►
From YouTube: Ceph Crimson/SeaStore Meeting 2023-04-05
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/
A
All
right,
let's
get
started,
let's
see,
I
don't
have
much
reporting
Crimson
this
week,
I'm
gonna
be
working
I'm
working
at
Qs
at
the
moment,
that'll
probably
be
true
for
another
week
or
two
that
work
will
eventually
transfer
over
to
Crimson
now
so
worthwhile
harsh
how's
it
going.
B
Okay,
so
a
quick
update
from
the
QE
side
we
have
released,
or
you
could
say,
file
two
more
bucks
with
Crimson.
One
is
related
to
the
OSD
logs,
not
having
log
level
20
entries.
So
basically,
what
happens
is
if
you,
if
you
enable
the
debugging
to
20
on
both
MGR
and
osts,
the
OSD
logs
do
not
contain
the
entries
for
log
level,
15,
10,
15
or
20,
whereas
the
MGR
logs
to
contain
all
of
that.
So
that
has
been
raised.
B
It
was
previously
observed,
like
two
two
months
back
when
I
was
initially
doing
the
testing
with
the
basic
rathers
test,
but
it
wasn't
raised
back
then,
and
I
noticed
it
again
during
the
last
week
when
I
was
doing
the
testing
again
another
round
of
rado's
test.
So
that
has
been
raised
and
there
is
another
new
one
that
I
believe
is
a
potential
bug.
It
is
that
pool
compressions
do
not
take
effect.
B
So
if
we
have
compression
enabled
on
a
few
pools,
it
could
be
Snappy
or
c-lib
the
can.
B
We
just
run
rados
bench,
the
amount
of
data
which
gets
written
and
the
number
of
objects
with
which
gets
written
on
these
pools,
as
well
as
a
pool
where
no
compression
is
enabled
is
exactly
the
same,
whereas
on
a
downstream
or
you
could
say,
a
red
hat,
CEF
cluster,
the
amount
of
data
written
as
well
as
the
number
of
objects
written
on
the
pools
which
have
compression
and
which
are
having
no
compression,
is
pretty
much
different.
B
So
both
of
these
issues
have
been
raised
in
the
option
tracker
itself,
one
additional
update
is
that
are
Downstream.
Ti
framework
is
now
supporting
Crimson,
so
that
has
been
another
development
where
our
Downstream
CI
framework
can
now
run
tests
with
the
Crimson
build.
So
every
every
test
that
we
have
in
our
CI
pipeline
right
now,
we'll
be
running
on
the
Crimson,
builds
during
the
upcoming
weeks
and
yep.
That's.
B
And
yeah
there's
one
more
thing:
I
do
understand
that
there
might
be
a
lot
of
failures,
so
I'll
be
mindful
to
not
release
bugs
which
are
not
applicable
to
Crimson
as
of
now,
because
we
would
probably
have
a
lot
of
tests
which
would
not
be
applicable
to
Crimson
as
of
yet
so.
A
A
A
Yes,
but
I'm
forgetting
what
it
is.
You
should
ask
knit
some
I
think
what
we're
doing
is
We're,
translating
the
debug
log
levels
that
we
would
normally
set
for
classic
into
either
Trace
warn
error
or
debug,
so
debug
20
should
get
you
debug
logs.
You
should
see
log
lines
that,
with
the
word
debug
in
them
towards
the
beginning,
whereas
with
debug
of
xero,
you
should
really
only
see
stuff,
that's
error
or
worn
debug
I'll.
A
Out
yeah,
it's
a
little
confusing
to.
A
A
B
There
is
a
follow-up
question
from
Milton
I'm,
not
sure
if
I'm
pronouncing
his
name
correctly,
but
he
has
asked
me
to
share
the
results
when
the
max
object
is
set.
So
I'll
follow
up
with
him
as
well.
With
this
results,
I
do
not
have
a
cluster
with
rados
configuration
yet
so
I'll
share
these
results
in
the
in
this
week
itself.
A
Cool
wow,
it's
a
ton
of
progress,
nice
yep,
yeah
all
right
show
me
how's
it
going.
C
Last
week,
I
updated
the
medical
system
up
here
and
go
to
English
new
round
review
comments
and
just
you're
working
on
the
comment.
It
seems
almost
a
finished
review.
Every
part
is
reviewed
just
to
the
final
modification
and
I
still
debugging
under
medical
oski,
I.
Think
a
30
code
is
debugging
the
industry
or
working
on
it.
So
that's
all.
C
Today
is
Pierce
quality,
Pierce
holiday,
and
this
is
kickoff
for
the
holiday.
A
Everyone's
having
a
good
time,
talk
to
you
guys
later
bye.