►
From YouTube: 2019-04-09 Rook Community Meeting
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,
the
recording
has
started-
and
this
is
the
April,
9th
2019
Brook
community
meeting.
We
are
headed
towards
the
1.0
release,
so
that
will
be
the
main
focus
of
today.
So,
let's,
let's
take
a
look
at
the
milestone
and
the
project
is
well,
so
we
had
taken
a
pass
at
a
lot
of
these
issues,
but
we
still
have
a
fair
amount
open
now.
So
what
I
want
to
do
is
kind
of
maybe
go
through
by
by
functional
area
or
by
storage,
writer
and
kind
of
get
an
update,
talk
about
what
the
risks
are.
A
A
So
we
so
his
comment
is
targeting
the
week
of
April
22nd,
so
that
gives
us
I
believe
two
weeks
here
to
work
through
the
various
issues
that
we
still
have
open
and
that
are
in
scope
for
the
1.0
milestone
and
what
was
on
the
1.0
roadmap,
including
the
updates
that
Blaine
had
made
recently
to
get
the
one.
The
full
roadmap
updated
with
the
latest
thinking
and
planning,
and
also
gives
us
time
to
work
through
a
backlog
of
pull
requests.
That
Travis
just
mentioned
here.
Jared.
B
I
have
I
have
one
suggestion
to
make.
If
you
could
please
go
back
to
the
milestone,
so
I
think
one
of
the
things
which
kind
of
we
should
ensure
good,
stable
release
right.
It
was
important
for
1.0
rather
than
you
know,
all
the
features
in
the
world
I
think
first
thing
which
I
can
see
from
here
is
that
there
are
a
lot
of
features
which
some
of
them
actually
I,
think
some
bugs
actually
also
probably
features
I,
don't
see
like
a
full
trashing
like
Francis,
dynamic
provisioning
for
an
affair.
B
Storage
like
I
do
not
see
whether
this
I
feature
or
bug
right
so
I
think.
The
first
thing
we
need
to
do
is
very
simple:
just
go
through
that
list.
It's
29
and
just
put
bark
or
feature
in
then
be
I
mean,
and
that's
just
a
suggestion,
then
just
be
very
pitiless
so
to
speak,
to
the
features
and
just
essentially
just
remove
them
from
the
list,
and
does
that
mean?
That
means
that's
a
lot
of
features
which
for
share,
also
features
which
are
for
edge
of
s.
B
B
Just
simply
move
it
to
like
what
about
the
next
one
is
so
that
you
have
this
edit,
because
if
you
want
to
aim
for
April
22nd,
which
seems
to
be
a
good
date,
and
we
need
to
be
on
target
before
the
bars
film
events
right.
That
means
that
we
really
need
to
keep
just
bug,
fixes
and
be
very
picky
less
to
a
feature.
So
we
just
need
to
you
know
just
kill,
kill
all
the
fishes,
essentially
I.
Think.
A
A
I
think
it
would
be
difficult
to
remove
all
the
features
from
the
1.0
milestone
that
you
know
have
that
are
pretty
close
to
being
done,
or
you
know
have
a
you
know
it
kind
of
in
the
final
phases
thereof
getting
merged
in
that
would
make
things
more
difficult
from
a
tracking
perspective.
I.
Think,
though,
being
judicious
about
you
know
the
scope
of
features
of
ones
that
you
know
don't
really
make
much
sense
or
bring
a
lot
of
risk
to
the
you
know
to
the
potential
release.
I
think,
that's
something
that
I
would
be
much.
A
B
B
So
one
thing
you
probably
need
to
do
is
just
go
and
mark
what
does
feature
what
is
not
kind
of
thing
and
if
the
future
indeed
is
kind
of
been
in
development
for
quite
some
time,
why
don't
we
just
go
ahead
and
let
it
be,
but
if
it's
just
not
even
started,
we
can
I
feel
that
if
you'll
take
still
amount
of
time
to
discuss
this
etc,
it's
just
better,
just
kill
it
off
and
just
move
move
to
the
different
milestone.
Why.
A
Does
this
keep
refreshing
I'm,
not
sure
the
page
keeps
refreshing
on
me.
That's
weird
yeah
I
think
yeah
that
it
gets
down
to.
You
know
like
at
what
stage
or
phase
are.
Are
these
items
right
yeah?
So
that's
why
I
wanted
to
dig
in
with
you
know:
go
buy
each
store
provider
and
kind
of
get
an
idea
of
you
know
where
what
is
the
status
of
of
these
features
here,
or
these
implementations
and
and
see?
C
A
A
Yeah
here
you
go
so
in
the
metadata
for
each
one
of
the
templates
there,
it
will
apply
that
particular
label,
the
feature
label
or
the
bug
label
when
you
use
a
template,
so
that
is
and
I
think
that
was
kind
of
recently
yep.
Here
we
go
22
days
ago,
so
that
is
there
any
change
that
is
now
in
place.
So
you
know
it
didn't
retro
actively
go
and
apply
that
to
all
the
issues
that
have
been
open
previously,
but
going
forward.
C
A
A
Otherwise
you
know
each
I
think
that
you
know
there
is
some
sense
of
autonomy
for
each
storage
matter
of
you
know
they
can
kind
of
manage
some
of
their
own
tickets
and
kick
some
of
them
out
that
you
know
they
feel
don't
don't
make
sense
of
scope
for
1.0,
but
at
a
high
level.
Here
you
know,
as
a
group
like
calling
out
what
the
big
risks
are,
is
probably
my
biggest
my
biggest
priority
now
so
so
sorry
I
guess
we
can
start
with
edge
of
s
cuz.
That's
the
presidency,
so
Dimitri
for
HFS.
B
So
those
are
the
only
one
which
I
care
about
as
far
as
1.0
is
a
transition
to
build
better
and
I.
Think
I'm
tone
is
working
on
this.
He
also
working
on
CSI
for
I,
scuzzy,
I,
think
maybe
end
of
this
week.
We're
gonna
have
a
pool
request
for
it
and
that's
basically
it
those
other
really
features
I'm.
Actually,
you
can
close
the
integration
tests
because
we
need
to
just
integrated
that
yeah.
B
A
A
B
As
far
as
bugs
I
mean
we
not
aware
of
like
any
critical
ones,
so
there
are
like
small
ones
which
we
chasing,
but
they
not
reflected
in
here.
Yet
you
know,
you
know
in
a
you
know,
put
a
few
more,
maybe
but
to
us
a
DFS,
isn't
a
stabilization
of
course,
so
to
speak.
So
the
before
one
point
zero.
You
have
a
window
of
a
couple
of
weeks.
We're
gonna
be
focusing
mostly
on
fault,
tolerance
like
restart
start
ability.
A
A
A
C
C
The
critical
ones
for
upgrades
obviously
are
actually
doing
the
upgrade
documentation
using
aggregated
rolls
to
improve
that
upgrade
is
something
I
just
grabbed
and
started.
Looking
at
yesterday,
the
upgrade
doc
not
working
for
old
Nathan
spaces.
That's
also
something
that
like
kind
of
has
to
get
done.
Otherwise,
as
far
as
things
that
are
in
the
to
do
column,
I,
don't
think
there
are
any
things
in
there
that
will
hold
up
the
actual
release.
A
C
A
That
sounds
great
Blaine
yeah
like
I,
can
see.
Some
of
them,
like
you,
know,
support
for
dynamically
resizing
volumes
and
doesn't
even
have
an
assignee
yet
like
that
would
be
one
that
looks
like
a
clear
case
of
you
know.
Punting
out
of
1.0
right,
yeah,
agree
and
I
did.
I
was
gonna.
Make
another
comments
to
you.
Oh
yeah,
I.
Thank
you
for
you'd
made
some
comments
on,
like
you
know,
at
a
support
for
adding
and
removing
discs.
C
A
Thanks
yeah
and
then
so
what
do
you
also
played?
What
do
you
think
is
the
biggest
risk
in
in
these
tickets
here,
like
you,
had
mentioned
kind
of
the
ones
that
are
necessary,
like
the
upgrade
stuff,
that's
necessary,
which
I
totally
agree
with.
But
what
do
you
think
is
like
the
biggest
risk
that
maybe
we
want
to
raise
any
rate,
little
red
flags
about
right
now,.
C
C
A
Got
it
okay,
so
it
sounds
like
even
though
the
view
here
has
a
high
quantity
of
issues.
The
long
poles
or
the
the
risky
items
seem
to
be
well
identified
and
the
items
that
we
think
are
critical
are
well
identified
as
well.
So
following
up
with
Travis-
and
you
know
taking
another
pass
here,
it
kind
of
you
know,
constraining
down
or
cleaning
up
or
converging
on
the
1.0
issues
sounds
sounds
like
a
good
idea.
Cole
look.
It
looks
like
this
is
a
general
feature
here.
A
C
A
C
A
C
A
Okay,
cool:
let's
go
back
to
issues
not
all
of
them,
just
that
there
you
go.
Oh
that's
easy
on
us
on
the
call.
Today,
no
I
do
not
see
honest.
He
had
mentioned
I
think
in
the
dev
channel
that
the
Cassandra
items
that
there
he
may
need
to
call
or
punt
on
some
of
those,
but
so
we
could
follow
up
with
him
on
that.
A
I
think
that
we're
trying
to
figure
out
what
the
timing
was,
but
if
Travis's
suggestion
of
converging
on
the
week
of
April
22nd,
then
that
might
actually
have
had
the
time
that
Yanis
needed
to
converge
for
Cassandra
as
well,
and
then
with
the
discussion
we
had
about
the
roadmap
recently
on
that
pull
request
that
you
had
Blaine
I,
think
that
the
scope
I
think
is
fairly
well
understood
now,
so
are
there
any
other
issues
that
anybody
wanted
to
talk
about
or
identify
in
the
1.0
context?
Here
before
we
move
on.
C
A
So
and
so
I've
kind
of
gotten
a
little
bit
of
exposure
to
this
process
here
in
from
the
way
that
SIG's
storage,
you
know
walks
through
and-
and
you
know,
their
set
of
features
and
scope
and
everything
that
they
want
to
include
and
then
how
they
deal
with
these
dates.
So
one
thing
that
that
you
know
it's
probably
good
to
clarify
for
everyone,
because
it
was
counterintuitive
for
me
was
how
you
know
a
feature
freeze.
You
know
what
that
means
is
kind
of
to
me,
at
least
from
my
experience.
A
What
I'm
so
I've
seen
is
that
a
feature
freeze
means
that
that
after
that
date,
you
know
new
product
proposals
for
teachers
to
be
in
scope
for
the
release
or
will
no
longer
be
accepted.
I
Ridgely
thought
it
meant
that
you
have
to
be
done
building
on
implementing
all
your
features
by
that
date,
which
is
not
okay,
just
to
clarify
for
everybody,
which
is
really
important
to
understand
it's
more
like
after
this
date,
we
won't
change
the
scope
of
the
mouse
there
anymore.
We
won't
accept
new
features
as
proposals
there.
A
You
know
so
that
I
think
which
is
important,
which
it
kind
of
can
you
know,
that's
a
good
constraint
on
the
scope
of
the
milestone
fairly
upfront
so
that
you
can
then
focus
on
delivering
and
converging
as
opposed
to
continuing
to
have
scope
creep,
which
obviously
then
ends
up,
but
you
know
pushing
dates
back.
So
one
thing
I
think
that
really
really
I
think
where
we
failed
on
this
on
1.0-
and
you
know,
I,
take
responsibility
for
this
myself.
A
Is
that
there's
so
there's
so
we
felt
we've
had
a
debate
about
feature,
release
of
scope
versus
date,
you're,
sorry
date
driven
versus
feature
driven
releases,
and
so
you
know
we
haven't
really
before
done
like
a
specific
date.
We
will
ship
here
and
help
that
we've
kind
of
been
like
these
are
the
features
we
want
when
they're
ready,
we'll
ship,
but
really
what
we
failed
at
I
think
one
of
the
things
we
failed.
A
That
is
that
we
haven't
had
good
communication
and
understanding
amongst
the
community
about
the
dates
you
know,
I
kind
of
like
talked
to
Travis
and
mentioned
and
casually
like
it
community
meetings
of
oh
we're
targeting
this
time
frame,
but
it's
not
published
anywhere.
It's
not
well
disseminated
information
of
you
know.
This
is
the
date
that
we
want
to.
You
know
be
done
with
with
with
features
or
coding,
and
this
is
what
we
want
to
be.
You
know
doing
a
release
candidate
and
all
that
stuff.
A
That's
not
really
talked
about
so
there's
when
you
don't
have
that
well,
published
well,
understood
well,
talked
about
you
know.
You
end
up
in
a
situation
with
surprises
for
people,
okay,
let's
what
we
won't
a
ship
next
week,
and
then
you
know
some
people
or
two
haven't
heard
anything
about
it,
and
you
know
they
are.
You
know
not
ready
for
that
at
all,
so
it's
gonna
be,
as
you
know,
blame
you
suggest.
A
A
A
At
it,
I
definitely
think
this
is
a
very
good
model.
You
know,
because
it
sets
expectations
and
it
has
things
very
clear
that
people
can
continue
to
to
drive
towards.
So
I
think
this
is
a
good
proposal
that
you
know
a
direction
that
we
should
go
for.
41.1
did
you
have
other
thoughts
on
that
Blaine
or
anybody
else.
C
Iii
mean
my
my
thoughts
are,
you
know,
I
think
the
kubernetes
schedule
is
a
little
more
defined
than
I
I
think
we
can
really
get
out
like,
especially
with
things
like
alpha-1
release.
You
know,
I'm,
not
I,
don't
know
that
I
necessarily
want
to
put
that
for
sure
in
like
our
release
timeline,
but
just
you
know
just
starting
to
have
an
overall
idea
of
like
you
know.
This
is
when
we're
cutting
off.
A
Yeah
yeah
definitely
there's
there's
no
way,
we'd
be
able,
we
don't
have
the
infrastructure
or
the
you
know,
manpower
to
to
do
something
of
this.
The
granularity
yeah.
We
definitely
couldn't
do
that,
but
yeah
the
the
general
idea
of
Giddings
dates
in
place
of
you
know
when,
like
a
code,
freeze
should
be-
and
you
know
when
the
stability
portion
should
start
all
that
sort
of
stuff.
You
had
that
totally
a
min
support
of
that.
A
C
A
A
Well,
thank
you.
Blaine
you're,
welcome
I,
say
say
just
not
on
the
call
right,
yeah
I,
don't
see
sage
there
so
put
sage
added
a
a
note
here.
That
is
a
reminder
for
everyone
who
will
be
in
Barcelona
for
cube
con
next
month.
There
is
also
a
co-located
cephalic
on
conference.
That's
you
know
in
Barcelona
as
well,
which
is
the
the
two
days
before
cube
con.
Actually
it's
one
day
before
cube
con
on
the
19th,
which
is
I,
think
Sunday,
and
then
the
20th
is
Monday,
which
is
the
day
all
the
other.
A
There's
a
registration
button
right
there.
Now,
where
does
that?
Go
uh-huh?
Okay,
so
it
looks
like
you
can't
so
disregard
everything.
I've
said
about
the
registration
process
because
it
looks
like
there
is
a
cific
registration
ability
here
as
well.
I
thought
you
had
to
do
it
through
the
co-located
events,
but
you
can
you
can
got
it
so
you
have
both
of
those
options.
I'm.
A
Cool
thank
you
for
clarifying
that
cool,
so
yeah.
So
there's
a
link
in
the
community
meeting
here
and
we
also
added
one
to
the
announcements
and
general
channels
and
slack
as
well,
so
we're
hoping
I'm
expecting
to
be
there
and
I
know
a
lot
of
you
folks
on
the
call
here
will
be
there
as
well
so
that'll
be
great
to
see
everyone
in
person
and
and
get
the
Barcelona
week
kicked
off.
A
A
Unfortunately,
it's
the
CN
CF
reached
out
to
us
and
you
they
do
for
each
coupon.
They
do
a
maintainer
track
and
we've
had
an
intro
session
in
a
deep
dive
session
for
four
rook
as
part
of
the
maintainer
track
of
each
cube
con
and
in
China
at
Shanghai
this
year,
that's
apparently
a
different
venue
than
it
was
in
Shanghai
just
last
year
in
November,
so
the
logistics
for
like
number
of
rooms,
a
number
of
parallel
sessions
that
could
be
run
apparently,
is
more
constrained
than
it
was
last
time.
A
So
they
can
only
give
one
session
out
per
per
CN
CF
project.
So
for
rook
we
decided
to
do
a
deep
dive
only
instead
of
the
intro,
because
Dimitri
here
on
the
call
has
a
a
cool
talk
and
InFocus
lined
up
about
multihomed
networking
support
for
for
for
Shanghai.
So
we're
gonna
be
going
with
the
just
that
deep
dive
talk
in
Shanghai,
which
I
think
it'll
be
really
cool.
A
Okay,
so
we
talked
about
the
1.14
support
where
we
saw
in
that
pull
request,
that
it
was
a
test
matrix
update
only
and
not
any
code
support.
So
Travis
had
a
note
here,
though,
that
it
looks
like
he's
having
trouble
with
it,
that
the
1.14
build
looks
fine
when
he's
locally
testing
I
guess
either
with
the
vagrants
multi-node
vagrant,
multi,
node
setup
or
mini
coud
setup,
but
our
integration
test
environment
is
failing
and
so
he's
asking
if
there's
anybody
else
that
is
available
to
investigate
and
take
a
look
at
that.
A
A
Yeah
that
this
build
failure,
apparently
it
continues
to
happen.
So
if
there
is
somebody
who
enjoys
looking
at
integration
test
failures,
then
you
are
welcome
to
do
so.
It's
look
sounds
like
help
is
needed
there,
every
other,
milestone
or
sorry
I
kubernetes
version
is
passing.
It
looks
like,
except
for
1.14,.