►
From YouTube: Kubernetes AWS Infrastructure - 02092023
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Hi
everyone
and
welcome
to
the
first
meeting
related
to
the
AWS
infrastructure
for
the
kubernetes
project.
Just
two
clarify
things
here.
This
is
a
dedicate
meeting,
that's
going
to
happen
if
we
possibly
every
week
or
every
two
weeks
until
the
month
of
April,
because
we
are
about
to
do
special
project
really
to
bootstrap
in
your
infrastructure,
on
AWS,
for
the
breakout.
I
just
want
to
just
basically
a
reminder
that
advice
gave
three
millions
during
kimco's
intro
of
last
year.
We
basically
want
to
work
on
make
sure
we
can
consume
them
this
year.
A
A
Thank
you.
Look
at
myself.
I
lost
I'm,
not
doing
the
introduction
here,
because
this
is
the
first
call
there's
going
to
be
a
lot
of
people
doing
that
solution.
I
want
to
focus
on
exactly
what's
needed
to
be
discussed.
A
A
C
A
Okay,
so
I
have
like
three
updates
to
give
science.
We
got
those
credit,
the
adding
the
first
thing
is
we
transfer
and
I.
Think
Rhian
will
give
confirmation
later
we
transfer
all
of
the
AWS
account
related
to
the
kubernetes
project
to
the
new
organization,
so
that
was
I
think
that
was
done
last
week.
Thank
you
for
that.
We
on
so
basically
now
all
the
go
job
ran
against
the
AWS
own
organization.
A
The
second
thing
is
like
science.
Yesterday
we
now
by
default,
serve
the
image
layer
from
S3
directly
for
everything.
That's
not
gcp
and
we'll
probably
see
this
in
the
future
in
the
building
of
AWS,
so
I
hope
we
can
show
something
in
the
next
case.
Information
happening
next
week
and
the
third
base
is
is
like
and
Justin
is
walking
on,
make
sure
all
the
binaries
serve
through
ad5.kids.io
will
be
served
by
from
S3.
There
was
some
kind
of
work
in
progress
about
this
I
think
we
just
need
to
go
in
production
and
I
hope.
A
D
Yeah,
sorry,
are
we
looking
at
spitting
up
our
Pro
build
clusters
in
AWS
next
month.
A
This
month,
in
fact,
there's
an
issue
I
create
what
is
that
yeah
I'm
gonna
zoom
in
so
yeah
again
we
can
put
that
in
the
dark,
so
I
create
initial,
and
that
was
I,
hope,
cubematic,
look
into
this
and
start
looking
to
this
I
think
GP
put
a
tag
for
the
folks,
so
we
can
start
take
a
look.
I
mean
anyone's
interest
to
basically
work
on
this
feel
free
to
yeah.
D
I
already
started
the
legwork
for
that
work,
so
proud
should
work
natively
us
and
we
just
need
to
spin
up
a
cluster
and
start
running
jobs
on
there.
I
can
drive
that
work.
It's
not
particularly
difficult
to
get
it
going.
Oh
we'll
just
need
an
account
created
for
this
cluster
for
the
build
clusters
and
then
they'll
live
in
there
and
that's
it.
A
D
A
D
A
A
Think
that's
the
first
thing
once
we
do
that
and
using
some
kind
of
infrastructure
called
tooling,
my
preference
will
be
tear
from,
but
if
people
are
comfortable
with
something
else
feel
free
to
work
on
it,
so
I
think
my
I
want
to
basically
be
because
we
have
a
contract
with
kubomatic,
give
that
to
Jiffy
and
Jiffy
drive
this
and
try
to
either
bootstrap
a
new
team
with
non-cubomani
people
and
keeping
my
people.
So
this
is
happening,
and
the
second
phrase
is
plug
that
ground.
A
So
I
think
my
first
thing
is
like
I
trust
this
issue
and
bootstrapper
eks
cluster
first
and
later
Vlog
with
brow,
and
we
basically
enter
it
over
time.
If
that's
easy,
it
should
be
quick.
We
should
be
out
all
of
this
done
by
end
of
this
month.
I
have
project.
In
next
month.
We
have
some
project
running
on
APS,
so
yeah.
A
Okay,
I
will
try
to
create
an
account
and
give
access
to
any
more
interest
to
walk
into
this
and
Jiffy
you
driving
this.
B
I
have
already
linked
it
to
the
kubernet
folks
and
got
an
act
from
them.
I
mean
Mario
has
his
hand
up,
so
hopefully
he's
going
to
go
next,
but
it
sounds
like
they're
going
to
start
looking
at
that
soon,
like
this
week,.
A
B
A
F
Yeah
yeah,
so
regarding
regarding
this
task,
we
already
linked
up
with
with
Jiffy
and
what
we
want
to
do
or
what
we.
F
What
we
propose
to
do
is
that
we
actually
take
over
this
this
job
to
create
the
cluster
and
also
to
take
care
of
the
of
the
power
installation
there
as
we
are
already
using
it,
or
we
are
already
using
Pro
internally
and
our
folks
are
have
quite
some
experience,
and
this
will
also
basically
work
into
the
proposal
that
we
want
to
give
eventually
in
the
next
official
Sigma
in
the
next,
not
AWS
migration,
stick
case
in
for
meeting.
A
A
A
A
Okay,
the
I
forgot
to
create
an
issue
about
this.
That's
like
that.
Also
another
one
item
I
would
like
to
Kiev
to
people
at
all.
I,
don't
have
any
issue
for
this,
so
I'm
gonna
say
it
by
the
way
I've
been
discussing
to
a
few
people,
so
I
think
I
would
like
to
say
Federate
access
on
AWS
using
the
the
existing
GitHub
organization.
A
So
for
a
long
time,
science,
the
bootstrap
of
the
project,
anyone
we
basically
handle
authentication
and
authorization
user
using
Google
Groups,
because
there's
a
time
integration
between
JCP
and
Google
Groups.
We
all
we
always
leverage
that.
So,
basically,
if
you
want
to
have
access
to
a
station
project,
we
use
Google
group
and
we
we
Define
an
ACL
that
gives
you
access
to
project.
A
So
whatever
else
is
different,
because
there's
no
identity
provider
and
the
project
don't
have
one
the
only
Pro
identity
provider
per
se,
it's
guitar
right
now
and
I,
don't
want
to
consider
that
as
an
identity
provider,
because,
basically
it's
individual
people
with
the
identity
that
we
use
in
GitHub
teams
to
basically
leverage
to
give
access.
So
the
one
thing
I
would
like
to
say
is
like
some
kind
of
integration
between
GitHub
and
get
address
at
some
point
relate
to
authentication
and
authorization.
A
Our
classification
is
the
most
difficult
part.
Authorization
is
something
else,
but
we
should
focus
on
the
notification.
How,
as
using
my
GitHub
identity,
I
can
authenticate
on
an
AWS
icon
and
basically
do
action.
How
that's
happening
I
need
to
put
that
in
some
kind
of
GitHub
issue
and
try
to
better
Express
what
I'm
trying
to
say.
A
C
If
yeah
I
I
haven't
yet,
but
that's
an
interesting
one,
I'm
gonna
reach
out
to
a
few
folks
on
it.
A
So
why
I'm
saying
this
is
because
we're
gonna
have
today
with
multiple
accounts
on
AWS,
so
we
need
to
be
able
to
delegate
access
to
those
accounts
at
some
point.
It's
just
painful
to
go
over.
Every
account
create
users
per
account
just
for
one
interviewer,
trying
to
basically
access
to
those.
So
I
would
like
to
basically
be
able
to
set
up
something
leveraging:
AWS
identity,
Center
and
saml
protocol
to
make
that
happen
or
oigs,
if
possible.
B
B
So
if
we
have
an
excess
of
effort
at
any
point,
I
think
it's
good
for
us
to
work
on
that,
because
that
is
like
a
sustainability
thing.
Managing
these
accounts
by
hand
is
a
pain
in
the
butt
yeah,
but
also,
if
like,
if
the
choice
is
coopermatic
working
on
eks
builds
clusters
for
you
or
kubermatic
working
on
IDP,
I'm,
going
to
say
the
build
clusters
right,
like
I
I,
just
wanna
I
just
want
to
make
that
clear.
A
Yeah
I
need
to
priorities.
Basically
yeah,
that's
right.
The
second
thing
is
like
okay,
I'm
gonna
phrase
this
differently,
like
my
I,
think
the
idea
is
like
having
people
focus
on
that
at
the
same
time
be
able
to
walk
on
prowl,
because
there
are
so
they
are
somehow
really
like.
At
the
same
time,
we
say
we
say
we
set
up
a
break
cluster.
At
the
same
time,
we're
trying
to
then
identify
issue
related
to
Authentication.
A
D
Oh
yeah,
so
I'll
just
put
out
a
few
things
there
right,
because
yeah
I
did
mention
something
about
IDP
a
long
time
demo
something
yeah.
The
short
answer
is
there
are
some
cool
commercial
products
that
can
bridge
that
Gap,
because
you
need
to
put
something
between
GitHub
and
AWS.
D
It
could
be
Azure
ID
or
it
could
be
something
else
from
all
all
to
zero,
for
example.
But
it's
a
question
that
we
need
to
resolve
at
some
point
this
year.
Otherwise
somebody
has
to
deal
with
manual
account.
A
A
A
Yeah
event,
options
available
yeah,
even
he
clock
at
some
point
is
limited
because
you
I
don't
I,
don't
remember
exactly
the
technical
detail,
but
plug
key
clock
with
GitHub
may
may
not
may
not
work
well
from
what
I'm
so
so
where
to
build
one
from
scratch.
Either
we
buy
something
but
I
don't
have
the
power
to
basically
say:
oh
I
have
money
for
that.
D
Yeah
I
agree:
it's
just
that
it's
a
conversation
that
we
need
to
have
yeah
experiment
with
key
gloves,
see
how
well
that
works.
First,.
D
Powerful
from
what
I've
read
a
commercial
offering
is
better,
but
it
needs
to
be
priced
up
and
see
if
it's
worth
the
cost,
but
it's
a
separate
work
stream
from
everything
else.
That's
going
on
so
IDP.
Isn't
that
important
right
now,
because
there's
a
small
number
of
people
working
on
the
account
so
they've
been
built
out
but
later
on
once
this
is
live,
there'll
be
a
larger
number
of
people
that
will
need
access,
and
then
we
should
probably
have
an
answer
for
that.
At
that
point,.
A
Yeah
and
that's
why
I
basically
want
to
put
that
into
that
effort
to
make
sure
we
have
this
under
the
radar,
and
we
have
to
address
it
at
some
point,
because
we're
gonna
be
I
basic
for
me
by
April.
We've
done
with
all
the
priorities,
and
we
now
start
to
do
migration
of
The
Proud
jobs
and
from
from
April.
We
need
to
like
come
back
to
this
and
say
and
see
what
need
to
happen.
A
B
In
this
case,
it's
a
little
different
because
hi
I'm
here
it
kind
of
Falls
all
under
the
Kate's
interest
stuff.
If
there
is
a
company
that
may
potentially
have
something
out
of
the
box
for
us,
that's
where
I
can
go
to
cncf
leadership
and
try
and
see
if
there's
a
relationship
that
we
can
use.
B
A
Okay,
am
I
empowered
to
go.
Do
the
negotiation,
because
so.
B
B
Okay,
again,
that's
that's
the
weird
thing
in
this
case
you
just
tell
me,
and
we
will
move
it
forward
because
it's
the
kubernetes
in
for
stuff.
If
it
was
something
completely
unrelated,
then
you
would
file
an
issue
in
kubernetes
staring
and
we
would
move
forward.
I'm
kind
of
like
your
Tam
right
now.
E
Wouldn't
bother
spending
too
much
time
looking
to
this
right
now
we
can
look
into
IDP
and
sort
of
farm
out.
What
might
actually
be
the
requirements
later.
A
D
I
can
write
up
what
we
need
right
and
what
I
commit,
what
good
commercial
products
are
out
there,
and
then
we
can
make
a
decision
on
what
to
explore.
Okay,.
H
A
Yeah
I
think
sure,
should
I
say
something
about
do
we
we
need
to
just
very
like
we
say
we
need
to
look
out
to
all
of
this
like
bootstrap,
something
it
could
understand,
trying
to
see
how
we
can
integrate
I,
don't
I,
don't
blog
on
anything.
A
What
I'm
putting
out
there
is
like.
We
need
something
we
need
to
be
able
to
authenticate
to
address,
using
GitHub
or
Google
workspace,
Google
group
to
be
to
be
more
specific,
I
mean
now
how
we
do
that
I,
don't
I,
don't
think
it's
the
right
moment
to
basically
talk
about
is
just
we're.
Gonna
open
an
issue.
People
can
drop
the
ideas
and
we
can
start
to
walk
into
this
later.
D
C
A
Out
yeah
yeah
I.
D
Will
demo
that,
just
to
complete
the
circle
right
and
also
propose
the
commercial
products
that
I
think
will
be
great
for
us
and
then
take
it
from
there.
A
Jay,
what
you
mentioned
is
one
bar
of
the
authentication
process
on
the
AWS
side.
What
we're
talking
about
on
the
basically
on
the
community
side
like
we
need
to
be
able
to
plug
identity,
Center
and
GitHub.
C
D
A
Yeah
I
would
like
to
basically
see
the
cost
of
wish
history.com
AWS
for
for
of
last
year,
because
I
need
to
put
that
in
the
annual
report.
I
forgot
to
give
I
forgot
to
give
to
give
the
detail,
because
I
don't
have
access
to
current
bill
of
this
year,
because
we
did
a
migration
before
last
year.
I
don't
have
access
to
eight,
so
I
need
that
to
put
that
in
the
annual
abroad.
I
Much
of
that
they
start
the
cncf
account,
and
so
when
we
migrated
over,
we
all
lost
access
to
that
at
the
cnco.
So
I
can
help
with
that
position.
A
H
B
I
just
want
to
make
sure
kubernet
knows
what
we're
throwing
at
them.
So
they
are
going
to
start
working
in
terms
of
AWS
stuff
they're,
going
to
start
working
on
the
eks
build
clusters
for
you
right,
yeah,
yeah,.
B
They
have
a
proposal
that
they
will
post
in
slack
whenever
they're
ready
to
push
it
out.
B
G
F
Also,
oh
sorry,
so
Patrick
in
the
call
here
would
be
the
person
to
reach
out
anything
regarding
regarding
Porsche
and
support
you
there.
You
can
contact
him
also
in
the
official
kubernetes
stack,
but
I
will
do
an
introduction
anyways
after
at
some
point
after
this
call
or
tomorrow
the
best.
Thank
you.
G
Thank
you,
I
just
want
to
say,
I
think
we're
in
we're
in
reasonable
shape
on
most
of
it,
I
think
they're.
One
thing
that
still
scares
me
is:
we
have
to
get
the
authentication
for
prowl
to
do
promotion
which
so
there's
already
promotion,
there's
a
similar
setup
for
registry.case.io,
but
we
have
to
do
the
comparable
thing
for
artifacts,
okay
study
or
just
to
be
able
to
copy
buckets
back
and
forth
from
the
trusted
proud
cluster.
A
I
One
of
the
things
that
I
want
us
to
take
into
account
is
the
AWS
request
that
we
spend
all
of
the
three
million
in
order
to
keep
it
I.
Think
one
of
the
things
that
we've
needed
as
the
organization
from
the
kubernetes
side
is
to
be
able
to
have
the
Headroom
to
grow,
and
if
we
get
to
a
certain
amount
and
that
gets
dropped,
our
ceiling
gets
dropped.
We
lose
our
Headroom
kubernetes
will
continue
to
grow
in
order
to
to
have
sustainability.
I
In
this
we
we
need
to
have
a
be
able
to
grow
and
not
need
to
Max
it
out,
because
if
we
do
go
to
and
we
get
the
full
three
million
we
spend
it,
then
there's
no
room
to
grow,
and
if
we
get
to
100-
and
you
know
1.5
million
and
spend
it
and
it
drops,
we
have
no
room
to
grow.
I
So
it
would
be
nice
to
in
our
in
our
relationship
with
AWS
to
ensure
that
we're
able
to
go
to
the
3
million
if,
if
needed,
and
if
we
can
get
there,
that's
eight,
but
we
need
to
bring
it
back
to
don't
use
it
or
lose
it.
It
sounds
like
that
would
be
unhelpful
in
the
long
run.
E
B
Sorry,
someone
at
the
door,
one
thing
that
I
might
do
for
the
next.
The
the
Sig
info
meeting,
not
just
the
AWS
one,
is
shift
the
order
of
what
we
like.
What
we
talk
about.
B
Sorry
Luna
is
going
crazy
because
someone's
here,
what
I
would
like
to
do
is
address
all
of
the
like
current
AIS,
that
we
talked
about
in
here
up
front
and
make
sure
that
there's
movement
and
if
there's
any
blockers,
identify
what
those
blockers
are.
A
B
C
B
That's
all
I'm
talking
about
like
okay,
hey
Justin,
what's
the
status
of
Porsche
and
artifacts?
Oh
still,
movement
cool!
All
right!
What's
the
status
of
the
eks
build
clusters?
Mario,
okay,
you've
got
the
scripts
going.
Do
you
have
any
blockers?
No
okay,
like
I'm,
talking
like
less
than
five
minutes,
but
I
still
wanna
shift
that
up
front.
So
it
kind
of
primes
the
pump
for
the
rest
of
the
meeting.
B
A
Okay,
I
have
no
hair
high,
so
I'm
fine,
with
whatever
you
say,
Jeffrey.
Okay,
thank
you
anything
else
before
I
think
I
think
we
can
call.
We
can
finish
the
meeting
now.