►
From YouTube: Ceph Crimson/SeaStor OSD 2020-07-29
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
Apart
from
it,
I
was
I'm
reviewing
sam's
fba
ping
yeah,
I'm
still
struggling
with
it.
A
A
A
A
B
Hey
everybody
I'm
doing
classic
scrubbing,
and
now
I
mean
I
already
updated
the
pad,
and
now
I
may
starting
to
run
the
standalone
stand-alone
test
suits
and
having
fellows
are
working
on
those
and
I'm
not
sure.
Yet
if
it's
well,
I
have
my
back
support
video,
but
also
I'm
probably
extracted
the
tests
might
have
made
progress
in
the
tests
themselves,
but
anyway
this
is
what
I'm
doing
and
working
daily
with
gabby
on
his
task.
D
A
A
C
Okay,
so
three
three
items
on
my
plate:
first
one
is
about
debugging
debugging,
the
thrashing
tests;
second,
one,
the
getting
back
to
the
to
the
sister
upgrade
thing
and
last
the
last
one
is
about
unit
tests
for
backfill.
The
infrastructure
is
ready.
I'm
up,
I'm
adding
third
test
case
be
would
be,
which
is
basically
a
quasi
randomized
test
that
will
generate
a
lot
of
a
lot
of.
C
Differences
between
rep
between
replicas
and
that
will
ensure
that
all
of
those
random
differences
are
nullified.
Focus
is
because
of
of
backfill.
This
is
quasi
random
because
thank
thankfully,
to
the
seat.
So
if
there
is
an
an
issue,
it
will
be
very
easy
to
to
repeat
to
reproduce
the
problem.
It's
quite
randomized,
it's
not
for
it's,
not
random.
C
A
C
C
And
the
the
sequence
will
be
much
much
longer
that
could
be
when
manually
generated.
That
will
be
the
main
difference
between
between
the
between
each
and
the
current
test
cases.
C
Which
means
that
I
will
need
to
okay,
I
will
need
to
add
one
small
piece
of
testing
of
the
unit
testing
infrastructure,
which
is
for
conveying
the
leads,
but
it's,
but
it's
pretty
straightforward.
The
most
of
the
mocks
and
fakes
are
already
there.
So
now,
they're,
not
a
huge
deal,
we'll
push
the
commit
soon,
and
will
your
job
won't
work
in
progress,
so
it
will
raise
too
much.
A
C
Okay,
let
me
let
you
it's.
The
infrastructure
is
pretty
straightforward.
Let
me
provide
a
link.
It
will
explain
more
than
a
thousand
words
one.
Second,
please
work
more
okay,
I've
got
the
dpr
and
getting
to
the
changes.
One
second.
C
All
you
all
you
need
to
do
is
to
define
actually
a
fake
object
stores
by
by
passing
by
unified
internalization,
with
with
list
of
strings
that
are
available
to
which
are
passable
to
the
hr
instances,
plus
the
e-version
d.
A
So
you
basically
mocked
everything
right.
C
A
Okay,
I'll
take
a
closer
look
at
it.
Thank
you.
Thanks.
Tommy
has
had
some
some
errands
to
run
tonight,
so
he
updated
me
with
his
status.
He
she
was.
He
was
updating
the
extensory
pr
according
to
sam's
comments
and
were
updated
to
accommodate
the
changes
from
according
to
sam's
lba
ping
patch
set,
which
I
I
I'm
also
reviewing.
A
I
pasted
it.
Let
me
find
that
backtrace
know.
A
I
I
picked
the
batteries
translated
by
the
address
to
line
script
in
the
in
the.
A
In
my
one
hour
appears,
I'm
posting
the
url
to
the
pi
in
the
comments,
so
it
should
be
something
related
to
to
the
structure
of
here.
I
I
I
speculate,
but
I
need
to
re
to
rerun
that
try
to
reproduce
the
issue
in
my
my
test
bed,
because
I
can
hardly
find
out
the
reason
why
we
we,
we
were
calling
slim
stream
column,
column
stream
in
in
remove
unscore
down,
underscore
peers.
A
A
E
Oh
sorry,
sorry,
I'm
late
this
week
I
was.