►
From YouTube: Welcome + Opening Remarks: The State of the GitOps Working Group - Christian Hernandez & Scott Rigby
Description
Don’t miss out! Join us at our upcoming event: KubeCon + CloudNativeCon Europe 2023 in Amsterdam, The Netherlands from April 17-21. Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.
Welcome + Opening Remarks: The State of the GitOps Working Group - Christian Hernandez, Red Hat & Scott Rigby, Weaveworks
B
A
In
person,
second,
one
North
America,
so
the
first
one
was
in
LA,
so
I'm
glad
to
see
this
grow.
It's
always
it's
great
to
see
all
everyone's
faces
in
3D
as
as
Scott
says,
so
we're
gonna
kick
things
off.
Just
like
a
few
logistical
area,
things
the
the
breakout
rooms
for
the
first,
the
second
track
right.
So
there's
two
tracks.
A
If
you
are
either
speaking
or
want
to
wanting
to
attend
a
track
in
the
second
track,
it's
actually
down
the
hall
in
251
right
there'll,
be
someone
there
in
the
front
in
rooms,
ABC
right,
so
251
ABC,
it's
literally
down
the
hall.
A
few
steps
down
that
way,
so
you
don't
have
to
travel
far
restrooms
I,
don't
know
where
they
are
I,
believe
they're
back
there.
Yes,.
A
Yes,
so
they're
they're
back
there
and
you
know
coffee
Refreshments,
all
of
that
is
just
outside.
If
you
didn't
get
a
drink
ticket
or
a
food
ticket,
the
lovely
ladies
outside
will
have
that
for
you
they're
the
ones
that
give
you
the
wristband.
So
if
you
didn't
get
one
make
sure
you
grab
that
so
that
way
you
can
get
lunch
so
I
just
I
believe
with
that
we
can
just
kick
it
off
right,
we'll
just.
B
Yeah,
absolutely
let
me
advance
these
slides,
so
so.
I'm
Scott,
Rigby
I
work
at
weaveworks
as
a
developer,
experience,
engineer,
I
and
one
of
the
co-chairs
of
the
good
Ops
working
group
and
amongst
a
a
good
group
of
maintainers
from
across
multiple
organizations
for
open,
getups
and
I'm.
Also
I
do
other
things
in
the
CNC
ecosystem,
I,
co-maintain,
the
flux
community
and
I
co-maintain,
the
helm
project
and
other
a
few
other
things.
So
that's
me
yes,.
A
B
B
Okay,
so
about
the
get
Ops
working
group
and
open
get
Ops.
A
bunch
of
you
may
know,
but
I'll
give
just
a
brief
intro
for
those
that
don't
to
get
up
to
working
group
is
a
working
group,
a
cncf
working
group
under
the
app
delivery,
technical
Advisory,
Group
we've
been
doing
this
for
some
years
now
and
it's
it's
there's
a
really
good
gang
of
folks
actually
raise
your
hand
if
you're,
if
you've
joined,
get
UPS
working
group
meetings
or
a
part
of
the
getups
working
group
yeah.
B
B
The
working
group
has
guided
the
open,
getups
project
and
still
guides
the
open,
get
Ops
project,
which
is
a
cncf
Sandbox
project,
I.
Think
as
as
most
people
know,
there's
sandbox
incubating
and
graduating
projects
within
cncf,
open,
get
op
says
that
and
the
open
gave
us
get.
Ops
project
is
meant
to
be
the
place
where
we,
where
we
hold
all
persisting
things
that
relate
to
git
Ops
documents,
any
certifications
that
are
forthcoming
programs
and
events
and
stuff
go
through
the
open,
get
Ops
project.
B
Yeah
so
I
actually
just
said
some
of
this,
but
but
yeah
so
so
so
right
we're
the
main
premise
of
the
get
up
of
the
working
group
is
to
clearly
Define
a
vendor
new,
neutral,
principle-led
definition
of
get
Ops
and
we'll
go
over
the
get
Ops
principles
in
just
a
bit
and,
as
I
said
about
the
sandbox
project.
Yes,
it
holds
those
things
it
also.
B
The
the
documents
right
now
contain
the
cncf
sorry
contain
the
principles
and
the
glossary
specifically
for
get
Ops,
also
giving
a
shout
out
to
the
the
general
cncf
glossary
that
is
at
glossary.cncf.io
for
folks
that
are
interested
in
the
definition
of
get
Ops
there.
We
are
still
working
on
it.
We
have
it
was.
It
was
a
relatively
recent
invite.
B
But
the
point
of
that
is
it's
supposed
to
be
readable
by
someone
who
has
really
no
domain
knowledge
about
this
whatsoever.
So
let's
say
someone
in
upper
level
management
of
a
company
that
is
a
non
a
person,
that's
not
really
in
the
weeds
with
the
technical
stuff,
so
yeah,
if
you're
interested
in
that,
please
let
us
know
which.
B
A
The
one
that
helped
out
with
mapping
the
get
Ops
principles
right,
which
we
hope
most
of
you,
are
familiar
with
to
to
practices
in
Tech,
so
so
basically
modern
mapping
those
practices
to
the
the
get
Ops
principles
so
that
try
to
resonate
it
more
with
with
you
folks
out
there.
So
I
guess:
I'll,
kick
it
off
right.
So
the
the
first
principle
is
that
it's
declarative,
I'm,
actually
just
gonna,
do
what
I
don't
like
to
do
and
read
off
the
slide.
But
it's
a
system
managed
by
getups,
must
have
its
desired.
A
State
expressed
declaratively
right,
and
so
this
kind
of
matches
things
like
configuration
as
code.
We
have
infrastructure
as
code.
You
know
this
is
like
devops
devsecops
right.
Some
of
the
this
is
kind
of
like
the
Cornerstone
of
these
practices
and
also
get
Ops
right.
Get
Ops
also
uses
that
declarative
model
for
for
managing
a
system.
So
yeah.
B
Basically,
Soup
To,
Nuts
I
is
the
ideal
thing
where
you've
got.
You
know
like
not
only
your
infrastructure,
but
your
networking
and
your
policies
and
then
also
all
the
applications
that
run
on
that
pretty
much
anything.
That's
required
to
to
reproduce
a
system,
your
system,
exactly
the
way
you
want
it
or
exactly
the
way
it
was
if
there
was
say
a
disaster
or
just
just
for
replication
purposes,
multi-cloud
Etc
yeah.
So
then
do
you
want
to
keep
yeah
yeah
so.
A
The
second
one
is
version
and
immutable
right,
so
the
desired
again,
the
desired
state
is
stored
in
a
way
that
enforces
immutability,
versioning
and
retains
a
complete
version.
History,
and
then
I
always
like
to
point
out
that
notice,
we
didn't
say
get
even
though
it's
called
get
Ops
right.
So
it's
like
we
try
to
make
these
as
generic
as
possible.
A
A
B
A
B
Yes,
and
and
also
notice
that
this
is
just
a
screenshot
of
our
website
with
these
little
practices
mapped
the
stuck
on
under
each
principle.
So
please
do
go
to
the
website
for
anyone
who
hasn't
already,
you
know
gone
into
the
principles
and
the
glossary,
because
these
little
yellow
notes
like
say,
stored
and
and
so
on.
They
link
to
glossary.
A
Cool,
so
the
next
one
is,
it's
pulled
automatically,
so
software
agents
automatically
pull
the
desired
State
declarations
from
The
Source
I
always
like
to
add
here
the
the
word
pulled
kind
of
trips
people
up
a
little
bit.
We
don't
we
don't
mean
like
as
in
pull
and
push
as
in
like
the
configurations.
Right
like
we
mean,
as
in
your
declarations,
right,
the
decorations
are
pulled
from
a
source,
how
you
apply
them,
whether
it's
in
a
push
or
pull
method.
A
That's
we
don't
have
a
strong
opinion
on
that,
but
we
do
have
a
strong
opinion
on
how
you
get
those
decorations
and
those
declarations
are
pulled
really,
and
it's
really
to
differentiate
them
from
like
web
hooks
right
so
from
you
know,
you
know
classic
CI,
CD
sort
of
web
hooks
kind
of
you
know,
I
hit
a
web
hook
and
then
something
happens.
No
like
this.
B
Yeah-
and
this
is
a
super
important
Point
here,
because
you
know
like
usually
when
you're
going
through
the
principles
with
folks
that
are
already
really
involved
in
technical
practices
and
and
devops
practices,
and
so
on,
we
start
describing
that
everything
must
be
declarative.
They're,
like
okay
infrastructure,
is
code,
cool,
cool
I'm
doing
that
right.
It's
version,
immutable.
Okay,
we've
got
our
stuff
in
git
cool,
and
you
know
pretty
pretty
much
after
that.
B
This
is
where
it
starts
to
diverge,
I,
think
as
a
community,
because
or
not
as
a
community
but
yeah
like
with
various
practices
and
where
traditional
cicd
really
the
point.
The
point
of
this
that
Christian
was
mentioning
I.
Think.
Is
that,
like
event,
driven
things
are
great
things
happen
faster?
It's
awesome
with
get
get
UPS.
You
can
still
have
event.
A
Yeah
and
and
like
like
Scott,
said
they
kind
of
like
build
on
each
other
because
then
like
it
leads
to
this
next
principle,
right
and
and
I
like
to
actually
mention
the
practice
before
the
actual.
So
this
is
the
one
that
kind
of
like
differentiates
a
little
bit
like
I.
Guess
our
Crown
Jewel
I
would
imagine
what
you
know.
A
Is
it
continuously
going?
Is
continuously
monitoring
your
system?
It's
you
know
it's
it's
also.
The
aspect
of
observability
as
well
is
like
do
I
need
to
do
something
to
the
system.
I
think
that
also
is
an
aspect
of
the
continuously
reconciled.
I'm
getting
maybe
a
little
bit
too
specific
there,
but
just
generically
speaking
system
managed
by
get
Ops
is
continuously
reconciled.
Yeah.
B
I
think
that's
right
where
we
say,
like
literally
the
software
agent's
attempt
to
apply
the
desired
state.
That
attempt
can
be
most
of
the
time
you
you
want
it
to
be
a
closed
loop,
it's
not
in
the
principles
because
that's
not
required.
You
can
make
a
Cron
job
that
just
does
Cube
cuddle
apply.
You
know,
while
true
do
right,
sure
sure
it
wouldn't
be
a
very
intelligent
one.
You
know
it
wouldn't
really
be
able
to
respond
to
any
feedback
or
or
previous
attempts,
but
but
you
could
do
that.
B
So,
however,
during
that
attempt,
various
actions
are
made
by
those
operations.
On
behalf
of
of
of
you
to
like,
say,
alert
on
Divergence
to
you
know,
do
you
think
this
is
very
similar
to
kubernetes
core,
let's
say
back
off
options,
and
so
on
there
there
are
different
like
ways
to
make
your
your
agents
more
intelligent
and
the
main
tools
that
are
out
there
to
do.
This
do
do
that.
B
A
Cool
yeah
and
well
I
think
that's
I
think
he
put
that
out
beautifully,
yes,
cool,
yes,
announcements.
These
are
awesome.
These
are
cool
cool
things
are
happening
right.
So
first
I'd
like
to
say
we
have
new
maintainers
right.
This
is
actually
it
was
a
was
a
process,
but
we
actually
got
a
lot
of
people
who
have
been
attending
regularly
and
helping
out
with
the
project.
So
we
decided
to
expand
the
maintainer
list
to
include
more
maintainers.
A
So
sorry,
if
I
butcher,
some
of
these
names,
Jamie
majera
who's
was
actually
here
at
kubecon,
but
he's
actually
doing
openshift
Commons
he's
from
the
University
of
Michigan
he's
a
new
maintainer
Nicholas
met.
He
met,
I
think
someone
who's
German
couldn't
tell
me
how
to
pronounce
that
who,
who
is
actually
I
believe,
runs
his
own
consulting
firm,
so
he's
he
maintains
actually
a
lot
of
the
website.
Stuff
he's
doing
a
lot
of
automation
with
the
website,
so
he's
been
helping
out
a
lot.
A
Robert
strand
who's
been
involves
since
the
beginning,
so
it
was
only
natural
for
him
to
become
a
become
a
maintainer
and
William
as
well
as
we
mentioned.
William
caban
from
from
Red
Hat
he's
also
been
contributing
a
lot,
especially
around
like
sustainability
and
stuff.
Like
we'll
talk
about
in
a
little
bit
so
and
then
we
have.
B
B
Okay,
okay,
yes,
so
the
governance
for
open
git
Ops
has
changed
from
from
there
being
a
top
level
maintainer
status
for
for
co-chairs
for
the
get
Ops
working
group
to
the
maintainers
of
of
open
get
Ops.
So
that's
just
a
good
new
thing.
If,
if
you
don't
know
anything
about
that,
then
you'll
be
like
cool.
But
if
you
do
it's
kind
of
nice
because
it
helps
things
move
a
little
bit
more
smoothly.
Procedurally,
so
yeah.
A
So,
essentially,
the
the
power
it
now
resides
with
the
maintainers
now,
whereas
before
a
lot
of
the
power
was
with
the
co-chairs,
I
guess
for
bootstrapping
purposes,
but
now
that
the
governance
has
been
updated
approved
that
PR
was
approved.
It
was
awesome
now
now
the
the
top
level
is
the
maintainers,
so
yeah.
B
A
Yes,
yes
and
there's
a
session
about
that,
and
also
like
AIML
right
so
like
how
does
get
Ops
fit
into
like
AIML
workloads
right
I
believe
that
they
they
they
there's
been
a
lot
of
work
already
done
with
that.
But
now,
like
the
the
subgroup,
has
started
for
people
who
are
interested
in
the
in
that
as
well.
A
Also
subgroups,
there's
I
guess
by
de
facto
my
subgroup
right
for
for
the
events
for
for
things
like
this
right.
We're
also
talking
about
like
other
kind
of
more
open,
Forum
type
of
events
right
where
we've
been
kind
of
toying
with
right.
So
that
way,
maybe
we'll
have
some
of
these
meetings,
maybe
more
often
and
more
of
a
bird
of
a
Feather
type
of
of
events,
smaller
more
more
frequent,
but
anyway,.
B
B
B
However,
there's
still
a
lot
of
marketing
out
there
that
you
know
that
will
say
you
know
we're
doing
something
on
githubs,
but
it
really
doesn't
adhere
to
the
github's
principles
at
all.
So
this
fact
checking
group
is
really
just
a
helper
group
not
like
to
call
people
out
like
policing
them,
but
just
a
helper
group
to
say:
oh
you
know
you
all
may
not
have
known
that
that
that
these
principles
were
released
by
cncf
and
Linux
foundation
and
agreed
upon
by
all
these
other
people
Etc
and
please
join
us.
B
A
So
we
there
there's
been
some
events
centered
around
get
Ops
right,
one
of
them
being
argocon.
Argocon
2022
happened.
It
happened
in
in
the
Mountain
View
area
in
the
computer
science
museum
there.
So
if
you've
ever
been
in
the
Bay
Area,
you
know
where
that
is,
and
it
was
the
second
articlecon
for
the
first
time
in
person.
So
that
was
that
was
pretty
cool.
B
Yep
yeah
get
upstairs
focusing
on
focusing
on
flux
happened.
It
was
really
great
yeah
check
out
the
doc
or
check
out
the
videos
from
both
of
those
events
for
sure
and
I'm
gonna.
Try
to
speed
us
up
a
little
bit
because
we
just
hit
the
9
30
Mark,
but
we
still
have
a
few
more
minutes
before
the
next
talk.
So.
A
A
Right
so
we're
always
looking
for
sponsors.
They
help
out
a
lot
they
help
out
to
put
in
this
event,
so
either
ask
me
or
Scott,
if
you're
interested
in
sponsoring
the
next
event,
so
cool.
So
next
one.
So
what's
next
right,
looking
at
the
Horizon
right,
as
Scott
said
fact,
checking
right.
We
need
kind
of
folks
to
help
out
not
not
only
like
do
actually
doing
the
fact
checking
but
like
to
develop
a
I,
don't
know
code
of
conduct
or
I'm
trying
to
find
the
right
phrasing.
A
If
there's
there
may
be
a
lot
of
issues,
but
not
a
lot
of
PRS
and
try
to
like
fix,
fix
that
ratio
a
little
bit
so
things
like
you
know
best
practices,
white
papers,
green
papers,
we
even
talked
about
blogs
right.
All
of
that
is
stuff
that
we
invite
all
of
you
to
contribute
to
as
well
right.
So
don't
don't
feel
that
you
know
you
have
to
be
a
maintainer
or
always
presence
right
like
we
can
do
things
asynchronously,
that's
what
gets
for
so.
B
That's
right
and
the
website's
just
it's
just
pull
request
based,
so
we
can
discuss
it
in
slack.
We
can
discuss
it
on
the
meetings
open,
an
issue
or
a
discussion
topic
if
you
have
them
and
want
to
do
that,
otherwise,
just
set
pen
to
paper,
you
know
yes
yeah
collaborate
or
do
it
on
do
it
solo,
also
under
the
the
con.
B
The
certification
side
I
just
wanted
to
mention
that
we
are
working
with
the
the
CD
Foundation
to
basically
Linux,
basically
Linux
Foundation
is
is
offering
a
or
creating
a
course
on
gitups,
and
so
the
get
UPS
working
group
and
open
getups
project
from
cncf
is
working
with
CD
Foundation.
To
do
that.
So,
if
you're
interested
in
finding
out
more
about
that,
I
mean
we'll
be
announcing
it
as
it's
coming
closer,
but
if
for
for
anyone
interested
in
getting
involved
with
that,
even
potentially,
let
us
know
cool.
A
And
then,
with
that
I
believe
it's
it
yeah
the
more
get
involved,
the
more
the
merrier
yeah.
You
know
where
to
find
this
right.
We
can
join
a
meeting
join
the
subgroup
you
know
get
involved
with
us,
you
can,
you
know,
go
to
the
website
is
probably
the
best
place
to
start,
because
that's
where
all
our
links
are
there
yeah.