►
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
Yes,
all
right
welcome
to
the
February
6
2023
cluster
API
AWS
infra
provider
meeting
just
you
know
that
I
guess
says.
As
we
always
say,
when
we
start
off,
we
are
a
kubernetes,
six
cncf,
sorry
yeah
project
and
we
follow
the
cncf
code
of
conduct,
which
basically
means
be
excellent
to
one
another
and
please,
if
you
are
attending
the
meeting,
add
yourself
to
the
attendees
list.
If
you've
got
any
items
for
the
agenda,
please
yeah,
please
add
them
to
the
meeting
notes.
A
We
sometimes
have
PSAs.
Today
we
don't
and
yeah
this.
This
might
be.
This
might
be
a
brief
one
also.
If
anybody
is
new
here
and
you
would
like
to
introduce
yourself-
we
always
leave
a
little
bit
of
time
in
the
beginning,
so
I'll
mute
myself
for
a
minute
and
feel
free.
If
you'd
like
to
introduce
yourself.
B
Hey
I
can
jump
in
real,
quick,
so
I'm
Tim,
Gilcher
I'm
on
the
team
with
Mike
tujeron
I,
believe
you
know
him
from
previous
meetings.
He
couldn't
join
today.
I've
been
doing
some
work
on
the
same
team
to
do
some
Gap
analysis
for
the
Kappa
project
as
compared
to
our
own
internal,
tooling,
so
very
interested
in
in
getting
more
involved
and
so
far
been
quite
impressed
with
what
I've
seen
so
yeah.
That's
me
good
to
be
chill.
C
And
then
I'll
go
next,
so
I'm
Andreas
from
transform
just
started
working
there
at
this
I
finished
kubernetes
company
in
December
and
I
straight
got
into
a
lot
of
work
on
cluster
API
and
Kappa,
specifically
and
yeah,
while
upgrading
to
2.x
for
a
couple.
I
have
already
filed
a
maybe
five
cars
last
week,
so
quite
involved
right
now
in
things
so
just
figured
out
my
join
here
to
see
what's
going
on.
Otherwise
so
hi.
A
A
So
yeah
I
think
at
this
point,
since
we
don't
have
any
PSAs.
If,
if
anyone
has
anything
they'd
like
to
talk
about
or
any
questions
Tim,
if
you've
got,
you
know
if
you're
comparing
internal
tooling
to
Kappa,
we
would
love
to
to
hear
you
know
some
of
your
thoughts.
If,
if
you
know
if
it,
if
it's
a
full-fledged
or
feature
request,
we'd
love
a
PR
or
or
a
bug,
you
know
if
you
find
some
issue
or
something
that
doesn't
behave
as
expected.
A
A
A
B
So
so,
if
it's
okay,
I'd
I'd
go
first,
just
because
I
don't
think
I
have
as
much
perhaps
to
talk
through
mainly
at
this
point
you
know
some
of
the
initial
stuff
is
done.
I'm
going
to
be
doing
deeper,
diving
later
on,
so
I
may
have
some
follow-ups
in
future
meetings,
but
it
seems
like
I'm
finding
equivalents
for
pretty
much
everything
that
we've
been
doing
to
operate
our
clusters
over
the
last
five
years.
B
So
that's
really
encouraging
I
think
the
biggest
thing
that
I
haven't
really
gotten
a
good
answer
from
the
documentation
on
yet
is
just
how
viable
and
and
how
supportable
is
a
migration,
a
Brownfield
migration
of
clusters
to
be
managed
by
Kappa.
It
seems
like
there's.
You
know
some
Alpha
functionality
around
that
any
any
sort
of
insight
into
how
common
that
is,
and
any
things
we'd
want
to
be
aware
of,
would
be
super
helpful.
C
B
Okay,
that's
helpful
yeah.
We
have
several
dozen
clusters
that
are
either
in
AWS
or
eks,
we're
migrating
currently
away
from
AWS
to
the
eks
platform
and
so
I.
You
know
I
initial
conversation
I
had
with
my
management.
They
were.
They
were
kind
of
Eyes,
Wide,
Open
thinking
we
might
be
doing
migrations
again
so
yeah.
Hopefully
we
can
just
adopt
our
existing
eks
clusters.
Aws
clusters,
I
think,
would
just
be
going
away
from
our
side.
That's
at
all
helpful.
A
I've
I've
also
have
you
know,
I
guess,
maybe
some
encouraging
news
to
share
on
you
know
on
that
use
case.
So
we,
our
the
previous
version
of
our
say
life
cycle
stack,
was
not
cluster
API
based
and
we
did
migrate
our
users
to
to
Kappa
well
for
for
the
AWS
provider,
and
one
I
would
say
the
I
I
would
say
the
the
biggest
constraint
is
I.
Think
the
the
topology
you
know
depends
on
how
you
have
your
you
know
your
VPC
setup,
your
security
groups.
A
There
are,
there
are
so
many
ways
to
slice
and
dice
to
get
equivalent
functionality
right,
but
it's
it.
There
can
be
some
issues
with
migrating
to
Kappa,
because
Kappa
may,
you
know,
expect
to
manage.
Let's
say
you
know,
security
security
groups
in
a
particular
way
or,
or
you
know,
established
security
groups
for
certain
purposes,
whereas
you
maybe
you
were
using
a
single
Security
Group
and
had
Security
Group
rules,
sort
of
for
those
same
purposes.
A
There's
also,
there
may
be
an
issue
of
migrating
and
how
your
load
balancing
is
is
set
up
and
that's
actually
something
that
we
found
during
our
migration
and
we
upstreamed
some
functionality
to
adopt
an
existing
control.
Plane
load
balancer.
So
that's
that's
there
and
that
that
that
that
was
essential
for
for
our
migration
use
case.
But
it's
definitely
it
it.
It's
definitely
not
not
only
possible,
but
you
know
I
would
say
successful
for
us
so
yeah.
A
B
A
About
at
the
meetings
you
know,
we
also
have
our
slack
channel
in
the
kubernetes
slack.
We
could
so
so
you
know
your
experiences.
You're,
not
I,
think
you're
not
finding
a
lot
of
information
in
the
docs
about
this,
and
if
we
can
work
together
and
get
some
of
this
in
the
docs,
you
know
it
would
yeah
it
would
help
it
would.
It
would
help
you
know,
sort
of
everybody
in
the
in
the
community
and
also
just
you
could
probably
figure
out.
A
C
I
can
quickly
talk
about
stuff
I
worked
on
last
week,
so
PR
is
almost
emerged
and
we're
continuing
was
basically
about
upgrade
issues.
I
have
because
so
I
tried
to
upgrade
from
Kappa
wondered
something
to
two
dot,
something
and
basically
took
the
commit
from
the
main
branch,
because
we
will
have
our
own
Fork.
So
we
want
the
latest,
but
I
took
the
crd
from
the
latest
release,
so
those
were
mismatching
and
I
ran
into
issues
because
of
that
not
a
biggie
to
talk
about
here.
C
So
the
managed
infrastructure
that's
broken
now
because
of
a
totally
unrelated
change,
and
we
talked
to
Richard
case,
for
example,
and
we
set
up
a
draft
document
for
basically
a
bigger
topic,
namely
rethinking
the
networking
code
in
Kappa.
So
this
might
be
something
bigger
where
we
as
an
example
might
introduce
a
separate
object,
AWS
subnet,
which
makes
it
easier
to
to
manage
stuff.
C
So
that's
what
I
yeah
we'll
work
on
next,
so
bringing
that
use
case
back,
because
many
users
want
to
not
bring
their
own
but
really
have
couple
manage
everything
like
ppcs,
subnets
security
groups
and
so
on
and
there
we
need
to
ensure
that
the
code
really
supports
this
and
doesn't
confuse
stuff.
So
yeah
I.
Guess
that's
enough
of
a
summary
just
for
you
to
know
that.
There's
an
issue
here,
I
think
it's
already
in
2.0.0,
so
it
is
released
this
problem,
so
people
have
an
issue
with
this
and
yeah.
C
A
C
Well,
for
my
company,
I
cannot
tell
okay.
I
cannot
tell
because
I'm
here
for
two
months
just
started
so
I,
don't
know
what
how
their
migration
or
our
expectations
were
for
for
switching
versions,
but
it
was
clearly
a
freaking
change.
So
there
was
a
PR
for
something
else
which
made
the
subnet
ID
field
required,
which
clearly
then
rules
out
of
this
case
where
Kappa
kind
of
managed
stuff.
C
But
the
code
is
still
there
to
to
manage
subnet
right,
so
it
was
kind
of
removed,
I'm
mistake
or
in
a
sense
and
I
think
what
copy
and
Kappa
overall
should
do
is
to
clearly
note
down
what
use
cases
are
supported
because
yeah
we
had
a
meeting
on
that
and
people
said:
yeah,
copper
and
copy
were
pretty
opinionated
in
the
beginning.
So
there
was
one
way
to
create
your
networking
and
stuff
and
then,
as
companies
joined
and
used
it
more
of
course,
as
always,
they
will
have
their
own
requirement.
C
They
want
customization,
and
then
you
get
so
many
use
cases
that
it's
hard
to
see,
which
ones
actually
supported
and
for
the
one
we
talk
about
here.
So
manage.net
I
found
that
there
is
not
even
a
unit
test
a
unit
test
for
animalist,
but
that
for
managed
and
yeah.
So
then
it's
hard
to
see
as
an
author
of
a
totally
unrelated
PR
do
I
break
something
or
not
so
I
think
that's
what
we
should
do
now,
while
authoring
changes.
A
Yeah
I
think
I
wonder
if
this
might
have
been
compounded
by
the
the
managed
yeah
the
managed
clusters
eks.
Oh,
that
was
in
an
experimental
I
guess
you
could
yeah
it
was.
It
was
experimental
for
a
long
time
and
it
was
in
a
separate,
separate
API,
and
so
there
there
may
have
been
yeah.
It
may
have
been
hard
to
see
some
of
the
interaction
or
or,
for
example,
you
know
lack
of
lack
of
testing
for
for
some
use
case.
A
D
Oh
hi,
this
is
Amit
Kumar
I
am
from
India.
I
am
a
negative
instrument,
so,
like
I
am
currently
volunteering,
Cappy
version
1.4
releasing
team,
so
there
I
have
a
related
Shadow
role
so,
like
I
have
just
applied
for
LFX
mentorship
in
Kappa,
so
like
I
want
today
so
like
I,
just
trying
to
upload
my
first
cluster
on
AWS
plus
more
like
AWS
cloud,
so
like
I,
found
that
in
getting
started,
there
were
many
links
broken
like
I've.
Just
like
left
a
link
in
the
chat.
A
A
Okay,
well,
if
you
can,
you
know
I,
don't
know
how
long
how
how
far
you've
gotten
with
you
know
with
clothing
the
repository.
But
if
these,
if
these
links
are,
you
know,
broken
in
the
in
the
cluster
API
AWS
book,
so
so
you
know
the
documentation,
then
it'll
all
be
in
the
in
the
repo
and
so
I
think
I
think
the
description
you
should
be
able
to
find
some
of
these.