►
From YouTube: Ceph Orchestrator Meeting 2021-11-09
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).
B
Sure,
although
I'm
not
sure,
there's
a
lot
of
updates
since
last
time
we
talked,
or
has
it
been
a
couple
weeks
already,
but
anyway
the
the
current
discussion,
so
it
was
two
or
three
weeks
ago.
I
opened
up
this
discussion
on
the
topo
lbm
operator
about
whether
they
could
handle
raw
devices
as
well.
So
it
was
linked
in
a
previous
agenda,
so
copy
it
up
here
too
yep.
B
Yes,
it
seems
possible,
but
it
would
it'll
take
some
time
and
investment
to
make
that
happen
and
the
and
so
then
the
question
becomes.
How
important
is
the
scenario
to
go?
Make
that
happen
for
what
we're
looking
at
for
ocs?
It's
really,
the
scenario
looks
like
well.
Lso
can
handle
the
scenarios
we
need
with
static
volumes
with
raw
volumes
basically,
and
that
top
lvm
will
handle
the
local
storage.
Well,
not
just
local
storage,
the
the
dynamic
provisioning
of
local
storage
that
we
need
for
ocs
and
that
we
can
use
those
both
together.
B
So
it's
and
that's
so
that's
what
we're
pursuing
in
ocs
right
now
as
far
as
we'll
use
both
of
those
solutions
and
not
pushing
the
short
term
to
get
an
update
to
topple
lvm
for
raw
devices.
I
I
still
agree
that
I
mean
that's
the
ideal
world.
If
there's,
if
in
one
place,
you
can
dynamically
decide
if
you
want
ride
raw
devices
or
an
ld,
all
right,
it'll
just
take
a
while
to
get
there.
C
Oh,
I
guess
I
guess
it's
a
good
summary
of
where
we
are
yeah.
I
just
want
to
highlight
maybe
that
the
ideal
scenario
we
have
in
mind
might
not
be
coming
from
the
vm.
C
C
This
doesn't
actually
change
the
decision
to
go,
move
to
point
vm
downstream
wise
for
the
single
load
openshift
use
case
that
is
targeted
for
410,
so
yeah.
That's.
D
Okay,
just
just
do
that,
one
more
thing:
okay,
that
is
that,
probably
well
we
last
week,
we
discussed
it
about
the
possibilities
to
do
modifications
in
topper
vm
project
or
in
the
topper
vm
operator
project,
in
order
to
make
possible
the
smart,
dynamic
processing,
okay,
to
have
drug
devices
and
to
consume
these
drug
devices
just
when
when
they
they
are
needed.
Okay-
and
it
seems
that
in
the
topology
project
directly,
is
something
that
what
as
sebastian
has
commented
and
answers
is
something
that
what
maybe
is
going
to
be
problematic.
D
In
this
moment,
we
are
working
with
a
topple
vm
operator
project
and
we
have
a
first
version
of
the
operator
working
with
the
original
match
of
the
vm.
I
think
that
is
a
good
progress.
Okay
and
after
that,
probably
what
we
need
is
to
just
to
to
be
sure
that
everything
is
working
for
the
run
environment,
the
it
has
no
to
what
to
deploy
that
in
single
nodes.
Okay
and
after
that,
or
just
just
in
parallel-
maybe
we
can.
A
A
So
I
have
four
other
topics
in
my
mind.
Right
now,
one
one
is
the
pull
request
about
adding
a
docker
and
docker
development
environment.
A
A
E
A
A
Yeah,
I
was
able
to
see
the
comment
here
I'll
have
to
investigate.
I
can't
I
can't.
A
F
A
Okay,
other
than
that,
I'm
pretty
happy
with
that
docker
and
docker.
A
Environment
because
it
really
is
an
easy
way
to
create
a
cfldm
cluster.
G
Yeah
I
mean
this
doesn't
really.
This
is
like
a
work
around.
We
should
be
testing
this
anyway
testing
this
center
stream,
because
it's
the
upstream
for
for
braille
and
whatever
else,
but
it
seems
to
work
so
far.
I
need
I'm
rebuilding
re-running
the
the
tool
suite
to
make
sure
and
so
there's
a
patch
that
basically
adds
that
stream
and
switches
some
stuff
to
use
data
stream
and
there's
one
that
temporarily.
G
A
Maybe
an
ancient
put
my
container
to
its
two
version:
shippingpotman
one
dot.
I
don't
know.
G
G
I'll
hopefully,
have
a
run
to
confirm
that
and
then
we
can
hopefully
immerse
that
today,
mm-hmm
the
test,
we've
cleaned
up.
It's
been
headache.
A
Okay,
the
another
pr
that
I
would
like
to
discuss
is
the
system
so
ctld
location.
So
we
have
a
pull
request.
A
And,
and
their
idea
is
that
slav's
users
left
flip
might
be
read
only
in
in
some
circumstances,
which
is
not
the
case
for
clashes
to
see
something
proper
folder
seems
to
be
the
slash
edc
for
etc.
Folder.
A
It's
just
a
matter
of
migrating
from
the
old
path
to
the
new
path.
I
guess
what
was
being
written
there,
that
we
we
had
some
cdl
config
files.
A
A
A
A
A
A
E
G
A
Anyway,
maybe
we
should
just
delete
it.