►
From YouTube: 2018-09-11 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
A
B
B
B
C
B
Just
from
a
few
people,
commenting
on
that
issue
that
it's
still
happening
happens
has
happened
occasionally
on
1.10
I
haven't
seen
anybody
say
it's
happened
on
111,
but
maybe
those
in
those
clusters
they
just
haven't
been
running
one
at
eleven,
where
the
conditions
are
just
right.
I
didn't
see
anything
in
1.11
that
would
would
have
changed.
The
behavior
around
mountings
1.10
had
the
big
change
where
they
news
blockhead,
be
instead
of
LS
walk.
A
Great
yeah,
that's
that's
what
I
thought
would
have
affected
the
root
cause
that
would
you
know,
change
or
improve
a
fix.
The
way
that
you
know
our
BT
devices
are
detected
because
that's
that's
the
problem
here
right
there
we're
not
detecting
that
the
device
already
exists.
So
we
go
ahead
and
you
know
stomp
on
it
right.
Yeah.
C
B
B
D
C
D
C
D
D
A
There's
also,
probably
at
the
unit
test
level,
honestly
I,
think
that
the
the
logic
about
you
know
orchestrating
the
OSDs
and
waiting
for
them
to
finish
and
stuff.
I.
Think
that
there
it
is,
it
would
be
possible,
it
wouldn't
be
trivial,
but
it
would
be
possible
to
add
some
more.
You
know
testing
that
goes
through
some
more
complicated
scenarios,
different
timing
of
things
finishing,
and
you
know
race,
conditions
and
stuff
like
that.
With
some
mocking,
you
know
it's
not.
That
would
be
trivial
effort,
but
we
could
or
get
more
coverage
there.
D
C
I
mean
that's
that's
trying
to
get
to
there's.
There's
there's
definitely
needs
to
be
an
effort
around
testing
long-haul
testing
performance
testing,
large
cluster
testing.
That
needs
to
happen
as
a
you
know,
as
a
release
gate
for
every
release
we
make-
and
maybe
this
is
one
that
fits
in-
that
bucket
I.
C
A
I'd
be
nice
to
find
somebody
who
has
an
interest
in
that
came
to
drive
that
okay.
So,
let's
let's
go
ahead
and
move
on
then
to
0.9.
You
know
so
we
are
plugging
along
with
the
tickets
and
issues
in
the
0.9
milestone.
I
know,
there's
a
number
of
issues
that
we
still
you
know
don't
have
owners
for
that.
We
have
help
one
and
the
issues
tag
does
Help
Wanted.
A
If
we
all
remember
when
we
started
this
milestone,
we
made
a
specific
conscious
decision
to
be
very
aggressive
with
the
items
and
issues
that
are
in
this
milestone,
with
the
hope
that
we
would
find
owners
for
them.
As
we
continued
so
I
mean
we
still
have
a
number
of
these
issues
that
do
not
have
owners.
That
would
be
great,
if
you
know
people
from
the
community
with
interest
in
them
could
step
up
to
to
take
on
some
of
these.
B
A
You
know
there's
a
number
of
those
there's
a
few
issues
that
came
up
that
he
was
interested
in
while
he's
driving
that
design
in
implementation
effort
for
Cassandra-
and
you
know
those
are
things
we
already
want
to
accomplish-
like,
for
instance,
919
about
you
know
being
able
to
use
persistent
volumes,
as
the
underlying
you
know,
source
of
for
the
storage
fabric,
that's
being
orchestrated,
so
some
of
these
ideas
are
converging
together,
which
is
which
is
good,
I,
think
it
shows
it's.
You
know
some
things
are
you
know
kind
of
going
in
the
right
direction.
B
A
C
A
A
All
right,
so,
let's
go
to
the
community
topics.
The
first
thing
I
just
wanted
to
mention
here
that
this
is
there
was
no
announcements.
There
is
no
nothing
formal
about
this
at
all,
but
I
did
want
to
mention
that
the
voting
had
started
for
the
proposal
to
graduate
the
rook
project
within
the
cloud
native
computing
foundation
to
the
incubation
stage
from
the
sandbox
stage
and
the
technical
Oversight
Committee,
there's
nine
members
and
two-thirds
are
required
to
approve-
and
we
have
six
yes
votes
on
this
now
of
the
nine
members.
A
So
we
do
have
the
two-thirds
necessary
to
approve
the
the
graduation
take
evasion.
So
that's
a
very
big
step,
I
think
for
the
project
and
it
says
a
whole
lot
about
the
growth.
The
you
know,
effort
from
the
community,
the
you
know,
everything
that
everybody
has
put
into
this
project
to
you
know
elevate
it
to
the
next
level,
so
I'm
really
really
happy
with
this
effort.
I'm
really
really
proud
of
all
the
work
that
everyone
has
done
there.
So
we
should
feel
happy
about
this.
B
C
A
Just
wanted
to
mention
this
real
quick
here
and
see,
if
make
sure
we
had
a
conclusion
about
it.
So
for
this
particular
pull
request,
we,
it
was
open
directly
against
the
0.8
release
branch
and
the
normal
flow
for
for
backporting
fixes
to
release
branches
is,
to
you
know,
merge
them
into
master
first
and
then
cherry-pick
it
to
the
release.
Branch
and
I
prefer
that
that
that
approach,
because
it's
you
know
reasons
in
that
I
outlined
here-
that
you
know
that
slit
causes
less
less
ci,
builds
and
less
load.
A
On
the
you
know
all
of
our
CI
pipeline,
which
is
a
fair
assessment
kind
of
stretched
thin
and
also
then
it
keeps
all
the
discussion
about
the
the
fix
in
a
single
place,
as
opposed
to
fragments
at
a
cross
pull
request.
So
that's
my
preference
and
I.
Don't
know
if
you
know
you
all
have
a
disagreement
with
that
or
if
you're
you
know
in
alignment
with
that
thought.
A
Cool
okay!
Well,
it's
so
I!
Don't
have
any
other
topics
community
topics
here,
so
we
can
move
along
to
pull
requests
to
discuss
I
this
the
Cassandra
design.
I.
Think
I
personally
have
not
visited
this
again
after
I
gave
some
feedback
Oh
quite
a
while
ago.
So
I
would
like
to
I.
Think
Yanis
is
kind
of
working
on
implementation
stuff
now
so
I
I
personally
want
to
revisit
this
again
and
just
kind
of
clean
things
up
and
get
this
merged
in.
So
we
don't
have
a
dangling
design
doc.
A
F
Alright
I'll
get
off
mute
here.
Yeah
I,
don't
like
have
any
immediate
pressing
need
to
get
this
done
immediately,
but
I
wanted
to
let
everyone
know
that
pull
20
95,
which
is
pretty
polished
now
now
initializing
the
monitor
in
init
containers,
the
Ceph
monitor
in
a
knit
containers
is,
is
pretty
well
done.
I'm
looking
for
feedback
and
I
think
at
this
stage
it's
good
like
if
you
want
to
be
as
nit
picky
with
the
feedback
as
possible.
F
A
A
F
I
think
this
is
the
the
time
for
that,
like
it's,
not
manohar,
not
holding
anything
back
by
waiting
on
this
particular
PR
and
I.
Think
it's
gonna
be
good
too
yeah
just
to
make
sure
that
the
patterns
and-
and
you
know
how
we're
we're
writing
these
things
looks
good
on
this
attempt
to
take
all
of
those
lessons
and
move
them
forward.
Yeah.
A
I
think
that's
that's!
Why
is
to
Blaine
of
you
know
if
this
is
the
first
one
in
a
set
in
a
pattern,
a
set
of
them
that
we're
going
to
do
then
you
know
making
sure
that
up
front
we've
kind
of
vetted
and
got
things
to
a
you
know
a
quality
position
before
we
continue
with
you
know,
duplicating
or
replicating
that
another
other.
B
F
B
A
A
B
B
A
E
A
C
B
A
E
A
B
A
Okay,
let's,
let's
follow-up
them
on
that
on
the
ProQuest,
open
ones
and
see
if
there's
if
there
are
ones
that
are
stale
or
need
to
be
closed,
or
you
know
ones
that
are
waiting
for
feedback.
You
know,
I
I
could
definitely
hold
myself
personally
accountable.
That
I
have
not
been
entirely
on
top
of
all
of
the
active
requests
and
I've.