►
From YouTube: 2020-05-11 :: Ceph Orchestration 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
C
C
D
C
C
A
A
A
Yeah
stuff,
like
that,
URI
any
anything
from
your
side.
Oh,
who
wants
to
who
wants
to
begin
I.
E
E
You
want
to
understand
itself
radium
like
Yuri
mentioned,
and
what
are
the
features,
and
we
have
few
queries.
How
is
it
is
different
from
the
previous
installers
and
what
are
the
advantages
and
how
this
can
be
integrated
into
the
pathology,
so
those
are
some
of
the
things
team
is
curious
to
know
before
we
start
testing
these
things.
E
A
A
A
A
We'll
be
self-id,
MSS
are
two
parts,
one
part
is
a
safe,
ATM
binary
and
the
other
part
is
the
orchestrator
module
or
the
safe
ATM
manager
module
which
lives
within
the
safe
manager
to
deploy
a
cluster.
You
first
bootstrap
your
cluster
on
a
single
node
which
includes
a
safe,
monitor
under
safe
manager,
and
then
you
will,
from
from
that
point
on,
you
will
use
they
two
operations.
A
A
The
the
user
experiences
some
somewhat
similar
to
give
a
netizen
to
look.
So
we
have
a
list
of
service
specifications
which
define
where
a
service
is
going
to
be
deployed
and
how
it
is
going
to
be
deployed.
So
we
have
a
service
specification
for
monitors.
We
have
a
service
specification
for
managers.
A
B
A
Have
three
ways
to
specify
the
location
of
services?
One
is
explicit
specification
like
I
want
to
have
this
service
running
on
this
node.
Then
you
can
use
labels.
You
can
label
the
few
a
few
holes,
for
example,
monitoring
or
I.
Don't
know
they
are
user
defined,
and
then
you
can
say
I
want
to
have
maybe
three
demons
running
on
all
nodes
that
are
labeled
with
I,
don't
know
monitoring,
for
example,
and
then
there
was
a
third
way
by
using
block
patterns.
A
Maybe
a
goal
for
self
ATM,
the
ego
force
F
ATM,
has
to
be
enterprise,
ready
to
make
it
possible
to
really
use
safe,
EDM
for
make
extent
safe
ATM
what
we
need
to
do
to
do
it
to
deploy
it
on
enterprises,
let's
see
goal
for
safe
ATM,
and
it's
also
meant
to
be
less
complex
than
Cuban.
It
is
and
drop
from
a
from
a
user's
perspective.
A
Okay
and
let's
continue
with
the
total
integration,
we
have
an
extensive
QA
suite
already
that
contains
a
few
tests
that
will
test
safe,
IDM
itself.
That
will
create
a
small
cluster
using
the
safe
IDM.
You
a
task
and
a
smoke
test
using
a
safe,
IDM,
cute
aq8
house
in
the
in
a
role
as
Madrid.
That
sage,
thankfully
created.
A
If
you
expect-
and
it
also
covers
an
upgrade
to
a
front
from
an
early
octopus
release
to
the
latest
master
or
to
the
latest
state
of
that
pull
request
and
Russia
as
working
on
integrating
deploying
NFS
testers
with
safe
idiom.
A
A
E
A
Thanks
very
much
diversity,
their
work
up,
your
any
questions
from
you.
B
A
That
would
be
awesome
and
and
I'm
really
open
to
any
enhancements
to
our
testing.
Suite
I'm
I
have
to
admit
that
I'm,
not
really
so
an
expert
in
totality
I
can
maintain
it
and
I
can
enhance
existing
tests,
but
I'm
not
really
the
right
person
to
actually
improve
the
overall
structure
of
the
safe
adium
within
in
general,
oh
yeah,
if
you,
if
anyone
has
some
some
improvements,
I'm
really
keen
on
on
integrating
that
into
the
into
technology,.
E
A
There
is
an
adduct
workflow
to
convert
an
existing
cluster
to
safe
ATM,
which
is
still
a
rather
manual
task.
That's
not
really
automated
by
safe
ATM
itself.
Any
improvement
that
would
be
awesome.
A
There
is
nothing
really
preventing
anyone
from
upgrading
and
Nautilus
cluster
directly
to
octopus
using
safe
IDM.
If
you
combine
the
adaption
step
and
the
upgrade
step
to
Nautilus
Tim
is
is
working
on
making
sure
that
actually
works.
The
only
difference
is
that
you
need
to
be
able
to
upgrade
services
that
are
offline.
A
A
Okay,
another
thing
just
wants
to
make
sure
that
it
is
possible
for
every
component
to
make
use
of
the
safe
ADM
tautology
task.
I
think
we
are
there
already.
So
if
you
have,
if
any
component
wants
to
make
use
of
this
FF
ADM
tautology
task
instead
of
the
safe
totality
task,
I
think
it's
already
working
more
or
less.
A
Regarding
the
communicate
Emmure,
you
asked
about
the
communication
protocol
of
Def
ADM.
It's
pure
SSH!
Well,
your
when,
when
you
spin
up
your
cluster,
you
will
make
sure
that
you
have
SSH
access
to
all
hosts
attached
to
that
to
the
cluster
and
then
deaf
ATM
will
do
remote
SSH
connections
and
then
execute
the
ADM
on
the
remote
host.
A
Yeah,
please
also
have
a
look
at
the
affirmations
right
now.
There
are
plenty
of
them
monalee,
mainly
regarding
all
those
themes
that
I've
tend
to
the
meaningless
user
experience.
Things
like
making
sure
everything
properly
works.
Everything
is
probably
checked
properly
proper
error
messages,
then
QA
enhancements,
QA
bugs
or
things
that
came
up
when
users
actually
testified.
Iam
like
lowing
custom,
URLs
photographer.
Now
you
for
the
Griffon
you're
sitting
in
the
dashboard,
so.
F
Thank
Sebastian,
one
of
the
note
about
the
use
of
self
ad
I
mean
more
speeds
like.
As
far
as
finding
looks
like
there
may
be
some
additional
bugs
to
work
out
with
the
pathology
integration
there.
You
know
that
I
could
be
good
to
see
what
what
happens
as
we
start,
trying
to
replace
the
use
of
the
yourself
tasks
with
the
Hadean
in
work
with
more
sweets
beyond
just
readers,
I
guess,
pressures
already
poking
at
rtw
and
everything
you
may
have
some
additional
team
and
said
maybe
not
be
well
covered.
F
F
A
Okay,
that's
classified
en
travis.
You
want
to
update
on
the
right
side,
yeah.
C
The
topic
to
bring
up
here
is
around
the
new
integration
test.
Juan
Miguel
had
for
the
manager,
those
tests
so
just
realized
that
had
emerged
yet,
but
it
was
ready
last
week,
so
it
is
merged
now
and
that
test
will
run
as
part
of
all
tough
rookie
ours
now
with
the
CI,
and
the
next
work
item
then,
will
be
to
create
a
separate
daily.
We
want
to
have
a
daily
job
that
runs
basically
tests
against
that
mastery
and,
oh,
my
god,
you
were
going
to
work
on
that.
That's
one
of
your
next
items.
G
G
C
C
G
The
30th
we
did
with
this
test
is
just
to
to
to
be
shortened
and
to
have
a
very
shown.
Audit
of
something
is
broken
in
the
in
the
interface
okay
and
there
are
very
basic
tests.
So
I
think
that
is
good
to
have
this
this
problem
man
and
to
be
a
left
of
this
problem
as
soon
as
possible
and
then
to
to
react
and
to
fix
the
problem
immediately.
Okay,
so
my
aim
is
just
to
to
take
a
look
every
day
on
this
test
tool
to
avoid
this
kind
of
problems.