►
From YouTube: Cloud Foundry for Kubernetes SIG [February 18, 2020]
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
A
A
There
is
one
additional
topic
that
was
brought
up
up
front,
I,
think
by
Sai,
and
that
is
we
initially
established
the
slack
channel
cf4
kate's
for
this
specific
signaling,
and
by
coincidence
or
not,
there
is
now
a
github
artifact,
called
by
the
same
name,
and
that
makes
people
join
this
lecture
hall.
If
they
want
to
talk
about
the
artifact
and
I
guess
we
wanted
to
raise
it
up
for
discussion
like
if
that's
a
good
thing
or
a
bad
thing.
A
If
you
want
to
keep
it
like
this,
that
we
have
maybe
zeke
related
discussions
and
like
github
artifact,
created
discussions
in
the
same
channel.
If
that
makes
sense
for
everyone
or
if
it
doesn't
yeah,
how
does
everybody
feel
about
this
site
since
you
brought
it
up?
Do
you
have
a
specific
opinion
on
this?
Actually.
C
D
And
I
think
my
my
expected
objection
is
that
this
might
lead
people
to
not
know
about
cube
CF,
that's
sort
of
what
we
would
generally
understand
the
thing
to
be
that
we've
overloaded
the
name
of
this
channel,
but
it's
not
a
strong
objection.
I
mean
I,
think
we
all
know
who
each
other
are
and
know
where
the
projects
are.
So
we
can
direct
people
who
might
be
seeking
one
or
the
other
I'm,
not
not
too
bothered
by
it.
You
can.
E
F
Was
just
going
to
link
a
post
from
from
Fenton
Brian,
who
posted
yesterday
a
little
bit
confused
about
fissile
and
its
status
and
I
tried
to
clarify.
But
his
point
in
the
end
is
yeah.
It's
a
little
difficult
to
discover
from
from
the
website
like
fissile
became
quirks
and
lives
in
the
CF
operator
repo
and
is
related
to
cube
CF
in
the
certain
way.
Yeah.
E
Yeah
we
yeah
yeah
I
would
I
would
like
to
say
that
I'm
also
like
like
I.
If
anyone
should
know
this,
it
should
be
me
and
I've
lost
track
of
what
the
various
nouns
in
that
world
me
now
so
yeah
well,
I
had
I
had
a
similar
conversation
in
the
day
and
I
realized
that
I
don't
even
know,
which
is
the
correct
noun
noun
for
things
now.
D
D
The
quarks
project
has
included
fissile
I
mean
that
it's
the
same
team
that
maintains
it,
but
it's
actually
more
closely
bound
to
keep
CF
and
keep
CF
won't
be
living
in
the
Boche
pmc,
and
that
this
is
the
problem
right
because
we've,
as
we
discussed
in
the
last
meeting,
the
CF
operator
is
more
of
a
Bosch
operator,
which
is
why
it
belongs
in
the
Bosch
at
PMC
and
cube.
Cf
belongs
in
run
time,
which
is
where
it's
gonna
hopefully
go
for
incubation
soon.
We've.
E
If
the
CF
operator
is
in
fact
the
budget
operator,
we
might
want
to
get
ahead
of
the
ball
renaming.
Some
of
that
I
mean
I,
think
I'm
at
the
risk
of
creating
even
more
names,
which
is
not
something
I
want
to
do
by
then.
Maybe
we
need
to
think
this
I
mean
I,
I,
nominate,
Roy,
but
I
think.
Maybe
we
just
think
to
think
through
this
pick
three
leaves,
or
you
pick
the
minimum
amount
of
names
that
covers
all
the
things
and
redirect
other
stuff
so
adducing
this
inspire
I
kind
of
agreed.
E
F
A
From
the
official
website,
if
I'm
looking
for
Cloud
Foundry
coronaries,
what
is
the
thing?
I
should
look
at
and
I'm,
not
sure
who
mind
can
or
should
be
taking
this,
but
this
seems
like
a
thing
we
should
fix.
G
G
I
was
also
going
to
reach
out
to
you
en
Baron
to
see
if
they
would
be
able
to
summarize
all
the
efforts
going
on
in
the
CF
and
kubernetes
front,
maybe
in
a
block
was
kind
of
thing
that
we
can
use
as
an
anchor
to
point
people
to
different
places
in
the
in
the
website,
because
right
now
there
is
I
mean
we.
There
are
pieces
of
information
in
different
places
and
for
us
to
have
some
kind
of
a
meaningful
story.
I
think
you
need
some
kind
of
anchor.
G
F
D
Pretty
high
confidence
that
in
this
meeting,
we
have
a
we're
starting
to
come
to
a
consensus
on
on
how
it
is,
and
you
a
and
Eric
and
I
have
had
side-channel
discussion
with
Jamie
Gomez,
so
I
think
we're
on
the
same
page
but
yeah.
If
we
could
start
a
Google
Doc
discussion
that
turns
into
a
blog
post.
That
will
tell
people
where
to
find
things
in
that.
From
that
guidance,
Swarna,
the
maybe
the
quarks
content
and
the
Irani
content
could
be
updated
on
the
website
to
point
to
the
right
places.
D
D
Let
me
just
double
check
that
that
my
understanding
is
the
same,
that
we
have
CF
on
K,
it's
a
project
maintained
by
the
relevant
team,
and
it
is
new
and
experimental.
We
have
a
cube,
CF
repository
that
is
going
to
go
for
incubation
in
the
run
time
pmc,
which
is
parallel,
work
and
hopefully
cross
pollinating.
We
have
the
quarks
work
which
is
under
the
bosch
pmc,
and
that
is
the
what
will
be
renamed.
The
bosch
operator
figure
out
the
best
way
to
do
that,
and
then
that
leaves
fissile,
which
is
an
implementation
portion
of
it.
D
That's
how,
if
my
that's,
how
cube
CF
is
built,
so
that's
the
part
that
I
need
to
sync
with
Ladd
and
see
where
he
feels
that
belongs
or
what
the
best
way
is
to
describe
that,
so
that
people
know
that
it's
still
part
of
the
part
of
that
ecosystem.
But
it's
specific
job
is
more
limited
because
I
quite
often
hear
people
talking
about
fifth
aisle,
as
if
it's
Sousa
Cloud
Foundry,
like
the
previous
version
of
cube
CF,
and
that
is
a
misunderstanding
of
what
it
is
actually
doing.
So
that
part
has
to
be
clarified.
D
E
E
It
feels
know
that
the
problem
isn't
so
much
I
mean
we
can
explain
all
these
nouns
are,
and
the
document
sounds
like
a
good
idea,
but
I
feel
like
the
missing
thing
is
like
an
entry
point
like
what
is
the
entry
point
that
gives
you
CF
on
Cuba,
Nettie's
and
I
feel
like
there's
a
really
well
named
entry
point
which
is
CF
for
kubernetes,
and
it's
kind
of
a
shame
that
it
refers
to
one
of
them.
Not
all
of
them.
E
I
feel
like
that.
Repo
is
kind
of
the
entry
point
and
the
fact
that,
right
now,
that
repo
is
actually
I
mean
what
were
what
we
have
at
the
moment.
Right
is
a
thing
called
CF
for
kubernetes,
which
is
not
in
fact
the
entry
point
to
getting
CS
for
Cuba
nettings.
It
is
an
experimental
collection
of
Bosch
releases
that
currently
won't
even
do
CF
push
end
to
end.
E
If
you
see
what
I
mean
all
right,
so
as
a
user,
if
I
go
to
Cloud
Foundry
/cs
for
Cuba
Nettie's,
that
isn't
actually
the
correct
entry
point
right
now
and
I
kind
of
feel
like
it
should
be
a
correct
entry
point:
whether
that
means
that
CF
or
kubernetes
the
current
thing
should
be
elsewhere
or
whether
it
just
means
that
that
readme
should
link
to
the
other
things
as
well.
I
don't
know,
but
if
that
feels
weird
to
me
right
now
alone
is
like
Seattle
cubing
it
like
housing
house.
D
We
could,
in
the
readme
say
this
is
experimental
work
that
is
currently
undergoing
if
you're
looking
for
some
CF
for
kubernetes
that
you
can
deploy
now,
here's
cube
C
F,
and
this
is
pointing
to
something
which
will
soon
be
better
than
it
is
right
now
in
terms
of
documenting
how
things
how
to
get
going.
So
we
have
a
CF
for
cube,
CF,
Docs
repo,
which
is
getting
closer
to
completion,
which
would
be
a.
D
E
I
mean
maybe
maybe
all
it
would
honestly
take
is
like,
if
all
of
the
various
things
had
a
link
at
the
front
of
the
readme,
that
link
to
a
common
doc
sites
which
is
like
here
are
the
dogs
for
CF
on
kubernetes
and
that
had
that
document
we
were
discussing
it.
Maybe
that's
enough,
maybe
we
just
like,
maybe
maybe
all
the
various
places
like
arena
repo
fits
our
repo
cork
sweeper.
E
All
these
things
should
have
a
link
at
the
start
of
their
readme,
and
that
says
here
are
the
docs
for
cloud
foundry
on
kubernetes
right
like
and
and
then
the
rest
of
things
for
that
repo
right,
but
for
people
who
hit
any
of
these
places,
we
need
to
like
get
them.
Two
to
one
consistent
said:
you
know:
cs4
kubernetes,
dot,
io,
sort
of
thing,
yeah.
F
I
H
H
H
B
A
All
right
cool,
then
we
have
the
container
to
container
networking
I'm,
not
sure
who
brought
this
topic
up
initially,
because
there
is
no
handle
behind
the
topic
on
slack.
If
that
person
is
here,
maybe
you
want
to
say
one
two
sentences
about
it.
What
made
you
put
that
up
on
the
agenda.
A
H
E
A
E
E
There's
this
various
ways
you
can
try
to
do
it,
but
to
me
they
eat
the
the
simplest.
You
just
run
these
things
side
by
side
right,
not
like
try
sharing
databases
showing
all
that
stuff's
just
have
two
of
them,
but
then
you
run
into
the
problems
of
networking
between
the
two
and
that's
that's
why
I
was
going
because
it
becomes
difficult.
So
if
we
could
easily
get
conscious
turnout
looking
between
the
two,
it
just
makes
migration
and
staged
migration.
E
A
slow,
slow
migration
doing
something
similar
to
we
did
what
we
did
with
das
in
Diego,
where
you
can
kind
of
have
them
bugs
running
at
the
same
time,
and
slowly
move
apps
over
before
you
shut
things
off,
make
sure
that
a
lot
easier
then
trying
to
sort
of
like
shut
one
down
stall.
One
up.
It's
like
way,
too
scary
to
do
at
scale
so.
H
E
You
see
whatever
you
want
in
this
topic.
Apparently
I
mean
I
guess
it
could
also
mean
just
like
I
mean
one
big
missing
piece
from
ireenie
right
now
is
the
container
networking
stuff
right?
That's
that's
one
thing
that
people
moving
from
one
to
another
will
carry
any
heavy
edition
experience,
partly
because
you
know
kubernetes
already
has
overlays
and
stuff.
So
what
is
containing
that
working
exactly
mean
in
that
world?
It's
not
under
simply
fear
that
that
is
definitely
another
variable
topics.
I
would
love
it
if
someone
had
solved
and
would
happily
what
should
their
more.
A
H
Don't
I
don't
say
anyone
from
the
networking
team
represented,
but
I
think
Shannon
and
Gabe
would
be
the
best
equipped
to
weigh
in
even
on
issues
of
feature
parity
like
having
things
like
AB
security
groups
or
container
network
policy.
Connectivity
should
be
there
across
the
cluster,
except
for
those
those
rude
ireenie
network
policies
that
drop
in
and
lock
everything
out.
E
E
H
So
potentially
have
the
expectation
of
the
pool
closing
on
a
staggered
schedule.
If
we're
going
to
continue
to
coordinator
on
that,
just
have
a
close
week
before
and
or
so,
and
then
try
and
drum
up
whoever's
going
to
represent
that
topic
or
find
out
that
we
don't
have
anyone
but
I,
think
that's
that's
a
good
idea
for
this
topic.
We
can
also
figure
out.
You
asked
that
question
so
that
we
know
what
they
were
actually
referring
to
since
consistent
has
inconsistent
understanding
across
all
of
our
minds.
A
H
H
D
Can't
believe
we
have
a
shortage
of
things
to
talk
about
Eric,
we're
gonna
talk
about
incubation
in
the
run
time
PMC
later
and
our
time
so
just
to
let
everyone
know
that
we
we
thought
we
were
gonna
skip
a
step
with
cube
CF
and
just
like
put
it
right
in
the
foundation.
We
redirected
back
to
the
process
and
the
process
is
incubated,
it's
discussed,
and
so
the.
D
D
Pmc
meetings,
so
we'll
just
give
people
some
time
to
take
a
look
at
that
it
will
go
into
incubation
the
reason
one
of
the
reasons
I
was
pushing
so
hard
for
it
to
get
in
quickly
because
I'd
love
to
have
something
to
talk
about
suza
cons
coming
up
and
I'll,
be
there
and
I
was
wanting
to
have
something
foundation
related
to
talk
about
there,
and
so
we
can
talk
about
the
incubation.
Hopefully,
if
all
goes
well,
we
could
say
cube.
Cf
was
accepted
for
incubation.
D
I
can
make
a
bit
of
a
noise
about
that
raise
its
profile,
raise
the
profile
of
CF
working
on
kubernetes
a
little
bit
helps
with
my
product,
but
it
also
helps
I
think
with
the
foundation
in
the
community's
effort
and
then
hopefully,
we'll
see
a
nod
your
head.
If
the
timeline
sounds
right,
we
if
we
have
it
in
the
incubation
for
a
short
time,
we
may
be
able
to
announce
if
it
goes
well.
Maybe
I'm
able
to
announce
graduation
by
CFS
I'm
at
Austin
question
mark
not
vigorous,
not
it
but
yeah.
H
Yeah
and
I,
just
in
terms
of
just
closing
up
the
incubation
proposal
relatively
quickly,
I
reminded
folks
in
the
runtime
PMC
to
weigh
in
on
it
and
let
them
know
that
I
kick
off
an
out-of-band
approval
request
later
this
week,
so
I
think
on
the
20th
and
leave
that
open
for
a
couple
of
days,
business
days
for
people
to
comment,
but
that
should
be
resolved
by
really
next
week.
That's
great.
D
So
yeah,
that's
thank
you
very
much.
Yui
and
yeah
thanks
ray,
unfortunately,
I
don't
see,
Jamie
Gomes
on
the
call
I
could
kick
him
to
haven't
joined,
but
maybe
that
could
be
another
topic
for
next
meeting
and
we
can
have
maybe
a
little
bit
of
time
to
prepare
for
it.
The
idea
that
was
announced
was
at
last
last
meeting
for
a
technical
program
manager.
D
D
D
C
So
there
has
been
some
feedback,
but
right
now,
I'm
just
migrated
to
the
to
the
github
repo.
So
it's
now
you
know
God
font,
incubator
kind
of
like
owned
by
a
drinking,
but
not
really
because
it's
diverse
I
don't
know
if
I
really
will
own
it,
but
yeah
I,
basically
copy-paste
him
expanding
I
did
a
bit
more
references
to
the
document.
C
A
B
Marco
so
I
think
primary
today,
as
of
today,
a
primary
objective
is
to
integrate
build
packs,
so
we
can
have
a
full
C
apush
experience,
so
we
are
working
actively
with
Cappy
for
on
Capac
integration,
we're
also
working
with
the
logging
and
metrics
team
to
incorporate
the
logging
capability
into
Steve.
Okay,
so
those
are
those
are
the
two
primary
set
of
tracks.
We
were
working
on
right
now.
We
also
have
our
first
public
pipeline.
Don't
know
what
you
had
a
chance
to
take
a
look
at
it.
B
B
We
would
like
to
fast
follow
that
with
additionally
with
building
of
a
full
integration
pipeline
with
the
teams
so
established
in
the
contract
between
us
and
the
contributing
teams
on
how
we
would
build
the
release,
safe,
consistent
components,
release
and
how
do
we
ship
those?
So
those
are
those
are
sort
of
the
next
set
of
milestones.
People
intend
to
to
work
on.
D
D
F
D
Think
their
roles
are
different
like
it,
it
wouldn't
necessarily
be
in
CN,
CF
incubator,
or
something
like
that.
As
long
as
it's
under
the
the
cloud
native
build
PACs
governance
as
part
of
CN
CF,
that's
fine,
and
we
can
pull
that
in
it's
just
if
we
start
using
that.
We
just
need
to
be
sure
that
it's
not
yeah.
D
A
I've
one
more
thing
around
the
lock
regulator:
artifact
integration,
that,
as
I
mentioned
before
I,
think
triggered
by
one
of
the
mergers
there
I
was
wondering
if
we
could
like,
given
that
you
have
a
pipeline
now
have,
in
that
repository
at
least
some
indication
for
which
versions
of
kinetise.
This
whole
thing
is
supposed
to
work
and
for
which
it
is
like
actively
tested
or
something
right,
because
there
was
some
issues
with
116
because
of
the
removal
of
deprecated
kinds
and
so
on.
So
this
this
would
be
highly
appreciated.
A
B
I
Another
good
idea
might
be
to
activate
configuration,
although
that
will
make
the
test
go
red,
but
like
there's
at
least
one
Network
policy
coming
from
from
ireenie
that
that
doesn't
work,
and
it
only
shows
up.
B
I
I
Just
like
we
just
made
it
green
for
us
without
all
the
activating
network
policies,
probably
like,
as
I
suggested
on
slack,
that
we
should
probably
talk
about
who
is
actually
responsible
right
for
these
network
policies
only
going
to
be
either
see
if
FERC
aids
as
the
aggregating
thing
or
CF
Cades
networking
as
the
networking
thing,
but
I'm
not
sure.
Let's
have
a
discussion
on
that.
Yeah.
I
J
Here
for
gates
I've
been
wondering:
do
we
have
a
model
yet
for
the
kinds
of
things
that
used
to
be
optional
in
CF
deployment,
so
in
other
words,
you
have
to
point
in
how
to
hold
bag
of
ops
files
that
you
could
pick
and
choose
from?
Are
we
gonna
repeat
that
combinatorial
explosion,
or
are
we
gonna,
keep
those
elsewhere
or
do
we
have
opinions
about
that?.
I
J
It
isn't
currently
included
in
in
CF
or
Kate's,
so
that
needs
to
be
added
and
then
once
you've
added
that
that's
just
an
extension
point,
so
you
knew
the
other
bits
that
actually
turn
on
a
volume
service.
So
in
our
case
that's
like
an
SMB,
CSI
plug-in
and
service
broker
yeah.
So
all
that
stuff
isn't
necessarily
stuff
that
everybody
wants,
but
some
people
want
to,
and
so
we
used
to
do
that
without
spots
and
now
there's
a
ytt
version
of
an
ops
file
that
we
could
just
kind
of
pile
on
in
there.
E
B
As
I
Julien
to
your
question,
we
have
not
yet
explored
that
our
goal
right
now.
The
short-term
goal
is
to
get
the
K
back
integration
on
and
then
nothing
fast
fall
over
it
sort
of
building
opinions
on
that
lessons
learned
from
CF
deployment.
What
not
to
do
well,
we
can
carry
the
good
things,
but
no
not
yet
you
know
there
yet.
Okay.
J
F
J
Thank
you
yeah
some,
you
know.
Minor
configuration
stumbles
around
kubernetes
versions
in
that,
but
once
you
get
it
working,
it's
very
quick.
It's
very
refreshing,
a
suicide
said
it
starts
up
in
four
minutes.
I
got
it
running
on
kind
on
my
on
my
laptop
with
some
minor
tweaks
that
was
cool,
including
volume
services.
So
ya
know.
I
I
E
I
I
That
way,
not
no
one
has
to
really
worry
about
certificates
and
whatnot
and
know
that
configuration
yeah,
which
potentially
would
work
out
also
delayed
for
local
development,
because
you
know
we,
you
don't
have
to
run
we
still
as
locally.
Let's
say
right.
So
as
long
as
we
were
able
to
tease
that
apart
nicely
and
drop
all
of
the
certificate
management
out
of
the
components.
E
Yeah
I
was
just
gonna,
say:
I
think
it
would
be
like
I
mean
if
we
could
get
like
a
nice
dogs
page
that
shows
people
how
to
get
started
on
keeping
at
ease
in
Dhaka
on
kinds
with
the
CF.
That
would
just
be
insanely
great
big,
huge,
like
stand
this
up
on
your
laptop
just
need
docker
that'd,
be
pretty
great.
J
Yeah
have
to
make
some
modifications
like,
for
example,
even
well
before
I
guess:
I
must
have
got
it
before
things
scaled
up,
but
you
know
vo
has
a
load
balancer
on
the
front.
End
and
kind
doesn't
support
that
stuff
like
that,
so
we
would
have
to
guess
decide
that
that's
a
priority
to
make
to
make
that
a
target,
but
I
agree
just
also
for
developers
able
to
being
able
to
just
spin
stuff
up
yeah.