►
Description
OpenShift Commons Gathering 2022
Speaker: Dan Garfield (Codefresh)
Full Agenda:
https://commons.openshift.org/gatherings/OpenShift_Commons_Gathering_on_GitOps.html
Learn more at OpenShift Commons https://commons.openshift.org
A
If
you,
if
you
don't
know
me,
I
am
the
co-founder
and
chief
open
source
officer
of
code
fresh
and
we're
a
company
that
focuses
really
on
git
ops
and
development
tools
that
help
people
to
deliver
deployment
tools
to
help
people
deliver
software
more
quickly
and
more
reliably,
and
we've
been
doing
this
for
five
or
six
years
now,
and
it's
really
grown
in
a
very
dramatic
way
and
been
very
exciting
and
I've
been
very
lucky
to
participate
in
the
founding
of
the
of
the
get
ops
working
group
and
open
get
ups.
A
So
we're
gonna,
I'm
gonna,
give
kind
of
a
community
update
today
and
we'll
cover
like
kind
of
a
smattering
of
topics
and
it's
somewhat
informal
and
and
so
feel
free
to
to
blow
up
the
chat
I'll,
be
watching
the
chat
and
looking
for
for
comments
and
questions
and
happy
to
answer
any
of
those.
So
cornelia
gave
a
really
good
kind
of
background
about.
A
These
ideas
existed,
they've
been
around
and
you
know
we'll
talk
a
little
bit
about
some
of
the
ideas
that
came
out
of
the
gals
principles
and
I
know
christian
covered
some
of
those
as
well
and
if
you're
not
familiar
with
them,
we'll
do
like
maybe
just
a
quick
rehash
if
we
have
time,
but
I
think
that
these
ideas
of
git
ops
are
really
a
collection
of
best
practices
and
principles
that
people
have
been
doing
for
a
long
time,
but
they
have
been
very
difficult
to
implement,
and
so
what
we
found
is
that
there
were
a
few
kind
of
savants
who
had
implemented
this
kind
of
stuff
you
know
years
ago,
but
that
largely
these
things
were
really
inaccessible
to
to
everybody
else,
and
so
what
we're
really
working
on
is
doing
what
henry
ford
did
for
the
car.
A
He
didn't
invent
the
car,
but
he
he
got
it
in
everybody's
house
and
that's
what
we're
trying
to
do
with
with
get
ops
and
open
get
ops
is
really
make
this
stuff
super
accessible.
Kubernetes
is
a
key
unlock.
You
know,
like
cornelia
mentioned,
you
don't
have
to
be
doing
kubernetes
to
do
to
implement
the
git
ops
principles,
but
kubernetes
just
does
make
it
damn
easy,
that's
for
sure.
A
So
so
there
are,
you
can
implement
it
with
within
in
other
systems,
but
but
kubernetes
definitely
makes
it
easier.
So
yeah
excited
excited
to
share
some
of
this
stuff.
Now,
with
the
get
ops
working
group
which
was
founded
under
the
cncf,
it's
a
cncf
sandbox
project.
This
was
really
what
we
needed
to
bootstrap.
A
What
the
current
project
is,
which
is
open,
get
offs,
and
so,
with
the
get
ups
working
group,
this
is
founded
by
code.
Fresh
weave
works,
red
hat,
github
azure.
I
feel
like
I'm
forgetting
somebody
which
is
like
never
good
you.
Never.
You
ever
want
to
forget
somebody
when
you're
mentioning
all
the
names
yeah.
A
Oh
amazon,
I
was
freaking
amazon,
so
we
basically
got
together
and
we
said:
okay,
let's
formalize
these
principles,
because
a
lot
of
people
are
talking
about
git
ups,
but
there
is
a
big
problem
in
the
industry
of
not
understanding
what
git
ops
is
and
not
having
kind
of
a
consensus
around
what
git
kidops
is
and
one
of
my
colleagues
costas
capilonus.
If
you
haven't
followed
him,
he's
great,
he
did.
He
gave
a
talk
where
he
said
he
asked
at
the
beginning.
A
How
many
of
you
are
doing
get
ups
today
and
it
was
like
you
know,
80
of
the
audience
was
like
oh
yeah,
we're
doing
we're
doing
get
ups,
we're
fully
implemented
on
get
ups,
we're
all
in
on
the
get
ops,
train,
no
worries,
and
then
he
gave
his
talk
on
the
get
ops
principles
and
then
afterwards
he
said.
Okay.
How
many
of
you
now
you've
seen
the
principles
are
doing
get
ops
today
and
suddenly,
instead
of
80
of
the
audience,
it
was
like.
Four
percent
of
the
audience
was
like
oh
yeah,
we're
doing
that.
A
It
turns
out
that
there's
just
a
lot
of
misconceptions
about
what
git
ops
is
and
people
kind
of
hear,
and
they
say:
oh
get,
ops
means
operations
by
pull
request
and
that's
part
of
it
and
they're.
Like
yeah,
I
mean
you
know:
we've
got
terraform,
we've
tossed
it
in
a
git
repo,
and
you
know
if
we
want
to
make
a
change,
we
make
a
change
and
we
committed
to
get-
and
you
know
that's
deployed
then
automatically
and
so
we're
doing
get
ups
well,
they're
missing.
A
Actually
some
pretty
critical
key
components
like
like
was
mentioned
earlier,
so
that
education
and
we'll
talk
about
those
in
a
minute,
but
that
education
is
just
really
really
critical
and
we
need
help
to
do
it.
I
mean
within
the
get
ops
within
open
get-ups.
You
know,
there's
a
handful
of
maintainers.
There
are
a
number
of
people
participating
in
the
community,
but
we
need
people
to
continue
to
spread
and
evangelize
these
messages
and
help
build
it
out
and
help
get
it
out
there.
A
So
one
of
the
things
that
I
want
to
ask
from
the
community
side
is
definitely
for
your
help
for
anybody
who's
listening.
If
you
want
to
get
involved
in
get
ops
I'll
I'll
share
a
little
bit
more
about
how
to
do
that,
but
we
definitely
need
that
help.
So
that's
kind
of
the
quickest
introduction
to
to
like
our
mission
is
really
to
educate,
and
it's
also
provide
principles
and
standardization
that
allows
companies
to
innovate.
A
So,
for
example,
with
these
cloud
operators,
we
want
the
cloud
operators
to
be
able
to
look
at
these
principles
and
say:
hey:
are
we
enabling
people
to
do
get
ops
with
our
offering
if,
if
you're
at
a
company
and
I've
talked
to
a
number
of
companies
who
reach
out
to
me
around
this
kind
of
stuff
to
say,
hey,
you
know,
can
you
can
you
present
on
on
the
github's
principles
and
share
this
stuff,
because
we
really
want
to
be
get
ops
oriented?
But
we
maybe
don't
fully
understand
what
that
means.
A
That's
something
that
we
can
do
so
from
an
open,
get
ups
priority
standpoint.
We
really
want
to
do
a
lot
of
education.
That's
a
super
high
priority,
so
we
run
we
run,
get
ops
con
as
well
as
some
other
get
ops
oriented
events.
This
event
that
christian
has
put
on
red,
hat's
a
member
organization
and
so
they're
they're
putting
this
on
under
the
auspices
of
of
the
of
open
get-outs,
which
is
fantastic.
A
We
did
last
year
we
had
a
cncf
ambassador
from
brazil,
reach
out
and
say
hey.
We
want
to
do
a
get-up
stay
and
we
worked
with
them
to
to
help
make
sure
that
they
had.
You
know
all
those
principles
present
and
making
sure
that
all
that
education
was
happening.
So
education
is
a
big
thing.
A
They
were
working
on
and
there
are
a
number
of
different
initiatives
within
the
project
to
help
enable
education
beyond
the
events
now
two
of
those
that
I'll
mention
which
are
kind
of
currently
going
through
the
the
committee
visual.
You
know,
process
stuff,
but
on
the
open,
get-offs
website,
we
have
the
principles
and
sometimes
people
look
at
those
principles
and
they're
like
yeah.
This
seems
really
simple,
they're
designed
to
be
simple
and
they
may
miss.
You
know
some
of
the
nuance.
That's
that's
reflected
in
those
principles,
but
basically
what
those?
A
What
what
were
you?
What
we'd
like
to
do?
In
addition
to
those
principles?
And,
of
course,
we
list
events
on
the
website
and
stuff
like
that.
We
also
like
to
get
out
a
section
so
that
people
can
share
and
look
at
patterns
and
reference
architectures.
So
if
you're
using
aws
and
vault-
and
you
know
my
sequel-
we
want
you
to
be
able
to
look
and
find
reference
architectures
that
people
have
contributed
to
say.
Here's
how
we
set
up
get
ops
here
are
some
of
the
considerations
we
have
for
stateful
applications.
A
Here's
some
of
the
considerations
we
had
for
using
aws
here's
some
of
the
considerations
we
had
for
using
bitbucket
or
whatever,
so
that
it's
easier
to
adopt
this
stuff,
because
again
we're
trying
to
do
what
you
know.
Henry
ford
did
for
cars,
we're
trying
to
make
them
really
accessible,
we're
trying
to
make
git
ops
really
accessible.
So
that's
that's
one
of
the
initiatives
and
then
another
is.
A
There
are
a
number
of
open
questions
within
the
get
ops
principles
that
the
br
that
or
I
should
say,
the
getups
principles,
don't
cover
so,
for
example,
secret
management
is
something
that
we
talk
a
little
bit
about
in
the
principles
or
we
reference,
but
how
to
do
that,
how
to
version
it,
how
to
keep
it
secure.
Those
things
are
actually
really
important
and
there
are
a
lot
of
nuances
to
how
you
handle
secrets
in
both
a
secure
way
and
like
a
scalable,
git
ops
way
and
the
principles
purposely
don't
cover
that.
A
So
we
actually
have
a
an
initiative
to
create
what
is
essentially
like
peer
review
journal
within
the
githubs
project,
so
that
people
can
submit
these
white
papers
at
exploring
these
topics
that
could
potentially
come
up
for
community
adoption,
but
they're
they're.
A
So
when
we
look
at
the
gaps
principles
just
to
share
kind
of
the
the
process
of
this
and
actually
and
we'll
come
back
to
the
events
in
a
second
but
the
get
off
principles
version
1.0,
we
had
over
96
different
interested
parties
involved
in
that
we
had
over
60
different
companies.
This
was
to
create
four
principles.
A
It's
very
short:
it
that's
the
effort
that
it
took
to
get
there,
because,
when
you
create
standards,
creation
is
is
tricky.
So
when
we
put
out
those
principles,
we're
doing
that
with
the
full
backing
of
these
60
companies,
these
96
interested
parties
and
these
dozens
of
co-authors
who
have
looked
at
these
principles
and
said
yeah.
We
agree
like
we
vote
to
adopt
these
principles,
and
so
when
we
put
that
out,
there
is
a
really
really
high
bar
and
degree
of
confidence
in
those
principles.
A
That
is
not
necessarily
fully
endorsed
that
this
is
a
place
where
we
can
explore
kind
of
the
ideas
and
some
of
the
rough
edges
and
some
of
these
patterns
that
were
that,
like
christian,
was
talking
about
you
know,
obviously
you
don't
have
to
use
customize
to
do
get
ups,
but
it
is
definitely
a
very
popular
pattern
that
we're
seeing
and
so
discussing,
like
the
challenges
of
dealing
with
overlays,
for
example.
A
These
are
these
are
things
that
we
want
to
do
so
if
those
initiatives
sound
interesting
to
you,
we'd
love
to
have
your
involvement
get
involved,
and-
and
we
can,
we
can
talk
about
those
principles
a
little
bit
in
a
second,
some
of
the
other
things
that
we
have
coming
up.
If
you
missed
it,
get
up.
Con
la
was
the
most
popular
and,
in
my
view,
successful
co-located
event
at
kubecon
la
it
was.
It
was
the
largest
co-located
event
done
with
kubecon,
and
it
was
very,
very
popular.
A
So
we
definitely
encourage
you.
If
you
haven't
seen
the
talks,
I
highlighted
the
talk
by
ricardo
roca.
He
gave
a
talk
about
how
cern
is
using
git
ops
with
argo
cd
to
and
cross
plane
to
manage
massive
infrastructure,
and
I
just
saw
an
article
today
talking
about
how
they've
sort
of
gobbled
up
all
of
the
gpus
in
the
cloud
for
all
of
the
stuff
they're
doing.
Chick-Fil-A
gave
a
great
talk
about
how
they're,
using
argo,
cd
and
git
ops
at
all
of
their
locations
to
manage
all
of
the
services
in
every
store.
A
Starbucks
gave
a
talk
discussing
how
they're
using
flux
and
argo,
cd
and
git
ops
to
do
services
in
their
stores,
so
there's
an
edge
use
case
here.
That's
really
interesting.
State
farm
gave
a
talk
focused
on
flux.
Now
this
is
another
thing
right
now.
Flux
and
argo
cd
are
really
the
leading
get
ops
solutions
out
there,
but
they're,
not
the
only
ones,
and
there
are
a
lot
of
kind
of
commercial
partners.
We
would
love
to
get
them
involved.
A
So
if
you,
if
you're
looking
at
hey,
we're
doing
this
get
ops
thing
from
our
company,
we
want
to
have
representation
within
get
ops.
Con
go
shop,
get
in
get
in
here,
you're
welcome.
This
is
a
this
is
an
open
source
initiative.
It
is
not
tool
specific,
it's
it's
it's
vendor
agnostic,
so
we
want
to
hear
about
how
you're
doing
get
ups
and
we
want
to
hear
what
tools
you're
using
so
please
come
do
that
if
you
haven't
seen
the
youtube
playlist,
it's
fantastic.
A
If
you
just
search
get
ops
con
la
on
youtube,
you'll
find
the
playlist,
and
maybe
maybe
somebody
can
post
it
in
the
chat,
but
the
playlist
is
very
good.
The
talks
are
very
good,
so
check
those
out
and
share
them,
because
there's
so
many
patterns
and
really
interesting
ideas.
There
are
a
number
of
talks
about
security
that
like
blew
my
mind
with
how
deep
they
were
and
and
how
insightful
they
were
so
check
those
out
and
upcoming.
Of
course,
we
have
at
another
get
ops
con
happening
in
may
may
17th
in
valencia.
A
There
will
be.
I
believe
that
I
I
I
can't
remember
exactly
what
the
status
is
of
the
hybrid
nature
of
the
virtual
event.
I.
B
A
Perfect,
thank
you,
diane,
yeah
and,
and
and
and
shout
out
to
to
christian
and
red
hat,
for
taking
the
lead
on
this
event
and
getting
it
organized
cfps
are
still
open.
They
close
on
february
14th.
So
if
you
have
an
awesome,
get
ops
use
case
or
you
know
someone
that
does
go,
get
them
to
submit
their
cfps.
We
want
the
best
talks.
We
want
the
best
real
life
use
cases
best
reference
architectures
and
when
I
say
that
you
know,
if
you
think,
hey
man,
I'm
just
doing
a
small
thing.
A
It's
not
that
cool,
submit
the
talk
anyway,
because
a
it.
We
may
actually
think
that
it's
better
than
you
do
and
and
want
you
to
come
and
give
that
talk
but
b.
We
also
look
at
these
talks
and
there
are
a
number
of
get
ops
events
that
happen
and
meetups
and
things,
and
we
look
at
those
talk,
submissions
and
say
hey.
Actually,
maybe
we
can
have
this
person
come
present
in
a
community
meeting
or
maybe
we
could
have
this
person
come
just
record,
a
video
or
or
whatever.
A
B
B
The
talk
dan
everything
and
to
add
to
that
we're
gonna
have
dual
track
this
time
in
get
ops
con.
So
don't
so
don't
think
so,
if
you
think
like
like
dan
said,
if
you
don't
think
you,
you
have
a
good
idea
submit
it
anyway,
because
we
have
a
dual
track
now.
So
there's
not
even
a
more
of
a
chance
for
your
talk
to
be
submit
to
be
accepted
all
right.
So.
B
B
A
So,
thank
you.
Thank
you.
Everybody
for
joining
please
get
involved,
go
to
opengetouts.dev
click,
the
get
involved,
tab
we've
got
a
subreddit.
We've
got
a
slack
channel
lots
of
get
up
discussions.
There's
open
meetings,
we're
looking
for
more
help.
We
need
you
to
do
it.
Thank
you.
So
much
take
care.
Go
get
ups.