►
From YouTube: Foundational Infrastructure Working Group [May 12, 2022]
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
B
A
Yeah,
maybe
you
can
just
edit.
I
will.
A
I
actually
provided
here
a
regular
again
to
the
latest.
Changes
constantly
did,
but
I
didn't
get
any
feedback
here.
So
this
is
basically
writing
still
waiting
for
changes
and
constantine.
C
B
A
That
means
additional
changes
are
required.
D
C
So
yeah.
A
So
this
will
be
closed
soon,
should
we
bump
it
or
it's
waiting.
B
I
would
say
we
can
just
leave
it
in
this
state.
I'm
gonna
quickly
check
if
they
maybe
have
the
issue
open,
but
they
in
the
meantime
release
a
non-pre-release,
because
I
think
that's
the
blocker
right
we're
waiting
for.
D
That's
still
the
the
latest,
the
latest
yeah.
A
A
C
E
On
a
few
days
each
week
so
yeah
it's
still
in
the
pipeline,
but
it
will
take
some
time.
So
I
guess
I
can
remove
the
stale
from
time
to
time
until
it
gets
done.
B
I
learned
that
the
us
government
is
interested.
They
have
something
where
they
want
all
their
software
to
use
ip6
internally,
so
that
has
at
least
made.
I
mean
we're
still
scoping
that
out,
but
maybe
it's
interesting
for
to
look
at
bugs
around
ipv6,
like
that's
a
government-wide
mandate
or
something
from
a
while
back.
I
learned
it's
not
aware
of
that.
E
B
Yes,
so
I
I
learned
of
like
a
requirement
of
having
to
to
use
ipsec
we're
still
trying
to
figure
out
like
what
that
actually
means
like.
Is
it
till
the
load
balancer
or
is
it?
Do
you
also
have
to
go
after
the
load
balancer,
and
we
need
to
do
it
at
like
the
container
to
contain
a
network
level
but
yeah?
I
don't
know
yet.
A
Okay,
let's
move
on.
B
So
yeah
I
mean
this
cannot
be
merged
right.
This
will
break
bosch
dns.
So
it's
just
an
investigation.
Probably
it
will
be
closed
eventually.
B
E
Can
that
be
reviewed
by
one
of
us
or
should
better
be
reviewed
from
non-sap
forms
or
since
approved
it
already?
It
would
be
sufficient.
A
A
B
B
B
The
problem
is
it's
in
the
base
os
image
and
I
don't
like
conditionals
in
the
base
of
os
image
right,
because
the
idea
is
that
the
base
of
s
image
is
shared
across
all
the
stem
cells
right.
So
so
I
tried
looking
into
uninstalling
a
package
so
installing
the
kernel
and
then
in
a
later
stage
and
installing
the
kernel
and
installing
the
vips
kernel,
but
that's
not
pretty.
B
B
So
they
take
the
kernel
install
stage
out
of
the
base
os
image,
and
I
think
that
would
be
a
good
pattern
because
that
would
allow
us
to
maybe
if
we
want
install
ir
specific
kernels,
because
I
think
we
have
talked
about
this
before
right-
that
that
might
be
a
good
thing
to
do.
To
use
the
optimized
kernels.
B
E
I
remember
when
we
had
the
flappy
nick
issue
also
aws
support.
It
was
the
first
thing
to
ask
like:
can
you
use
an
aws
specific
kernel
instead
of
a
generic
one?
And
you
say:
oh
not
so
easy,
but
it's
one
of
the
questions
you
can
get
rid
of
immediately
right
if
you
have
a
yes,
a
specific
kernel.
So
yes,
besides
that
it
might
have
some
some
advantages.
We
don't
know
right
now.
E
B
C
B
It
does
break
a
bit
of
the
like
stem
like
a
lot
of
releases
just
test
one
stem
cell,
because
the
idea
is
that
all
stem
cells
are
similar
or
the
same
across
iss,
and
this
would
break
that
right.
So,
in
theory,
garden
should
start
testing
against
all
the
different
iss,
because
the
kernels
will
be
different
and
that's
the
component
that
is
integrating
with
the
kernel
more
or
less.
B
A
B
Yes,
so
that's
what
I'm
like,
but
I
for
vips.
I
have
to
go
that
route
anyway,
because
it
requires
a
different
kernel.
Also,
a
different
kernel
pair
is
even
so
yeah.
B
C
B
A
B
B
A
This
ppo
state
itself-
this
is
quite
old,
should
we
go
now
over
all
the.
B
B
A
Sounds
good
because
this
one,
for
example,
is
quite
old.
C
A
Yeah,
it's
four
five,
four
five
years
old,
yes,
yeah.
Let's
add
automation
to
this
new
rebels
and
yeah.
I
don't
see
anything
which
is.
B
D
A
D
B
C
B
It's
in
the
beginning,
so
the
problem
with
the
cpi
releases
with
bosch
create
m
is
that
the
manifesto
you
compile
the
cpis
once
on
your
host
machine
that
you're
deploying
from
and
then
you
also
compile
them
on
the
stem
cell,
the
same
release
to
put
on
the
director
yeah.
But
if
you
go
deploy
the
watch
with
abort
a
full
brush,
then
you
don't
have
that
problem.
D
D
B
In
the
top,
in
that
you
have
type
issues
area,
vm
deployment
life
cycle
like
that,
a
bit
yet
there,
and
after
that
you
can
start
typing.
B
Okay,
so
there's
nothing
in
the
inbox
yeah.
It's
nothing.
E
Right
and
ramon
is
also
a
way
and,
regarding
the
whole
stem
cell
publishing
topic,
we
decided
internally
to
pair
one
of
our
engineers
with
ramon
to
hopefully
release
the
stem
cell
next
week
and
then
also
be
able
to
look
into
a
few
things
in
his
absence
or
during
his
absence,
and
potentially
work
continue
to
work
with
him
afterwards,
and
one
aspect
is
that
he's
not
yet
part
of
the
cloud
foundry
organization
and
the
current
process
would
basically
be
to
create
a
pr
towards
this
contributor,
a
file.
E
E
B
E
E
B
E
I
will
do
both
and
let's
see
what
happens,
yeah
and
the
quest.
The
second
request
would
then
also
be
like
if
you
are
away
and
ramon
who
he
or
we
could
ping
in
case
like
we
need
a
quick
stem
cell
release
or
there's
some
kind
of
a
problem.
B
Yeah,
so
I
think
brian
connie
has
been
involved
with
the
gemmy
stem
cell
work,
so
he
and
also
joseph
palermo
and
constantine
I
mean
I,
I
think
all
of
those
people
should
have
access.
Okay,.
B
Yeah,
I
I
think
brian
county
has
the
most
context
on
like
at
least
that
whole
setup.
I
mean
it's
really
similar
to
what
we
have
internally
like
in
our
chrome
course,
because
it's
using
the
same
templates,
but
I
think
at
least
brian
connie
has
really
recently
got
the
gemmy
stem
cell
together
with
vermont,
and
that
should
be
the
same
as
the
bionic.
E
Okay,
yeah,
and
maybe
a
second
question,
it's
more
like
a
mid
long-term
question
is
so
we
currently
have
all
this
version
scanning
for
cbes
and
everything
and
for
most
golang
based
projects
on
the
vmware
side
of
things.
We
have
automated
bumping
right,
but
not
for
the
ruby
projects
like
bosch
and
a
few
cpi's.
I
guess-
and
I
was
just
wondering-
was-
would
that
be
at
some
point,
an
investment
area
or
if
no,
it's.
E
B
C
Of
the
manual
code
changes
for
bumping
things.
E
B
Yeah,
but
I
think
it
actually,
we
already
have
it
configured
in
some
places
that
I
I
did
work
to
exclude
all
those
bots.
So
this
view
that
we
are
looking
at
all
the
time
is
excluding
bots
because
they
are
a
bit
noisy.
I
don't
know
like
what.
How
would
you
want
to
go
around
about
doing
that?
Like
do
we
want
to
assign
people
to
review
those
bumps
or
dependency
bumps.
E
B
B
Yeah,
like
initially
my
thoughts
was
like
we
had
too
much
pr
or
too
many
pr's
too
much
work,
so
it
felt
like
yeah.
I
wanted
to
scope
it
down,
so
I
yeah
taking
those
out
of
scope
was
like
a
good
first
step,
but
I
mean
like
I
feel.
If
we
want
to
start
reviewing
all
of
those,
then
we
probably
need
more
reviewers.
B
E
Yeah
yeah,
so
for
me
it's
more
like
a
midterm
or
long-term
question,
as
I
said,
because
yeah
right
now
we're
bumping
a
few
things
here
and
there
and
potentially
this
go
on
and
go
on
over
a
few
months
like
we
would
have
this
discussion
like.
Should
we
move
it
somewhere
else
to
an
earlier
point
and
then,
of
course,
we
would
also
need
to
bring
the
people
who
review
it,
but
it's
the
same
work.
E
It
would
just
happen
a
little
bit
earlier,
not
only
after
the
fact
we
consumed
something
but
right
in
the
repository
right
and
just
wanted
to
yeah
get
some
feeling
about
like
what
would
be
options
in
the
future.
Yeah.
B
E
B
I
think-
and
I
mean
I
can
try
to
bring
in
some
people
from
vmware
that
that's
okay,
but
we
also
want
to
have
it
balanced.
I
think
that
was
like
one
of
the
things
that
we
want.
We
don't
want
to
it
to
be
like
a
vmware
only
party.
So
that's
why
I'm
a
bit
hesitant
to
do
that
at
this
point,.
B
E
Yeah
yeah
makes
a
lot
of
sense,
so
I
fully
understand
that
this
goes
together,
but
just
would
like
to
understand
options
for
how
to
go
forward
at
some
point
in
time.
So.
E
B
Yeah
first
step
would
be
to
get
someone
to
go
to
like
a
contribute,
get
contributor
rights
and
then
move
up
to
approver.
If
we
go
that
route,
I
think
it
will
be
fine
yeah.
Then
we
can
enable
it
in,
like
I
don't
know,
say
a
month
or
two
start
incorporating
that
in
the
review
process.
B
A
Please
have
an
option
manually
somehow
trigger
the
validation
yeah
at
the
beginning.
B
Yeah
sure,
or
we
could
like
a
cheaper
option
like
a
more
hybrid
approach,
would
be
to
maybe
have
some
github
actions
for
just
running
unit
tests
right.
That
would
already
catch
quite
a
lot
of
issues
like
at
least
the
most
basic
that
would
probably
like.
If
I
would
review
such
a
thing,
I
would
probably
start
running
the
unit
tests
myself
anyway
right.
B
So
if
we
can
automate
that-
and
I
mean
the
integration
test-
that's
gonna
be
difficult
right,
because
there
you
need
all
sorts
of
infrastructure,
but
the
unit
tests
usually
just
should
work
fine.
When
we
get
to
actions.
E
E
A
Okay,
unfortunately,
I
couldn't
update
the
notes,
but.