►
From YouTube: Foundational Infrastructure Working Group [July 6, 2023]
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
Okay:
okay,
let's
start
supporting
instance
group
text,
what
is
this?
We
need
a
value
here,
I
already
rebuilt
this
one
I
think.
A
A
It's
a
good
question
just
a
second
weeks
ago.
No,
not
yet.
B
Okay,
then,
it's
probably
gonna
be
with
the
next
release,
which
or
it's
going
to
be
with
the
next
release,
but
I
think
the
they're
gonna
probably
want
to
cut
a
release
once
the
Bosch
recover
stuff
gets
in.
A
Okay,
so
into
the
documentation
we
updated,
the
expected
version
actually
is
yeah
exactly
we
updated
that
but
of
course
not
yet
released
yeah.
So
we
cannot
merge
that
under
the
other
one.
B
A
So
the
discussion
was
about
whether
we
need
integration
tests
here
or
not.
I
see
that
Ruben
you
are
approved.
A
But
should
we
add
an
integration
test
here,
I
didn't
get.
Any
feedback
goes
from
yourself.
B
A
Scroll
yeah,
I
I,
described
here.
What
can
be
I
describe
how
an
integration
test
could
be
a
way
it
makes
sense.
I
walked
into
this
described
that
I
think
it
associate.
This
test
can
be
extended
with.
B
I
mean
yeah
I
agree,
so
maybe
you
can
just
a
request
changes.
Then.
Then,
it's
like
more
formal
that
we
would
like
to
have
that
added.
A
A
B
Yeah
I
believe
this
is
because
of
the
the
move
I
mean
we.
This
is
the
Bosch
packages,
cfcli
thing
and
that's
now
added
because
it's
in
the
working
group
in
the
cloud
fund,
New
York,
but
we
I
failed
to
be
work
on
a
try
and
move
this
into
another
working
group
right.
A
I
actually
didn't
asked
Greg
and
Paris
in
the
pr
which
we
have
to
regard
moving
the
repositories
into
the
code
for
Android,
but
there
was
no
reaction,
but
let's
create
up
here
and
then
yeah
we
can
I
can
do
that.
B
A
Yeah
I
can
check
the
I,
checked
the
latest
contributions
here
and
try
to
figure
out.
A
And
the
next
one
is
about
recovery
plan
created
quality
plan.
Can
you
provide
some
context
here.
B
Yeah
I've
asked
Chris
to
also
maybe
start
a
discussion,
but
it
would
have
been
nice
if
he
did
that
before
but
yeah.
So
what
we
found
is
that
our
cck
is
not
that
efficient
when
dealing
with
a
larger
deployment,
which
is
why
we
want
to
have
a
new
feature
that
is
more
geared
towards
like
recovering.
So
when
you
like
watch
deploy
is
optimized
for
if
you
have
or
keeping
things,
how
do
you
say
it
prioritizes
uptime
right?
B
It
tries
to
not
bring
things
down,
but
if
you
already
have
downtime,
then
you
just
want
to
go
as
fast
as
possible
and
that's
what
Bosch
recoveries
all
about
right.
So
it
wants
to
maximize
all
the
threats
and
everything
and
bring
all
your
instances
back
as
fast
as
possible,
but
then
do
it
in
a
way:
that's
not
as
interactive
as
cck.
Currently,
yes,
so
do
it,
for
instance,
group,
and
create
a
plan
to
recover
things.
B
So
the
idea
is
that
you
will
be
able
to
generate
a
a
plan
interactively
so
it
will
give
you
a
summary
per
instance
group
then
spit
out
a
yaml
file
and
then
that
yaml
file
you
can
then
pass
into
the
Bosch
recover
command,
which
will
use
cck
in
the
back
background,
basically
to
quickly
recover
the
instances.
B
Yeah
so
like
this
is
say,
you
have
like
a
400
instance
deployment
and
a
whole
AZ
is
down
or
something
but
you,
the
AZ,
is
coming
back
and
you
want
to
be
able
to
start
restarting
instances,
for
example,
so.
A
Hey
yeah,
but
we'll.
A
This
is
so
this
is
now
a
pending
pending
review,
disc
in
discussion,
state.
A
Oh
okay,
then
inbox
content
of
the
container
networking
performance.
A
A
Bryant
but
I
know
he
was
actually
going
to
reproduce,
but
and
this
in
discussions
did
there
was
no
feedback.
A
Which
is
more
this
year
as
I
Outlet
to
this
sphere,.
A
In
the
connections,
but
this
is
more
obviously
related
and
still
discussion.
A
C
B
The
ones
of
four
six
foot
zero
released
was
small
forms,
just
unintentionally
created
by
an
area
error
release
pipeline.
Therefore,
we
focus.
A
B
But
you
hear
what
I
say
about
this
or
not.
My
phone
keeps
switching
to
the
car
and
then
now
it's
on
my
headset
again
now
we
are
here:
okay,
yeah,
so
I
I,
reached
out
on
slack
and
Riker
said
that
1.46
release
was
malformed.
It
was
unintentionally
created
by
an
error
in
the
CIA.
A
B
Like
half
an
hour
ago,.
B
Cell
yeah,
so
what
I
did
last
time?
I
created
a
milestone
and
added
it
there
like.
If
we
want
to
think
about
a
new
stem
cell
right
I
think
we
should
like
start
marking
some
things
for
a
new
stem
cell
line,
not
line
but
like
the
upcoming,
the
next
LTS.
If
we
want
to
go
with
a
a
new
LTS
every
two
years
that
we
should
probably
in
a
few
months
start
to
work,
I
mean
we
still
need
to
have
a
proposal
for
that.
B
But
I
thought
it
is
not
that
far
out.
So
maybe
we
should
just
start
collecting
things
for
that
release.
A
Yeah
sounds
sounds
good
actually
because
yeah
makes
sense
and
you
create
the
milestone,
wish
list.
Okay,
I
see
here,
20
false
wishlist,
it's
already
added
to
that
yeah!
Yes,.
B
A
B
There
we
go
so
I
think
we
can
mark
it
as
prioritized.
It's
just
prioritized
like
really
far
out
right.
We're
gonna
pick
it
up
in
the
next
stem
cell,
like
major
stem
cells
that
we're
gonna
do
the
next
LTS.
B
A
It's
too
fast,
we've
done,
okay
and
the
next
one.
I
think
this
is.
D
No
actually,
no,
it
turns
out
that
you
cannot
release
create
a
release
of
Posh
DNS,
it's
the
router.
If
the
root
directory
is
not
both
DNS,
that
was
the
issue
here
and
I
wanted
to
check.
I.
D
Exactly
and
if
you
clone
that,
if
you
clone
that
by
standard,
then
it's
named
Posh
DNS
release
right
and
right
afterward,
and
if
you,
if
you
want
to
have
two
in
parallel
for
some
testing
purposes,
then
you
rename
some
of
the
one
of
these
post
DNS
releases
right
and
if
you
then
run
this
release
script,
it's
simply
failing
yeah,
but
I
didn't
find
the
time
right
now
to
to
check
if
we
can
improve
the
script.
D
D
And
I
tried
in
different
ways
starting
tasks,
killing
workers
and
so
on
and
so
forth,
but
simply
couldn't
reach
that
point
and
the
major
difference
is
as
soon
as
you
start
the
task
and
that's
run
by
the
worker.
It
receives
a
timestamp
right
and
all
the
tasks
that
are
hanging
forever
simply
do
not
get
a
started
at
timestamp
yeah.
D
So
it's
still
unclear
how
that
that
situation
can
be
reproduced,
yeah
and
and
how
these,
how
these
tasks
could
be
attached
to
a
queue
with
some
usual
ID
yeah.
B
A
A
Yeah
I
was
going
to
so
then
welcome.
A
B
B
But
no
nothing
else
from
my
side.
A
Okay,
then
enjoy.