,
let's
just
go
ahead
and
cap.
A
A
A
Okay,
so
that
thing
got
installed
now,
one
of
the
things
I
love
about
octant
is
that
we
can
actually
go
in
here
and
look
at
custom
resources.
And
here
we
have
the
event
source,
and
we
can
you
know
it's
an
easy
way
to
view
the
ammo
here
and
we
can
have
their
spec
and
then
there's
a
whole
bunch
of
metadata.
A
A
A
Well,
I'm
a
little
bit
confused
on
what's
going
on
here,
so
the
Gateway
is
a
webhook
working
with
this
event
source
and
then
it
okay.
So
this
is
a
essentially
an
operator
that
will
run
a
set
of
containers
for
that
gateway,
but
then
it
also
it'll
set
up
a
service
for
the
gateway
of
type
equals
load
balancer.
We
should
probably
know,
let's
see
what
happens,
and
then
it
specifies
a
subscriber.
So
that
means
that
when
stuff
comes
into
the
gate,
gateway
it'll
actually
write
it
out
to
this
particular
event.
A
So
it
is
the
gateway,
so
the
gateway
converts
things
into
external
event,
sources,
okay
and
then
can
dispatch
it,
but
I
could
probably
like
could
I?
Have
this
be
a
subscriber
to
Scott's
thing
that
actually
shows
the
events?
I
don't
know
we
can
try
that
later.
But
let's,
let's
do
the
okay
and
then,
if
I'm,
doing
like
a
github
one,
there's
a
github
gateway,
okay.
So
this
is
the
thing
that
actually
knows
how
to
convert
things
from
the
github
format
into
the
cloud
events:
format:
okay,
that's
cool
yeah!
A
One
of
the
interesting
things
is
that
it
looks
like
you
know.
Kay
native
also
has
a
whole
bunch
of
capabilities
for
essentially
converting
events
from
external
sources
into
cloud
events
and
and
I.
You
know.
Maybe
that
should
be
part
of
the
cloud
events
thing
is
actually
having
these
translators,
maybe
as
libraries
or
as
as
you
know,
because
just
to
use
those
sort
of
translators
as
widely
as
possible.
Sorry,
the
dogs
barking
okay,
so
we're
going
through
and
now
we're
doing,
gateways.
A
A
A
A
A
A
Just
thinking
about
all
the
different
sort
of
you
know,
things
that
show
up
in
the
the
container
spec,
but
in
services
and
stuff,
but
yeah
looks
workable,
okay
and
then,
and
then
the
last
one
we
want
to
add
on
is
the
the
let's
see
the
the
web
hooks
sensor.
Okay,
and
so
let's
look
at
the
sensor
for
web
hook,
and
so
this
says
we're
going
to
run
a
sensor.
So
this
will
also
go
ahead
and
run
stuff
and
then,
when
this
happens,
it'll
actually
create
our
go
workflow
when
this
happens.
A
A
A
Okay,
I
can
trigger
a
pod.
How
about
we
just
do
the
trigger
the
pod,
because
I
don't
want
to
you
know:
I,
don't
want
to
I
just
for
time.
Okay,
so
trigger
source
is
no
no!
No,
the
standard,
kubernetes
resources.
Okay,
so
here
I'm
configuring,
a
pod,
a
deploy
or
a
deployment.
Okay,
so
we're
gonna
modify
this.
A
A
A
A
A
B
A
B
A
A
A
B
A
A
A
A
A
A
B
A
A
C
A
B
A
A
D
D
B
B
A
A
A
D
A
A
A
A
A
Yeah,
so,
okay,
so
this
one's
running,
just
like
normal
okay,
so
this
didn't
hit
any
errors
and
we're
up
and
running
okay,
and
then,
when
I
look
at
pods,
we
can
see
that
these
things
are
succeeded.
So
yeah
is
there
something
that
I'll
actually
garbage
collect
and
reap
these
things?
Yes,
if
it's
a
job,
it's
in
the
job,
spec
it'll
actually
go
ahead
and
do
it.
But
this
is
a
pod,
not
a
job,
and
so
I
think.
A
One
of
the
things
that
might
be
interesting
here
with
agro
events
is,
and
maybe
is
it
looks
like
the
standard
resources
are
pod
and
deployment.
I
would
expect
that
job
would
be
one
of
these
things
here
too,
and
then
you
can
do
custom
resource.
Ok,
let's
do
some
more
stuff
here
and
actually
figure
stuff
out
here.
A
A
Okay,
previous
sections,
you
have
been
dealing
with
just
a
single
dependency,
but
in
many
cases
you
want
to
wait
for
multiple
events
to
occur
and
then
trigger
a
resource,
which
means
that
you
need
a
mechanism
to
determine
which
triggers
to
execute
based
on
a
set
of
different
event
dependencies.
Oh,
this
is
cool,
so
you
can
say
like
when
this
happens
and
that
happens
then
I
want
to
actually
go
through
and
do
this
thing
alright.
So
let's
go
through
and.
A
A
And
trigger
when
either
of
these
things
Oh
fascinating,
okay
and
then
this
one
is
an
and
okay
and
then
we're
going
to
trigger
our
workflow.
That
does
a
cow,
say:
okay,
so
I
I
think
we
can.
Probably
let's
see
so.
The
question
here
is
that
the
dependency
okay,
we
can
do
an
event
name
coming
from
the
gateway.
So,
let's
see
so.
Where
does
the
event
name
come
from.
A
A
Interesting,
okay,
so
the
event
I
would
expect
that
it
would
pull
something
from
like
cloud.
Events
like
can
I
actually
do
something
like
we're
part
of
my
sensor,
I,
actually
yeah.
You
can
do
multiple
event
configurations
within
a
single
event
source
but
I'm
wondering
as
part
of
my
sensor
can
I
actually
go
through
and
key
off
of
stuff.
That's
in
the
that's
in
the
the
cloud
events
stuff,
like
the
the
cloud
events
type
or
something
like
that.
A
That
would
be
something
that'd
be
interesting.
Okay,
so
we're
gonna
do
example:
two
here
we're
gonna
do
web
hook
web
hook,
okay
and
then
we're
gonna
do
web
hook
and
we're
gonna
do
and
then
switch
trigger.
It's
when,
when
is
the
switch?
That
term
is
whether
to
execute
the
trigger
or
not
all
of
Group,
one
so
I'm,
a
very
I'm
kind
of
confused,
so
so
I'm
there's
the
circle.
So
the
difference
between
the
circuit
and
the
switch
is
confusing
to
me
a
little
bit.
What
is
the
difference
between
those.
B
A
Here
we
can
okay
circuits
and
switch
okay
circuit
is
a
boolean
expression
to
create
a
circuit.
You
just
need
a
switch.
Is
the
conditional
execution
gate
for
a
trigger
I
want
to
trigger
an
agro
work
flow
if
the
sensor
receives
an
event
from
the
web
gateway,
but
another
workflow,
oh
I,
see
so
you
can
actually
specify
do
this
one
or
the
other
one
interesting
I
would
love
to
actually
see
as
part
of
the
documentation
sort
of
the
examples
of
sort
of
like
when
this
stuff
comes
into
play.
B
A
A
Okay,
so
we're
going
to
update
the
event
source,
we're
gonna
update
the
sensor
so
now,
okay,
so
let's
look
at
octant
and
see
what
we
have.
We
have
a
couple
of
completed
hello
worlds.
We
have
the
gateway
controller,
the
sensor
and
all
that
stuff.
Okay.
So
if
I
go
through
and
I
do
okay,
you
may
want
to
change
the
name
of
the
sensor
from
circuit
sensor
to
web
hook.
Sensor:
oh
okay,
gotcha
on
the
yeah,
because
I'm
mixing
and
matching
the
different
tutorials
here
so
I'm
definitely
going
off
script.
Here.
D
B
A
Circuits
and
switches-
and
here
we're
doing
sensor,
Oh
1:00
a.m.
oh,
ok,
so
that'll
update
the
webhook
sensor.
Ok
now,
if
I
do
a
curl
to
example,
that
should
not
have
run
anything
right
because
we've
only
run
one
and
not
the
other,
so
we
haven't
run
anything
yet.
But
now,
if
I
do
another
curl
to
example,
2.
A
A
A
B
A
A
Well,
this
is
a
lot
of
data
in
terms
of
what's
happening.
The
sensor
needs
to
change
the
trigger
from
workflow
to
pod.
That's
the
problem
kind
pod.
There
you
go
and
then
this
is
like
did
I
I
thought
I
did
that
I'm?
Sorry,
let
me
copy
and
paste
from
the
sensor
pod.
Okay.
This
is
the
trigger
that
I
want
here
right,
because
this
is
pods
version.
V1.
A
D
B
A
Let's
actually
try
and
hit
this
again
and
see
what
happens?
Okay,
so
we're
gonna
hit
this
to
example.
Example:
two
pods
yep
and
then
it
launched
it
I
think
nine
seconds
yeah.
So
there's
the
one
and
it
succeeded
it
already
ran,
and
so
we
can
do
logs
and
that
worked
okay,
but
the
one
problem
that
we
had
is
that
we're
doing
inspect
our
containers
are
zero.
I
guess
like
I'm,
not
sure
where
the
arcs
come
from
there.
So
that's
actually
pretty
cool
okay,
so
that's
really
cool.
A
So
you
can
do
some
of
these
like
workflows
where
it's
like.
First
I
want.
You
know,
I
want
these
two
things
to
happen
and
when
that
happens,
then
I
actually
trigger
something.
And
so
it's
it's
interesting
because
there's
like
there's
definitely
state
going
on
in
the
the
sensor.
That
does
that
state
could
actually,
you
know
persisted
into
the
CR
D
I
expect
it
probably
does
and
then,
similarly,
we
have.
A
You
can
filter
based
on
data
context
and
time.
So
that's
cool,
yeah
I
definitely
feel
like
some
visualizations
would
definitely
help
here.
A
policy
for
a
trigger
determines
whether
the
trigger
results
in
success
or
failure,
HTTP
yeah,
so
I
think
you
know.
One
of
the
things
that
I
think
would
be
a
lot
of
fun
to
see
is
maybe
an
extension
for
octant
for
actually
being
able
to
view.
You
know
the
relationship
between
all
these
different
things.
A
What
the
current
status
is
that
would
be
super
cool
here
and
it's
one
of
the
things
that
I'm
sure
the
octant
team
would
love
to
hear.
You
know
if
you're
able
to
do
that
and
how
that
works,
because
you
know
you
could
have
like
another
thing
here
that
could
be
like
you
know:
Argo
events
and
Argo
workflows
and
actually
have
that
visibility
showing
up
there
that'd
be
super
cool,
okay,
so
I'm
gonna
call
it
here.
I
think
I
got
a
good
sense
of
how
this
stuff
works.
Oh
one,
more
thing,
I
wanted
to
try.
A
B
A
B
A
A
B
A
A
B
B
A
A
Well,
that's
pretty
cool
all
right,
so
I
think
I'm
out
of
time.
Here.
Let
me
sort
of
sum
up
some
of
my
thinking
here.
Cuz
I
think
you
know
I'm
still
I'm
a
total
noob
when
it
comes
to
sort
of
this
cloud
event.
Space
I
think
I
really
like
I,
like
the
idea
that
we
can
start
normalizing
events
bring
them
in
and
actually
start
dealing
with
them
coming
from
different
directions.
A
Right
and
I
think
that
example
of
like
well
I
want
to
wait
until
X
happens
and
Y
happens,
and
one
of
them
could
be
coming
from
Mineo.
The
other
one
could
be
coming
from
github,
something
like
that,
but
I
can
actually
view
you
know
deal
with
those
things
in
sort
of
a
universal
way.
I
think
that's
super
cool
agro
events.
A
I
think
you
know
it
took
me
a
little
while
to
understand
the
relationship
between
the
gateways,
the
sensors
and
the
event
sources,
and
so
that's
definitely
something
that
I'm
still
wrapping
my
head
around
and
actually
understanding
how
those
things
you
know
relate
to
cloud
events
right
like
the
event
source
thing.
Like
is
really
you
know
it
comes
down
to.
A
Let's
see
if
I
find
the
event
source
the
event
source
really
comes
down
to
essentially
the
name
that
you
give
for
a
path
that
you're
listening
to
I
think
it
would
also
be
interesting
as
part
of
the
sensor
to
be
able
to
you
know.
Essentially,
let's
see
as
we
look
not
that
one
I
was
looking
at
since
row,
one
you
know
as
I
look
at
the
dependencies
can
I
actually
have
dependencies
based
on
other
parts
of
the
cloud
event
schema.
A
Oh
well,
late,
I'm,
sorry,
you
missed
it
so,
like
you
know
so,
essentially
can
we
actually
make
this
be
sort
of
cloud
events
native
in
more
ways
right?
That's
than
that
I'm
thinking
about
and
I
would
love
and
I
think
you
know
we
didn't
have
time
to
explore
it,
but
are
there
ways
to
you
know,
have
our
go?
A
Events
actually
play
a
part
of
sort
of
a
cloud
events,
ecosystem
that
involves
things
like
Kay,
native
and
stuff
also,
and
so
one
of
the
things
that
I
think
might
be
interesting
and
it
didn't
I
didn't
see
this
maybe
or
HTTP
trigger.
So
what
does
this
do
so?
This
says?
Oh,
we
should've
tried
this.
Oh.
A
So
this
was
the
cue
bliss.
As
the
example
here
is.
Can
we
go
through
okay?
So
we
have
a
HTTP
okay.
So
this
is
cool.
I
know
we
ran
out
of
time
getting
to
this,
but
but
if
I
look
at
sensor
o1
here
here,
the
trigger
is
like:
oh
just
call
this.
Okay,
so
that's
pretty
cool
and
you
can
actually
sort
of
do
the
payload
stuff.
Okay,
so
that's
cool,
so
I
think
yeah.
It
would
be
really
fun
to
actually
see
like
hey.
A
Can
argo
events
actually
add
some
value
in
terms
of
providing
some
of
this
filtering
some
of
the
dependency
stuff
when
combined
with
something
like
k
native
and
actually
bring
these
things
together?
Oh
sorry,
I
wasn't
sharing
my
screen.
I'll
show
you
what
I
was
what
I
was
looking
at
sorry,
so
yeah
there's
a
HTTP
trigger
here
that
we
didn't
get
a
chance
to
dig
into,
but
then
that
actually
translates
to
like
sensor
o1
here
and
to
you.
A
There
is
a
sort
of
trigger
of
type
HTTP
trigger,
which
is
pretty
cool,
so
yeah
and
so
I
would
love
to
see
us.
You
know
more
examples
of
a
sort
of
ecosystem
of
things
that
can
pass
stuff
around
using
cloud
events.
Another
example
would
be
you
know:
how
does
this
interact
with
something
like
dapper
that
we
looked
at
a
little
while
ago
and
I
don't
know
if
dapper
supports
cloud
events?
A
Does
it
it
supports
I,
don't
think
it
does
not
yet
at
least
oh
wait
here
we
go
boost
interoperate
with
cloud
events,
1.0
and
zhirov
as
your
event
grid.
So
I
don't
know
about
dapper,
but
dapper
has
a
bunch
of
integrations
for
being
able
to
say
run
this
thing
when
you
like
receive
something
from
this
queue.
I
think
you
know,
I
would
love
to.
Actually,
you
know
look
at
more
examples
of
how
cloud
events
become
sort
of
this
interoperability,
language
between
different
systems.
I.
A
Think
that's
really
interesting,
so
alright,
so
that
is
it
for
me
for
today.
Thank
you,
everybody
for
joining
and
a
stay
safe,
stay
healthy,
stay
inside.
That's
what
I'm
doing
I
think
this
is.
This
is
rough
for
all
of
us.
Oh
and
here's,
my
kid!
Oh,
this
is
Theo
he's
joining
me.
Auntie
GI,
k
and
and
y'all
have
a
good
weekend,
and
we
will
see
you
sometime
later.
Thank
you.