►
From YouTube: 2019-Jan-09 :: Ceph Testing Weekly
Description
Weekly collaboration call of all community members working on Ceph Testing.
http://ceph.com/testing
A
B
C
C
C
Hey
Lea
red
had
had
a
meeting
yesterday
with
our
I'm.
C
One
of
our
QE
leads
to
talk
about
how
we
want
to
do
Brooke
testing,
unfortunately,
I
didn't
get
as
far
along
as
I
wanted
to
for
it,
or
even
for
this
reading,
but
I'm
planning
today
to
get
an
actual
interface
written
out
for
to
ecology
deployers,
the
one
we
have
right
now
being
you
know
our
SS
age
based
thing
that
goes
into
place
F
and
manages
the
demons
and
then
the
existing
ansible
one
that
doesn't
actually
satisfy
quite
the
same
interface,
but
I
want
to
write
down
what
the
interface
is,
that
we
need
to
be
fully
compatible.
C
C
C
Vos,
you
asked
for
it
because
oh
we're
getting
rid
like
you
know
we're
getting
ready
to
talk
about
it
from
our
downstream
product
perspective,
and
so
it's
coming
into
his
plate.
Okay,
I
got
it
thy
mother
upstream,
build
people
and
stuff
earth
not
upstream
to
downstream
and
upstream
RedHat
del
people.
Nothing.
C
C
Mail
scepter
well
as
soon
as
I.
Do
we
haven't
done
anything
reels
here?
You
said
we
were
just
I
actually
did
internal
II
were
just
like
dark.
You
wanted
to
know
what
rook
does
in
the
Oscar
Newman.
C
D
A
C
Need
to
allocate
enough
time
to
actually
it
could
happen,
but
I
think
it's
gonna
happen.
Yeah
a
major
portion,
I
think
I,
don't
know
about
Susan
but
like
brett,
has
people
III
think
you've
gotten
tired
of
not
doing
a
den
testing
with
our
sensible
and
we've
seen
enough
issues
with
not
testing
that
and
then
that
we
don't
really
want
to
do
that
on
our
next
one.
C
C
So
I
mean
my
son
should
be
that
both
technology
and
the
and
the
manager
Orchestrator
would
be
issuing
cube.
Cuddle
commands,
I,
get
and
quitting
at
the
thing.
So
yes,
that
once
we
haven't
set
up
I
would
I
even
have
to
work.
Fine
yeah,
that's
awesome!
That's
awesome
and
I.
Don't
know
if
we'll
be
able
to
share
any
of
that
code.
I
suspect
that,
but
maybe
I
feel
like
playing
track.
D
D
D
C
D
D
A
C
C
A
A
C
A
A
C
D
Yeah
one
thing:
do
we
have
documentation
how
to
Oh
what
the
what
a
workflow
of
testing
upstream
pull
requests?
I
know
that
has
in
turn
into
an
intern
recommendation
or
to
do
that,
but
is
there
an
don't
know,
upstream
version
of
of
a
workflow
how
to
push
things
to
SEF
CI
and
then
trigger
trigger
tautology?
Bye-Bye
vodka
samples
on
some
example
lines.
B
We
don't
have
that,
you
know
we
have
like
I'm
using
kind
of
process
and
I
have
a
cheat
sheets
in
terms
of
how
to
do
things,
and
we've
talked
about
that
and
I
think
Josh
wanted,
maybe
to
automate
some
stuff
around
things.
But,
generally
speaking,
you
know,
we
don't
have
it.
Why
would
we
want
to
have
it?
You
know
you
want
to
more
people
be
able
to
test
it.
A
A
A
Have
I
have
created
inside
our
organization
a
Jenkins
job
which
is
similar
to
technology,
just
instantiate
it
a
new
instance
of
technology
in
OpenStack,
like
NACP
and
and
all
and
it's
possible
to
use
of
age,
and
it's
run
like
basic
tests
like
dummy
or
some
soozee
soozee
dummy
suite,
which
we
have
only
downstream
recently
and
nathan,
is
working
all
on
the
deep-sea
job.
Very.
A
Mother
sighed
I'm
working
on
the
ansible
stuff
to
make
it
possible
to
deploy
it.
It
ology
to
OpenStack
openSUSE
based
distros
and
as
far
as
I
have
a
ansible
at
the
at
some
point.
Working
groups
I
will
be
able
to
start
preparing
that
pull
request
for
upstream.
So
we
can
use
this,
or
rather
all
testing
of
digital
needs
of.
D
A
A
C
For
the
totality
testing
I
think
Sebastian
was
asking
about
the
set
testing.
Workflow
and
I
mean
yeah,
maybe
about
three
neurons
passed
on
as
people
need
them,
but
there's
a
very
broad
overview
about
how
stuff
gets
merged
like
yes,
okay,
you
submit
a
PR
and
then,
when
a
reviewers
happy
with
it,
people
merge
it
into
their
integration
branches
for
testing,
and
when
that
passes
then
emerges
and
yeah
and
add
more
specific
questions.
Sebastian,
then
I
guess:
are
you
worried
about
setting
up
a
workflow
on
the
orchestrators
that
matches
the
rest
of
the
project
or
yeah.
C
So
fine,
if
you
have
questions
that
means
that
we
probably
well
it
might
just
be
nice
to
have
a
conversation
on
the
mailing
list
or
higher
seer
whatever,
but
it
might
also
mean
that
we
need
to
write
more
stuff
down
and
so
good
like.
If
you
can
write
down
your
questions,
then
we
can
maybe
like
discuss
them
and
then
prepare
document
with
that.
Okay,
we
do
like
you
know.
We
don't
know
what
we
need
to
write
down,
because
we
don't
need
to
written
down
anymore.
B
B
B
It's
a
little
bit
tricky
because
you
know
you
need
to
know
right
now
we
don't
have
real
concrete
system
in
terms
of
you
know.
How
do
you
know
what
to
run,
and
we've
been
talking
about
like
that,
if
people,
if
we
formalize
it
and
if
we
say
that
like
every
PR,
has
a
pretty
good
kind
of
label
that
indicates
what
to
run.
That
would
be
useful,
but
that's
the
sense
of
the
old
process:
okay,
yeah
and
then
your
look
at
it
and
then
your
prove
it
I
think
I.
B
Think
people
been
like
chatting
for
a
while
for
mini
like
several
years
how
to
automate
it.
How
can
you
do
that
stuff
without
human
being
interaction
like,
for
example,
on
my
part,
is
probably
most
of
my
time
going
through
this
process
and
testing
them,
and
once
once
we
go
through
integration
like
bunch
of
pr's
that
consider
to
be
for
next
release.
As
we
say,
integration
testing
is
done
and
you
release
it
for
qee
validation
and
that's
again
doing
that.