►
From YouTube: GitOps Guide to the Galaxy (E51) | Using GitOps to simplify your cloud native migrations
Description
Cloud Success Architect Christopher Nuland joins our Gitopsians and Guides to share about how he has used Konveyer, gitops and other tools to manage Enterprise Migrations, the GitOps way.
Want to learn more?
https://www.konveyor.io/
Try Red Hat OpenShift: https://red.ht/3sMaW7O
Follow us:
Hilliary Lipsig https://twitter.com/Int2Caffeinate
Christian Hernandez https://twitter.com/christianh814
[add guest handles as needed]
OpenShift https://twitter.com/openshift
Subscribe to Red Hat's YouTube channel: https://www.youtube.com/redhat/?sub_confirmation=1
What is GitOps Guide to the Galaxy?
Every other Thursday at 3pm ET host Hilliary Lipsig and guests dig in to discuss everything in the GitOps universe, from end-to-end CICD pipelines to creating Git workflows.
Learn GitOps https://red.ht/3saqlzS
What is GitOps? https://www.redhat.com/en/topics/devops/what-is-gitops?sc_cid=7013a000002qG1BAAU
About OpenShift:
Red Hat OpenShift is an open source container application platform based on the Kubernetes container orchestrator for enterprise application development and deployment.
OpenShift on redhat.com https://red.ht/3LGCIeD
OpenShift on GitHub https://github.com/openshift/
#RedHat #GitOps #Kubernetes
A
B
C
C
C
So
that's
like
one
big
thing
right
so
where,
instead
of
being
co-located
with
kubecon,
will
be
co-located
open
source
Summit,
both
North
America,
which
will
be
Vancouver
this
year,
which
I
think
is
really
cool,
haven't
been
haven't
been
back
to
Canada
in
pre-covered
days,
love,
Canada,
love
Vancouver
by
the
way,
and
then
we're
gonna
do
open
source
Summit
code
located
open
source
Summit
in
the
in
in
Europe
right
EU,
which
will
be
in
Bilbao
in
Spain
right.
So
we
will
still
have
get
Ops
con.
C
C
You
know
recent
recent
recent
years,
it's
been
two
tracks
right
one
day,
but
now
we're
gonna
have
multiple
days.
Multiple
tracks,
it's
gonna
be
a
lot
bigger.
We
are
partnering,
and
this
is
the
third
the
third
thing
about
this
announcement.
We
are
partnering
with
a
cdcon
right,
so
it's
going
to
be
a
CD
con.
Slash,
get
Ops
con
right,
it's
all
related
right,
CD,
the
CD
Foundation.
C
C
C
Cfps
are
open
for
CD
con
or
get
opscom.
So
if
you're
doing
anything
with
Argo
CD
with
tecton
with
Jenkins
around
get
Ops.
This
is
where
we're
looking
for
we're
looking
for
papers.
So
that's
the
big
announcement
we're
moving
but
moving
to
in
order
to
make
the
event
bigger
and
so
I
see
Gerald
on
chat,
saying
British
Columbia
is
the
best
yes,
I
do
I
do
like
BC.
Bc
is
cool.
C
B
C
B
Okay,
so
that
said,
yeah
submit
your
cfps
guys
minor
thing.
It
was
asked
to
me
privately.
What
is
the
green
thing
in
the
background
for
folks
who've
been
around
the
stream
for
a
while.
You
remember
that
my
unofficial
job
title
is
Chief
mermaid,
it's
a
mermaid
tail.
That's
it
that's
what
that
is!
Ask
it
answered.
You're
welcome.
B
A
Awesome,
thank
you.
Hillary
yeah,
my
name
is
Christopher.
Newland
I've
been
involved
in
migrations
for
about
going
on
about
14
years
now,
most
of
my
professional
career
has
been
in
Cloud
migrations,
I
actually
started
with
Oak
Ridge
Labs
just
went
a
lot
of
the
cloud.
Technologies
were
taking
off
and
container
Technologies,
so
this
was
pre-docker
kind
of
in
the
late
odds,
spent
some
time
in
the
finance
industry,
doing
micro
services
and
then
the
last
five
years,
I've
been
with
red
hat
as
an
open
shift
specialist
dealing,
mostly
with
large-scale
Fortune
500
migrations.
C
A
What
I've,
seen
mostly
from
Enterprise
is
that
they've
kind
of
tested
the
waters
they've
had
some
green
field.
They've
had
some
hybrid
Cloud,
maybe
a
few
different
things
running,
but
we're
seeing
a
large
effort
from
large
organizations
to
start
picking
up
a
lot
of
their
legacy,
applications
refactoring
those
and
then
getting
them
into
the
cloud
in
services
like
kubernetes,
eks,
openshifts,
Rosa
arrow.
A
C
A
C
Well,
I
I
actually
have
more
anecdotal
than
anything
else.
Like
do
you,
so
you
say
you
work
like
with
like
large
Enterprises,
and
you
know
you
know
smbs
right
who
do
you
find
how
do
I?
How
do
I
ask
this
question
I'll,
just
ask
it
the
way:
I'm
thinking
it
who
was?
Who
was
the
easiest
to
migrate
right?
So
a
lot
of
us,
so
you
know
a
lot
of
us
assume
big
Enterprises.
They
they
take
a
longer
time
to
shift,
but
you'd
be
surprised
in
some
places.
C
How
like
how
fast
they
can
go.
I
I've
known
like
smaller
mom
and
pop
type
or
smaller
startups
type
that
they're
actually
slower
to
to
be
because
they
try
to
move
fast,
they're
slower
to
change
Technologies
right
they
they
because
like
they
need
to
like
well
like
I,
can't
change
because
I
need.
You
know,
I
need
to
release.
You
know
in
a
few
hours
right
and
a
few
hours
after
that.
We
need
to
do
another
release
like
it's
hard
to
so
like
I.
C
A
It's
it's
important
to
make
sure
that
everyone's
on
board
on
the
same
page,
where
I
find
migrations
fail,
is
when
those
expectations
haven't
been
communicated
across
an
organization.
So
you
have
a
team,
that's
handling
this
migration
and
then
on
the
side.
You
have
these
app
teams
and
maybe
they've
heard
some
rumors
about
an
app
a
migration.
But
then
you
go
up
to
them
and
say:
oh,
hey,
we're
ready
for
you
it's
time
to
go
and
they
just
haven't
heard
anything
and
the
expectation
it
leads
to
hard
feelings
or
you're
injecting
work
into
their.
A
A
B
B
What
do
you
all
want
to
see
and
he
was
like
conveyor
and
I-
was
like
oh
work
on
it
right
which,
like,
fortunately,
that
was
very
easily
done,
because
you
and
I
had
already
had
some
levels
of
conversations
by
then
so
sad
that
he
can't
can't
stay
with
us
live,
but
he
knows
how
to
reach
me
if
he
has
questions
later.
So
that
said,
Chris
I'm
gonna
we're
just
gonna
turn
it
over
for
you-
and
this
is
this-
is
your
show
now
you
get
to
run
it.
A
Yeah
sure,
so
what
I
always
recommend
and
I'm
I'm
approaching
the
conversation
today
from
a
very
high
level,
more
from
a
methodology,
kind
of
theory,
approach
on
how
we
handle
migrations
and
how
good
Ops
plays
in
that
I
know.
Hillary
we've
already
talked
about.
If
people
are
interested,
you
know
I
think
there's
opportunities,
maybe
for
me
to
come
back
and
maybe
do
a
demo
of
how
I
would
then
structure
this
from
an
execution
standpoint.
A
A
You
know
the
Netflix
stack
was
very
popular,
so
there's
a
series
of
tools
that
you
would
use
like
Eureka
and
others
that
for
communication
between
between
microservices
and
a
lot
of
those
just
don't
play
nicely
with
with
kubernetes
kubernetes
has
its
own.
You
know
DNS.
It
has
its
own
tools
for
doing
that
type
of
communication
locally
from
to
from
container
to
container
from
pod
to
pod.
A
A
A
A
Things
are
going
to
be
going
at
a
certain
point,
they're,
either
on
the
train
or
you're
not,
but
it's
going
to
be
leaving
at
that
point
and
you're
going
to
have
to
catch
the
next
train.
Don't
fall
too
much
in
this
trap,
where
you're
you're
going
to
be.
If
people
will
I
always
think,
migrations
are
very
challenging
for
organizations
very
resistive,
because
most
of
the
people
aren't
going
to
be
wanting
to
do
this.
Like
I,
said
you're
you're
interfering
with
other
people's
work.
A
You
know
they're
trying
to
make
improvements
to
applications
and
processes
within
an
organization,
so
always
be
thinking
about
that
and
from
that.
That
point
make
sure
that
they
know
they
have
these
set
points
where
they
can
move.
I
I
know
a
lot
of
people
when
they
talk
about
migrations,
I
like
to
focus
more
on
the
dependencies.
A
C
A
A
Try
to
build
confidence
in
your
team
when
you're
doing
migrations,
especially
from
maybe
a
legacy
system,
or
maybe
you
have
on-prem
like
an
on-prem
system,
that's
running
like
Docker
swarm
or
maybe
older
version
of
kubernetes
and
you're
moving
into
something
like
an
openshift
Rosa
or
some
of
the
other
or
Cloud
native
platforms
that
are
in
the
cloud.
I
I
would
highly
recommend
just
building
that
confidence
find
the
low-hanging
fruit
find
those
applications
that
are
stateless
and
are
going
to
be
a
lot
more.
A
There's
a
few
few
advantages.
So
when
I've
used
Argo
Argo
CD
for
a
few
of
my
migrations,
you
you
end
up
having
what
I
call
a
receipt.
You
know.
That's
that's
the
beauty
of
git
is
that
you
have
this
account
of
when
things
went
and
what
what?
What?
What
were
the
effects?
After
that
you
know,
was
it
a
successful
deployment?
What
were
some
of
the
errors
were
some
things
that
we
saw
the
time
stamp
and
for
large
Enterprise
organizations.
This
is
very,
very
critical.
A
It's
very
critical
because
there
needs
to
be
an
account
of
when,
when
and
in
the
status
of
of
different
applications,
as
they're
going
in
and
essentially
in
a
large
organization,
you
have
a
lot
of
stakeholders
who
are
very
invested
in
this.
They
need
to
know
that
they're
not
interested
in
all
the
technical
details
they
just
want
to
know.
When
did
it
go
and
was
it
successful,
and
this
is
where,
like
I
said,
the
source
of
Truth
is
very
important.
A
A
spreadsheet
may
have
some
of
that
information,
but
it
may
be
stale,
it
may
not
have
been
updated,
and
this
is
the
beauty
of
good
Ops
and
the
model
I'm
really
pushing.
For
you
know
more
and
more
is
get
all
that
in
from
all
that
migration
information
into
your
git.
Now
you
may
have
spreadsheets,
but
let
the
spreadsheets
be
updated
from
that
source
of
Truth
and
that
then
kind
of
goes
into
the
next
conversation
about
conveyor
and
some
of
the
tools,
but
Hillary
and
Christian.
B
A
Absolutely
so,
in
that
analysis,
phase
I
always
recommend
to
have
an
individual
meeting
with
each
application
team.
So
one
one
thing:
I,
don't
like
I,
don't
like
those
big
bang.
Let's
all
get
in
a
room
talk
about
migration
and
then
we're
done
because
migrations
are
very
personal
and
each
each
application
team
is
going
to
have
a
different
need
and
approach.
A
Usually
I've
had
an
analysis
of
the
application.
Previously
to
this,
where
I've
used
like
conveyor
tackle
to
do
an
in-depth
analysis
of
some
of
the
technical
challenges
around
this
application,
but
from
a
dependency
standpoint,
the
application
team
is
going
to
know
the
most
a
lot
of
those
QA
SRE
people
are
going
to
be
the
ones
that
are
going
to
be
able
to
identify
those
dependencies
compared
to
what
what
a
dependency
graph
will
show.
So
it's
a
very
combination
of
both
I
would
say,
but
no
Wheeling,
very,
very
heavily.
B
A
B
One
of
the
things
that
we
did.
We
had
iot
servers
right
in
this,
so
we
went
from
a
monolithic
stack
for
the
iot
server
and
we
were
migrating
it
over
to
a
microservice
driven
kubernetes
based
base
stack
on
that
and
yeah,
even
in
even
in
something
like
that,
which
is
like
totally
not
Enterprise
right.
The
the
tangled
web
we
weave
into
our
software.
Is
we
don't
I?
Don't
think
we
really
think
about
it?
A
Yeah
and
that's
where
I
think
the
integration
testing
is
so
critical
and
that's
when
you
you
have
to
Define.
When
is
your
integration
test
and
usually
that's
going
to
be
after
multiple
phases
of
your
migration
are
going
on,
so
there
may
be
things
that
are
moving
into
the
cloud
that
you
can
test
on
a
smaller
scale,
but
usually
that
that
switch
over,
which
I'll
talk
about
in
a
second
doesn't
come
until
that
integration
point.
When
you
have
all
those
tightly
coupled
things
figured
out.
B
Can
I
define
the
testing
levels
for
those
who
do
not
know?
Do
it
yeah?
Okay,
oh
I,
wish
I
had
a
whiteboard
okay
folks
types.
There
are
multiple
types
of
testing.
There
is
Black
Box.
There
is
white
box,
there
is
gray
box.
What
is
the
difference
between
those
three
well
Black?
Box
means
you
have
nothing,
but
the
user
experience
white
box
means
you
have
all
the
things
access
to
the
back
end
to
the
logs
to
the
code,
to
the
commits
to
the
anything
you
could
possibly
want.
B
Gray
box
means
you
basically
only
have
logs
and
if
you're
very
lucky,
also
some
observability
metrics
okay.
This
is
important
because
then,
in
addition
to
those
like
three
main
categories
of
testing,
you
have
different
levels
of
testing
right.
So
folks
who
are
familiar
with
the
concept
of
unit
testing
your
code,
that
would
be
like
functional
testing.
Does
this
thing?
Do
the
thing
I
want
it
to
do
right,
regression,
testing,
when
I
added
more
things
to
that
function,
did
I
break
any
of
the
other
aspects?
B
Okay
and
that
that
regression
piece
actually
is
like
very
small
to
full
system
end
to
end
wide
and
because
what
you're
looking
for
is,
did
I,
introduce
a
problem,
and
that
happens
at
every
single
layer
right.
So
that's
a
regression
test.
An
integration
test
is
okay.
Now
that
I
know
my
thing
works.
The
way
I
want
it
to
work.
B
B
Okay,
so
when
he's
talking
about
then
doing
all
the
integration
testing,
especially
in
a
migration
where
you
actually
could
be
hot,
swapping
out
components
like
networking
components
from
actual
Hardware
to
Virtual
Hardware,
these
are
how
you
want
stage
those
things.
So
you
can
understand
what
they
do
before
the
move
and
then
two
verify
those
things
after
the
move.
B
C
I'll
I'll
bite
all
right
load
load
testing
is
for
me
as
an
admin
is
how
much
load
can
this
an
application?
Take
right?
What's
what's?
What's
the
what's
the
the
high
Mark
right
of
what
you
know,
what
you
you
want
your
low
and
your
highs
right
and
you
wanted
to
know
where
your
application,
you
know,
what's
the
minimum
amount,
I
need
and
what's
the
maximum
amount
I
need
in
order
to
run
this
application?
Let's
load
stress
testing.
B
Key
difference
between
the
two
things,
and
that
is
duration
should
occur
over
long
periods
of
time.
The
longer
even
the
better
stress
testing
is
very
acute,
and
so
a
stress
testing
is
how
you
identify
what
your
auto
scaling
rules
ought
to
look
like
right.
So
not
only
are
we
going
to
start
hammering
this
application
from
all
ends,
but
we're
going
to
do
it
really
really
fast
and
then
we're
going
to
do
it
fast
for
the
next
cycle
and
so
you're
not
going
to
do
it
for
long
periods
of
time.
B
You're,
it's
going
to
be
nice
yeah,
the
biggest
spike
it
can
handle
exactly.
This
is
your.
Can
my
can
my
Web
Store
handle
Black
Friday
traffic?
That
is,
that
is
what
you
do
to
determine
it,
because
you
will
have
very
very
sudden
spikes
and
you
want
to
know
how
long
does
it
take
to
scale?
How
does
your
application
behave
while
at
scaling?
What
is
the
user
experience
while
those
scalings
are
happening?
B
A
Absolutely
and
to
that
point
the
the
functional
testing
is
very
much
the
type
of
testing
when
you're
bringing
over
the
individual
pieces.
That's
where,
like
that
low-hanging
fruit
analogy,
you
can
functionally
test
it,
but
not
until
all
the
pieces
have
been
moved
over
that
it
relies
on.
Can
you
do
that
integration
type
of
test
and
that's.
A
A
A
Variations,
you
can
just
do
what
we
call
like
a
big
bang.
Cutover,
that's
just
everything
at
once
is
getting
moved
over.
You
just
press
a
button
and
that's
where
you
hope
a
lot
of
integration
testing
has
been
done
prior,
but
then
there's
also
like
an
A
B
type
of
deployment.
Where
you
know
you
can
deploy
from
a
percentage.
A
So
and
that's
always
what
I
recommend
if
you
have
that
capability
you'll
start
your
your
new
Cloud
platform
at
like
10
of
traffic,
and
maybe
you
have
an
old
on-prem
and
then
just
slowly
kind
of
build
that
number
up
over
some
agreed
upon
time
but
yeah.
That's
those
are
kind
of
the
three
major
phases
there's.
Obviously
some
Sub
sub
categories
that
go
into
that,
but
I
do
want
to
phase
a
little
bit
more
into
conveyor.
A
A
Acted
perfect,
so
conveyor
right
now
is
a
set
of
different
sub
projects,
which
I'll
kind
of
dive
into
right.
Now,
the
direction
that
the
community
is
going
is
the
more
consolidate
this
a
little
bit
more
into
one
one
project.
That
then
has
like
add-ons,
but
this
is
just
from
our
historical
reasons.
This
is
a
a
project.
A
That's
come
together
from
a
bunch
of
smaller
projects,
so
we're
kind
of
seeing
that
evolution
of
the
kind
of
maturing
into
its
own
singular
project,
but
just
to
dive
into
what
there
is
now
and
I'll,
be
focusing
more
on
tackle
here
in
a
moment.
But
crane
is
a
a
project
tool
that,
if
you're
going
from
a
kubernetes
instance
to
another
kubernetes
instance,
it's
a
very
powerful
tool
that
will
make
that
possible,
and
this
is
all
manifest
based,
which
is
right
in
line
with
good
Ops.
A
So
if
you're
doing,
let's
say
you're
going
from
like
a
vanilla,
on-prem,
kubernetes
and
you're
moving
into
like
openshift
Rosa
on
AWS
well,
this
is
a
really
amazing
tool
where
you
just
kind
of
point
at
the
two
locations,
and
then
it
will
not
only
create
a
migration
plan
which
is
itself
is
a
manifest.
So
this
is
like
a
this
is
not
a
receipt
there's
actually
a
post
thing
too.
So
this
is
more
of
a
a
plan
to
find
as
a
manifest.
A
So
this
is
a
something
that
you
can
track
through
git
and
you
can
even
trigger
your
migration
through
your
through
your
actual
commit
into
the
repository.
So
this
tool,
like
I,
said
also
there's
like
a
receipt
afterwards,
which
is
also
a
manifest
it'll
print
out.
All
the
the
details
from
it
was
it
successful.
Did
it
fail?
Where
did
it
fail?
A
So
that's
very
much
from
a
kubernetes
to
kubernetes
pathway,
forklift
I'm
not
going
to
talk
too
much
about
that's
more
taking
virtual
machines
into
into
cupert
or
openshift
fertilization.
If
your
flavor
you're
using
and
it
functions
similar
to
Crane,
where
there's
a
series
of
manifests
so
there's
an
operator
that
then
executes
on
that
migration,
the
two
that
I
want
to
focus
the
most
on
are,
is
this
move
to
Cube
and
Tackle?
A
This
is
where
we
hear
the
most
challenges.
It's
hey,
I,
I
know:
I
want
to
go
to
the
cloud,
but
I
don't
really
know
how
to
get
there.
I
know
what
I
know
what
the
business
people
are
telling
me,
but
I
don't
know
if
our
applications
are
actually
ready
to
make
this
move,
and
these
kind
of
tools
have
been
usually
done
through
different
consulting
firms.
They
have
the
kind
of
their
own
in-house.
This
is
really
the
first
open
source
tool.
A
That's
really
enabling
the
community
at
large
to
be
able
to
do
these
type
of
analysis.
This
is
like
I
said:
it's
open
source
tools,
so
all
the
rules
are
there.
You
can
take
a
look
at
them
right
now.
They're
very
Java
focused
be
just
given
the
history
of
where
this
tool
is
coming
from,
and
you
know
most
most
Enterprise
applications.
We're
dealing
with
right
now
are
some
form
Java
based,
but
that's
being
expanded
in
different
directions.
A
A
Is
there
opportunities
to
move
from
like
Java
8
to
Java,
11
or
future
versions
I'm
not
going
to
dive
too
much
into
this
I
think
this
is
good
for
a
separate
demo
in
the
future,
but
I
do
have
the
tool
up,
which
I
can
kind
of
give
you
a
look
into
it.
There's
What's
this
called
is
the
assessment.
The
assessment
phase
is
actually
it's.
A
It's
not
linked
to
a
repository,
it's
more
of
a
series
of
questions
about
how
your
team
functions
and
how
the
application
functions,
and
so
this
is
more
of
a
questionnaire
that
then
gets
combined
with
the
what
we
call
the
analysis
phase
and
then
this
is
the
tool
that
actually
goes
in
and
does
the
the
breakdown
and
I'll
go
ahead
and
just
show
you
what
one
of
these
reports
looks
like
and
it
breaks
it
down,
which
is
really
cool.
It
even
gives
you
like
an
estimated
story
points.
A
C
Well,
what's
cool
with
it
so
like
what
what
I've
always
wondered
and
and
purposely
being
asking
dumb
questions,
what
I've
always
wondered,
I
like
I
would
like
to
go,
maybe
a
little
deeper,
maybe
not
too
much
because
of
time,
but
so
what's
the
so
with
with
respect
to
application,
specifically
right.
So
let's,
let's
remove
the
VM
aspect
of
it,
like
you
said
before,
so,
let's
say
like
you
say
you
have
a
spring
application
right.
You
have
a
spring
up.
C
What's
the
difference
from
using
something
like
conveyor
versus
using
something
like
s2i
or
build
packs,
or
something
like
that
right,
because
the
idea
of
source
to
image
right,
it's
like
I,
just
take
my
source
code
and
you
build
a
container
out
of
it.
So
what's
the
difference
between
running
s2i
and
running
it
through
conveyor
right
can?
Can
you
kind
of
like
go
over
those
those
differences?
A
So
conveyor
is
going
to
be
conveyor,
is
more
of
a
re-factoring
and
re-platforming
Tool,
where
s2i
is
more
of
a
tool
to
help
speed
up
your
your
deployment
into
something
like
openshift
and
the
nice
thing
about
conveyor
2.
Is
that
it's
it's
not
locked
into?
You
know
any
specifics
around
any
version
of
openshift
or
any
other
version
of
kubernetes.
A
A
Re-Platforming
would
be
something
like
moving
from
something
like
pivotal,
Cloud
Foundry,
to
kubernetes,
where
there's
usually
a
little
bit
more
effort
involved,
but
not
any
change
through
necessarily
to
the
core
applications.
At
least
significantly.
There
might
be
some
middleware
changes,
but
the
hope
would
be
there'd
be
less
of
those
where
refactoring
is
really
going
in
saying
this
Eureka
service
discovery
in
my
Java
microservice
is
not
going
to
play
nicely
with
kubernetes,
DNS
and
I'm
going
to
have
to
just
rip
that
out
of
my
applications
to
make.
A
It
then
be
ready
to
play
nicely
in
this
in
this
type
of
environment,
and
there
could
be
things
where
it's
like.
Well,
this
type
of
Java
application,
or
this
particular
package-
that's
being
used-
is
non-compliant
with
another
technology
that
you're
using
in
kubernetes,
and
that's
really
where
it's
kind
of
focused
on.
It's
really
focused
on
that
refactoring
like
how
do
I
change
my
application
in
a
minimal
way
to
get
it
into
the
cloud
and
take
the
most
advantage
of
the
cloud,
which
is
how
I
want
to
present
this.
A
How
do
I
take
advantage
of
the
Technologies
I'm
moving
to
ultimately-
and
that's
where,
like
that
refactoring
element
comes
into
play,
so
tackle
is
a
very
cool
tool.
I
love
it.
Some
of
the
things
that
the
community
is
talking
about
right
now
is
to
be
able
to
track
more
of
your
migration
from
this
tool.
A
I
think
that's
going
to
be
awesome!
That's
where
your
your
spreadsheet
then
just
becomes
a
first-class
citizen
of
kubernetes
in
a
lot
of
ways.
It's
all
going
to
be
tracked
within
within
the
git
repository,
not
not
in
some
spreadsheet,
or
that
has
10
different
versions
floating
around
your
organization.
A
We're
coming
up
with
time
here
soon
I
want
to
have
some
time
for
questions.
The
only
thing
I
want
to
highlight
here-
and
this
is
where
I
think,
if
there's
an
interest
from
this
community
here,
we
could
talk,
but
maybe
do
like
a
demo
in
the
future.
I
move
to
cube
so
do
reach
out
if
you're
interested
but
I
do
want
to
talk
about
move
to
cube
a
little
bit.
A
Move
to
cube
is
pertains
a
lot
to
this
conversation
we're
having
about
good
Ops,
because
it's
a
tool
that
does
an
evaluation
of
different
different
application
sources.
So
the
example
I
just
gave
a
minute
ago
was
pivotal,
Cloud
Foundry,
so
let's
say
you're
doing
a
migration
from
pivotal
Cloud
to
kubernetes
or
openshift
and
you're
wanting
you're
very
new
to
kubernetes
and
you're
needing
a
way
to
generate
all
those
assets.
All
those
different
manifests.
A
Well,
that's
where
a
move
to
cube
comes
into
play.
It'll,
actually,
look
at
your
current
iteration
of
your
application
in
you
know,
PCF
or
other
platforms,
Docker
swarm,
maybe
you're
just
doing
Greenfield.
There's
a
few
variations
of
that,
and
then
you,
you
didn't
go
through
this
evaluation.
It'll
then
it
will
crank
out
your
assets
and
that's
where
I
think
it
combines
very
well
with
tools
like
Argo
CD,
because
you
can
then
take
those
assets
and
then
plug
it
right
into
a
git,
Ops
approach
and
the
best
time
to
do.
A
A
Tackle.
Has
these
features
called
add-on,
add-ons
they're,
just
like
plugins
that
you
can
bring
into
bring
into
tackle,
and
this
is
more
my
opinion
of
where
the
direction
of
the
community
is
going,
but
I
think
tools
like
move
to
cube
you're,
going
to
see
some
of
that
functionality.
Kind
of
breaking.
A
Apart
into
tackle,
as
these
type
of
add-ons
and
one
I
want
to
show
here,
was
this
one
that
one
of
the
the
community
members
has
put
together,
and
this
one
is
just
it's
happening
during
that
analysis
phase
that
I
was
showing
you,
and
this
is
assuming
a
it's
assuming
a
Greenfield
application
or
a
legacy
application.
That's
Java
hasn't
been
containerized,
it
hasn't
moved
to
to
anything
like
kubernetes
or
Docker
swarm,
and
it
does
part
of
that
analysis
that
I
was
showing
before.
A
But
then
it
will
also
generate
your
your
assets
like
move
to
Cube
and
I,
see
the
future
being
more
of
these
types
of
more
modular
types
of
approaches
where
maybe
you're
doing
this
with
a
python
application.
Maybe
you're
doing
this
with
a
Ruby
on
Rails,
and
this
is
kind
of
hype
right
I'm,
seeing
right
now
in
the
community
is
having
more
of
these
more
modular
add-ons.
That
will
do
a
lot
of
this
lift
and
shift
for
you
going
into
going
into
the
cloud
and.
A
A
A
Things
like
you,
know,
companies
plan
for
for
more
than
half
their
customer
custom
applications
to
be
modernized
for
the
next
year.
This
is
all
the
way
up
to
27
within
six
months,
so
that's
actually
right
around
now,
actually
given
when
we
did
this,
so
this
can
just
give
you
an
idea
of
of
just
the
impact
and
scale
what
other
companies
are
talking
about
and
where
their
concerns
are
so
things
like
that
and
then
right
to
our
Point
earlier,
like
difference
between
repot
post
refactor.
B
C
B
B
A
B
C
B
B
C
B
Our
chat
is
regularly
very
happening,
it's
not
usually
on
topic,
but
it's
always
happening
of,
and
that's
that's
a
fantastic
part
of
our
our
little
User
Group
yeah,
there's
sometimes
there's
references
to
things
of
episodes
past,
like
one
of
the
times,
my
cat
attacked
me
from
above
the
bookshelves
during
an
episode,
occasionally
I'll
get
asked
the
cat's
still
there.
The
cat
is
actually
down
there.
This
time,
although
she
threw
a
actual
book
at
me
yesterday
morning,
wow.
B
C
B
B
C
See
there's
so
here:
here's
here,
here's
a
good
I'll
cause
a
little
controversy
here.
There's
someone
saying
we're
running
key
cloak
and
kubernetes
cluster
I'm,
actually
a
fan
of
Dex,
so
so
someone
yeah
yeah
yeah,
so
shots
fired.
So
if,
if
the
key
cloak
team
wants
to
reach
out
to
me,
I'll
tell
you
exactly
what
you
need
to
do
for
me
to
change
my
mind.
C
B
I
was
gonna,
say
interesting
thing
is
they
were
talking
about
making
a
kubernetes
operator
for
key
cloak
using
Java
instead
of
golang,
which
is
a
thing.
So
that
was
that's
an
interesting
anecdote
that
I
don't
know
if
they've
done
it
or
if
they've
canceled
it
or
what
but
keep
an
eye
on
that
and
then,
let's,
let's
read
this
poem
I'm
going
to
show
it
it's
going
to
cover
my
entire
face.
B
B
C
B
There
we
go,
we
ended
up
with
so
much
time
to
just
kind
of
like
sit
around
and
be
silly
for
a
minute
here.
Chris
you
gave
us
a
lot
of
information.
It's
really
the
truth
of
it
and
I
I
still
feel
like
I
have
to
digest
things
to
be
smart
about
it,
but
I'm
gonna,
preemptively
say
we're
going
to
have
you
back
to
do
a
demo
of
some
of
this
stuff,
because
I
learned
things
today,
which
is
always
my
favorite
episodes
yeah.
B
B
A
B
B
From
yeah
yeah,
he
actually
had
talked
about
this,
specifically
the
different
different
Dynamic,
a
little
bit.
What
what?
What
is
that
about
single
replica
rules?
Single
replica
rules
is
like
a
terrible
thing.
Don't
do
that?
If
you
have
only
one
replica,
you
don't
have
high
availability,
you're,
not
a
reliable
piece
of
software
done
end
of
story.
The
SRE
has
spoken
that
the
gavel
is
down.
That
is
the
ruling.
B
C
There's
there's
there's
a
story
with
that
and
and
actually
Clayton
Coleman
actually
told
the
story
on
a
podcast
a
while
a
while
back
talking
about
how
like
open
this.
Oh,
this
is
how
old
the
story
openshift
online
right.
Their
upgrades
would
go
down
because
at
CD
like
needs
a
certain
amount
of
PODS
up
and
running
when
you
are
like.
Oh,
we
need
a
pod
disruption
budget.
It's
just
kind
of
funny.
How
like
these
things,
come,
come
to
mind
it.
It
came
from
SRA
by
the
way
Hillary.
C
B
B
Yep
double-edged
sword
on
that
one,
but
I'm
still
gonna
prefer
the
Pod
disruption
budget
over
the
knot.
Gonna
gonna
go
with
that
one,
but
you
use
use
your
technical
debt
wisely.
Choose
your
technical
debt
wisely
as
I
always
say.
Thank
you
chat,
people
for
being
so
chatty.
Thank
you
Chris.
For
joining
us.
We
will
have
you
back
for
the
the
deeper
stuff
Christian
I'll
just
see
you
around.
Whatever
dude
stay.