►
From YouTube: 2020-07-27 Multi Large Working Group
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
Yeah,
I
think
jegus,
yes,
gregor
updated
the
issue
with
the
next
steps,
so
there
were
some
minor
fixes
that
were
emerged
last
week,
so
we
expect
to
do
a
initial
rollout
this
week.
Camille
will
step
in
and
together
with
sean
they'll
work
on
rolling
it
forward.
A
So
that
is
progressing
well
and
then
another
item
that
we
have
is
jar
is
working
on
creating
the
run
book
to
manually
shift
workloads
between
clusters-
and
this
is
also
progressing.
So
this
is
his
priority
this
week,
so
that
will
allow
us
to
know
what
kind
of
operational
gaps
we
have,
which
will
be
helpful
regardless
of
this
group,
but
it
will
also
be
helpful
to.
A
To
feed
some
information
here,
so
that
is
also
what's
happening.
Next
we
have
action
items
this
week
and
then
one
blocker
is
jerry
still
needs
to
work
out.
Helm,
charts,
plus
operator
automation,
he's
back
from
his
pto
after
two
days,
so
he'll
work
on
that.
A
I've
added
one
discussion
point
because
I
think
I
have
the
right
audience
for
this:
we've
been
adding
or
we
are
adding
new
services,
and
as
part
of
that,
we
are
not
thinking
cloud
native
first
or
kubernetes
first,
which
is
fine.
It's
it's.
How
we've
been
doing
things
for
a
long
time,
but
I'm
wondering
if
it's
time
to
think
about
how
can
we
put
forward
whenever
we
add
a
new
service?
How
can
we
ensure
that
it's
ready
to
run
on
kubernetes
immediately,
because
what's
happening
right
now
is
development?
Add
support.
A
We
need
to
do
an
extra
overhead
of
introducing
something
that
will
migrate.
This
may
or
may
not
provide
sufficient
information
for
us,
but
ultimately,
we'll
have
to
do
the
same
thing
and
migrate.
The
service
later
on
so
initially
the
the
overhead
might
be
a
bit
higher
to
structure
the
future
from
the
beginning
to
work
inside
of
kubernetes,
but
it's
in
the
long
run.
It
might
save
us
some
time.
So
I'm
wondering
whether
people
have
opinions
on.
B
B
A
B
C
I
I'm
thinking
in
terms
of
I
think
I
agree
with
that.
I'm
really
thinking
where
we
can
test
it
that
it's
installation
cloud
native
cable
earlier
on,
like
maybe
italy
and
any
new
service
when
you
set
up,
you
probably
need
a
review
app-ish
that
you
know
test
the
integration
with
cloud
native
and
don't
wait
for
staging
or
production
to
to
like
know
that
your
your
things
are
working
with
cloud
native.
So
I'm
thinking
along
the
lines
of
that
but
yeah,
I
think
that's
that
makes
sense.
A
A
Forward
I'll
I'll
think
about
where
I
can
add
this
to
the
handbook,
but
expect
from
me
this
week
an
update
on
that
merger
question.
D
Yeah,
the
other
thing
marin
is
if
it's
just
validation
and
correction
of
mistakes,
but
if
you
have
general
guidance,
the
thing
I'm
concerned
about
is
is
that
the
classic?
D
D
For
what
it
means
to
be
cloud
native
in
this
context,
then
I
think
I
think
that's
definitely
something
we
can
accommodate
and
test
out
ahead
of
time.
A
Yeah,
that's
a
good
call
out.
I
believe
there
won't
be
a
big
change
in
at
least
the
initial
requirements,
because
one
of
the
initial
requirements
is
always
that
it
can
be.
It
can
run
right.
It
can
be
installed
somewhere,
so
people
need
to
engage
with
distribution
immediately
anyway.
This
would
just
reverse
the
roles
a
bit.
It
wouldn't
be
omnibus
first,
it
would
be.
It
needs
to
work
in
charts
first,
and
that
informs
a
lot
of
your
other
decisions.
A
Right,
for
example,
you
won't
be
able
to
use
shared
storage
like
nfs,
because
just
in
that
world
it
doesn't
work
with
that,
so
it
it
would
be
benefiting
omnibus
as
well.
A
Mark
all
right,
sid,
your
writing
is
next.
B
Yeah,
I'm
just
curious,
I'm
not
sure
it's
a
good
use
of
everyone's
time,
but
I've
been
thinking.
First
of
all,
I
love
that
we're
making
the
nfs
requirement
of
pages
a
smaller
problem
but
kind
of
working
around
it
and
by
constraining
it
to
just
that.
B
I
wonder
what
the
long
term
plan
is
to
get
pages
off
of
nfs.
Are
we
using
that
by
using
s3
like
object,
storage,
to
serve
files,
and
if
so,
will
that
be?
B
A
A
So
this
epic
has
a
couple
of
ideas
going
forward
so
they're
doing
a
bit
of
a
poc
just
to
understand
what
kind
of
penalty
we
would
have
to
pay
there
if
there
is
any,
but
I
think
right
now,
we
are
at
the
stage
of
tech
evaluation
only
because
the
concern
there
is
that
there
is
going
to
be
a
significant
performance
impact
as
well,
because,
right
now
it's
really
quick
to
serve
those
things,
how
it
would
work
when
you
introduce
another
layer
somewhere
else.
A
I
I
don't
know
whether
the
team
still
figured
that
out
yet.
B
E
Yeah,
so
this
was
I'm
trying
to
find.
There
is
an
older
epic
too,
that
I'll
link
in
the
agenda.
E
Put
that
right
there,
sorry
tim,
so
we
had.
We
had
done
a
kind
of
a
proof
of
concept
on
this
with
camille
like
a
year
ago
or
so,
and
we
were
trying
this
idea
of
taking
a
zip
file,
putting
it
in
object,
storage
and
then
kind
of
extracting
it
on
the
fly
to
serve
up
a
site
and
he
had
done.
I
think
there
was
a
an
mr
that
he
had
worked
on
a
little
bit
to
try
this
idea
out
and
and
kind
of
we'd
kind
of
explore.
E
That
idea,
I'm
not
sure
where
the
team
has
taken
it
now,
but
but
that
was
kind
of
the
original
idea
and
this
this
sort
of
predates
some
of
the
other
work
that
was
done
to
move
the
configuration
part
of
pages
off
of
nfs.
E
That's
done
now,
and
now
it's
just
the
actual
file
storage,
that's
happening
there.
All.
B
Right,
I
think
I
think
you
get
kind
of
the
worst
of
two
worlds.
If
I
understand
this
proposal
correctly,
wouldn't
you
just
like
get
out
of
the
business
of
serving
the
file
in
the
first
place.
Just
have
the
file
be
just
point
them
to
the
object
bucket
people
can
download
it
there
at
the
cloud
handle
yeah.
E
Yeah,
so
so
I
think
that
was
that
was
sort
of
what
we
were.
I
don't
know
it
was.
It
was
an
attempt
to
try
to
try
something
in
the
process,
because
we
were
we're
just
sort
of
like
looking
into
ideas,
but
I'm
actually
not
familiar
with
where
the
team
is
on
this
right.
Now
I
don't.
I
don't
work
with
that
team
anymore,
so
I'm
not
I'm
not
sure
if
they're
they're
still
pursuing
that
idea.
This
is
still
what's.
B
E
Yeah
I
would
need
to.
I
would
need
to
dig
in
with
the
team
and
see
what
their
what
other
sort
of
considerations
they're
tying
into
this
like
it.
Could
it
could
be
around
access
control.
It
could
be
around
some
other
other
things
like
domain
name
serving
but
yeah.
I
think
there's
there's
probably
some
some
questions
to
ask
about
this
approach.
E
I
can
take,
I
can
take
an
action
item
to
talk
with
with
daniel.
This
is
his
area
and
see
what
their,
what
they're
up
to.
A
There
is,
as
far
as
I
know,
I'm
going
to
butcher
his
name
now.
Haimea,
that's
right!
I
may
jaime
yeah.
I
think
he's
he's
directly
assigned
to
work
on
this
right
now
and
vladimir
is
pair
pairing
with
him.
B
Okay,
we
said
we
got
two
people
working
on
this
they're
working
on
a
year
old
idea
that
seems
a
bit
hybrid,
so
yeah
darby,
maybe
that's
a
good
idea
to
talk
with
them
yeah.
I
can
do
that
and
I
at
some
point
we
have
to
ask
like
how
many
people
are
using
the
private
pages
functionality
and
maybe
deprecate
it
in
a
future
version.
Anything
is
on
the
table.
We
get
need
to
get
rid
of
this
nfs.
C
I
have
a
quick
one
since
time
is
precious
and
are
we
planning
to
have
a
board
to
just
show
the
different
tracks
of
work
for
this
working
group,
and
and
so
we
can
get
status
async
at
a
high
level.
I
think
there's
a
is
there
any
plan
for
some
sort
of
representation
for
that.
C
A
The
issue
board
will
be
very
hard
because
we
have
part
of
the
work
in
infra
part
of
the
work.
Sorry
part
of
the
working
gitlab
com
group,
part
of
inget
labor
group.
So
it's
split
across
but
I'll.
I
can
talk
with
with
jerry
to
see
like
what
kind
of
plans
he
has.
Maybe
we
can
present
something
in
the
working
group
or.
A
All
right,
I
like
that
as
an
action
item
for
myself,
jerry
has
highlighted
three
items
and
I
highlighted
one
for
darby
and
I'm
gonna
add
one
for
myself
and
I
want
to
thank
all
of
you
for
your
time.
I
hope
you
have
a
good
week
ahead.