►
From YouTube: sig cluster lifecycle cluster api aws
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
A
B
Sure
so,
currently
we're
I'm,
mostly
trying
to
finish
up
like
the
routing
tables
m'dear
like
contributed
to
not
gateways,
then
with
the
Internet
gateways
and
now
like
throwing.
People
is
just
like
to
put
everything
together
after
this,
it's
mostly
around
it's
going
to
be
like
some
work
around
security
groups,
which
will
have
will
need
some
Doc's
around
it.
So
try
to
get
that
end
before
coat
it
out.
I
guess
just
to
get
some.
You
know,
okay
from
multiple
folks.
A
We
do
have
to
solve
some
of
the
problems
of
how
do
we
get
data
out
of
the
cluster
specifically
around
the
bootstrapping
information,
CA
certificates
and
then
cube
config,
potentially
bootstrap
tokens
and
all
of
that
stuff?
One
things
I
wanted
to
try
to
avoid
doing
which
has
been
previously
done
with
most
of
the
provider
implementation
so
far,
I
want
to
try
to
avoid
storing
a
set,
a
separate
set
of
SSH
credentials
that
has
the
closed
or
the
machine
actuator.
Actually
SS
aging
into
the
remote
instance
and
performing
operations
in
that
manner.
A
Yet
if
the
benefits
of
it
and
the
ability
to
run
commands
remotely
outweighs
kind
of
the
negative
aspects
of
those
additional
requirements
that
it
puts
on
there,
Wickham
is
teeny
to
run
well.
So
the
big
thing
would
be
is
how
do
we
validate
that
the
cloud
in
it
module
has
completed
either
successfully
or
failed,
and
then
how
do
we
get?
For
example,
like
the
admin
cube
config
from
the
install
done
and
and
kind
of
the
next
point
I
have
on?
A
So,
for
example,
we
can
just
pre
generate
kind
of
those
secrets
that
we
need
within
the
machine
actuator
or
the
machine
actuator
for
the
initial
control,
plain
instance
and
kind
of,
or
even
within
the
cluster
actuator
itself,
and
just
kind
of
inject,
the
needed
secrets
into
the
cloud
and
it
config,
and
then
we
can
just
kind
of
do
a
naive
pulling
of
the
instance.
You
know,
for
example,
for
the
control
plan
instance.
A
D
I'm,
some
of
my
colleagues
have
done
a
lot
with
SSN,
so
I
can
play
find
out.
What
is
I
would
imagine
this
for
you
key
things.
You
would
need
to
bootstrap
in
terms
of
like
training,
mum
thoughts
and
things
possibly
might
be
a
bit
annoying
I
think
I
would
definitely
instead
of
using
s3
is
parameter,
store
to
store
those
secrets
and
we
can
scope.
D
The
I
am
policy
to
maybe
some
prefix
that
that
we
use,
like
that
looks
like
they're
tagging,
had
standard
tags
that
we've
that
you've
written
in
document
around
how
we
do
item
pregnancy,
so
I
think
that's,
probably
that's
currently
best
practice.
If
you
look
at
the
web
architected
framework,
that's
what
I
did
yes
recommend
these
days.
All.
D
A
A
That
was
my
thought
like
having
a
manager.
An
extra
set
of
SSH
credentials
would
even
complicate
like
remote
access
to
the
machines
for
an
admin,
and
things
like
that.
So
I
was
trying
to
avoid
that,
and
we
don't
necessarily
want
to
take
just
user
credentials
and
put
them
into
the
actuator
and
have
them.
You
know
be
like
the
long-running
kind
of
requirement
as
well.
Yeah.