►
From YouTube: Ceph Crimson/SeaStore Meeting 2023-04-26
Description
Join us weekly for the Ceph Crimson/Seastore meeting: https://ceph.io/en/community/meetups
Ceph website: https://ceph.io
Ceph blog: https://ceph.io/en/news/blog/
Contribute to Ceph: https://ceph.io/en/developers/contribute
What is Ceph: https://ceph.io/en/discover/
B
Yeah
hi
everyone.
My
name
is
June
season
and
I'm
working
for
Hari.
Now
so
you
know,
I
just
attended
severeign.
B
A
A
All
right
never
mind.
Let's
see,
let's
get
started
this
week
for
me:
I've
been
catching
up
on
reviews
after
cephalicon,
mostly
engine
how's,
it
going.
C
Yeah
this
week,
the
mod
core
system
and
the
connection
reference
change.
One
marched
thanks
for
Sam
and
I,
have
reviewed
and
emerged
the
circular
Journal
space
for
random
block
manager
and
for
the
mod
core
messenger.
This
socket
accepts
and
connects
on
multiple
cores
and
the
unit
tests
are
done
and
the
current
challenge
is
to
make
sure
everything
in
memory
is
correctly
allocated.
The
allocated,
read
and
write
in
the
same
call
I'm
working
on
the
internal
incrementation,
but
before
OSD
I
see
that
the
cross
Cloud
constraint
for
OSD
operation
message.
C
Buffer
list
might
not
be
enforced
quite
correctly,
but
somehow
it
is
not
triggering
any
faults
which
looks
weird
and
the
last
I
will
skip
the
next
stand
up
for
holiday
and
that's
all
for
me.
A
A
A
Yeah
it
may,
it
may
become
a
performance
problem,
but
as
long
as
we're
not
messing
with
the
ref
counts
frequently,
it
may
never
be
worth
fixing.
So
it's
probably
worth
just
profiling
and
seeing
what
happens.
Yeah.
D
This
week,
I
already
did
clone
VR
first
historical
NPR
and
I,
suddenly
two
small
PRS
to
fix
the
corresponding
box,
with
with
respect
to
the
snapshot
functionality
in
crimson
right
now,
I'm
working
on
modifying
the
lva
point
of
PR
as
Quincy
and
Sam
suggested.
That's
all
from
me.
A
Okay
on
that
PR
I
think
when
I
went
over
last
time,
the
internal
logic
seemed
fine.
So,
from
my
perspective,
once
we're
satisfied
with
the
passing
testing
and
engine
is.
A
C
A
All
right
should
we.
E
So
last
week
a
medical
assistant,
PR
got
merged
and
Marco
estate
PR
is
provided
so
I
found
in
debug
build
the
the
code
for
the
get
connection
pipeline
to
the
enter.
And
if
the
width
barrier
is
a
sister
returned
the
sister
now,
so
it
can't
guarantee
the
sequence
for
the
io
request,
because
when
the
iteration
system
now
the
continuation,
the
then
continuation
is
scheduled,
not
exact,
not
executed
immediately
for
some
IO
request
and
so
it.
But
the
release
will
know
this
issue.
A
A
E
E
I
enabled
the
the
SMP
for
consolid
testing
and
with
this
PR
SMP,
set
to
three
and
cefca
test
that
has
the
same
result
as
without
this
PR,
so
I'm
not
sure
how
to
check
it.
It's
enabled
the
three
Sharda
how
to.
A
I
see
I
think
I
added
a
debug
output
during
the
startup
process.
That
indicates
the
number
of
cores
that
are
running.
You
would
also
see
debug
output.
That
would
indicate
that
it's
running
on
something
other
than
core
zero.
Each
debug
line
I
believe
indicates
which
core
it's
running
on,
so
you
can
probably
check
it
that
way
too.