12:45 p.m.
Hi,
I noticed that FID on OST is enabled since 2.4. With it, it seems the
OI files on OST are not used, unless old client connect w/o
OBD_CONNECT_FID flag, is it correct?
Just want to confirm my understanding of the code.
BTW, is there any documentation of FID on OST? I see that LUDOC-81 is
open but no patch submitted.
Thanks,
Tao
6:55 p.m.
I'm also curious about the FID on OST work.. I had thought the OST
should be creating files under the /oi.<N> directory, but that's not the
case for the file systems I've looked at here at LLNL. The files are all
under /O/0/d<N> on the OST.
--
Cheers, Prakash
On Wed, Jul 31, 2013 at 06:45:11PM +0800, Peng Tao wrote:
Hi,
I noticed that FID on OST is enabled since 2.4. With it, it seems the
OI files on OST are not used, unless old client connect w/o
OBD_CONNECT_FID flag, is it correct?
Just want to confirm my understanding of the code.
BTW, is there any documentation of FID on OST? I see that LUDOC-81 is
open but no patch submitted.
Thanks,
Tao
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
7:03 p.m.
that's because we wanted to make MDS objects visible under "normal"
namespace,
so we can't put them under /O. on the other hand, we wanted to make OST objects
visible via ZPL as well. so here is this trick with two different OIs.
thanks, Alex
On Jul 31, 2013, at 8:55 PM, Prakash Surya <surya1(a)llnl.gov> wrote:
I'm also curious about the FID on OST work.. I had thought the
OST
should be creating files under the /oi.<N> directory, but that's not the
case for the file systems I've looked at here at LLNL. The files are all
under /O/0/d<N> on the OST.
--
Cheers, Prakash
On Wed, Jul 31, 2013 at 06:45:11PM +0800, Peng Tao wrote:
> Hi,
>
> I noticed that FID on OST is enabled since 2.4. With it, it seems the
> OI files on OST are not used, unless old client connect w/o
> OBD_CONNECT_FID flag, is it correct?
>
> Just want to confirm my understanding of the code.
>
> BTW, is there any documentation of FID on OST? I see that LUDOC-81 is
> open but no patch submitted.
>
> Thanks,
> Tao
> _______________________________________________
> HPDD-discuss mailing list
> HPDD-discuss(a)lists.01.org
> https://lists.01.org/mailman/listinfo/hpdd-discuss
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
--------------------------------------------------------------------
Closed Joint Stock Company Intel A/O
Registered legal address: Krylatsky Hills Business Park,
17 Krylatskaya Str., Bldg 4, Moscow 121614,
Russian Federation
This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
4:42 a.m.
FID-on-OST is only enabled with DNE (i.e. if there are multiple MDTs). Otherwise, it
continues to use FID_SEQ_OST_MDT0 (== 0) on the OSTs for now.
Even when FID-on-OST is active, the OST objects are still created under O/{seq}/ since
they are not in the client visible namespace.
Cheers, Andreas
On 2013-07-31, at 4:46, "Peng Tao" <bergwolf(a)gmail.com> wrote:
I noticed that FID on OST is enabled since 2.4. With it, it seems
the
OI files on OST are not used, unless old client connect w/o
OBD_CONNECT_FID flag, is it correct?
Just want to confirm my understanding of the code.
BTW, is there any documentation of FID on OST? I see that LUDOC-81 is
open but no patch submitted.
Thanks,
Tao
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
3214
days inactive
3215
days old
3 comments
4 participants
participants (4)
-
Alexey Zhuravlev -
Dilger, Andreas -
Peng Tao -
Prakash Surya