Hello,
I have some strange problems with large files on Client side. In the
concrete case, a file with a size of 4504526848 Bytes lies on my server (kernel
2.4.24). If I mount the directory with this file via nfs, the client
(kernel 2.4.24) informs me that the file size is only 209559552 Bytes. For
"debugging" I copied the file and watch the grow of the file size during
the copy job on both machines. Nothing suspicious happens until the file size
reached a level of about 4300000000 Bytes. The the client began counting
from 0 Bytes up to the 209559552; some form of overflow. On the server
side, nothing went wrong.
On a previous posting (http://sourceforge.net/mailarchive/
forum.php?thread_id=3874012&forum_id=4930) Vincent ROQUETA reports
problems with files that exceeds 4294967294 Bytes. Is there may be the same
underlying problem? BTW: 4294967294 + 209559552 = 4504526846 -> this is
very, very close to 4504526848; so can this be only luck?
Thanks!
cu,
Gerald.
-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs
Le Vendredi 13 F=E9vrier 2004 14:21, Gerald Stieglbauer a =E9crit :
> Hello,
>
> I have some strange problems with large files on Client side. In the
> concrete case, a file with a size of 4504526848 Bytes lies on my server
> (kernel 2.4.24). If I mount the directory with this file via nfs, the
> client (kernel 2.4.24) informs me that the file size is only 209559552
> Bytes. For "debugging" I copied the file and watch the grow of the file
> size during the copy job on both machines. Nothing suspicious happens unt=
il
> the file size reached a level of about 4300000000 Bytes. The the client
> began counting from 0 Bytes up to the 209559552; some form of overflow. On
> the server side, nothing went wrong.
> On a previous posting (http://sourceforge.net/mailarchive/
> forum.php?thread_id=3D3874012&forum_id=3D4930) Vincent ROQUETA reports
> problems with files that exceeds 4294967294 Bytes. Is there may be the sa=
me
> underlying problem? BTW: 4294967294 + 209559552 =3D 4504526846 -> this is
> very, very close to 4504526848; so can this be only luck?
>
> Thanks!
>
> cu,
> Gerald.
Well... In my bug, the size of the big file is not 4294967294, that is the=
=20
=2E.. user name and group name (Really stranges too... !!! )
I don't rember the size of my file, I know it was a little more than 2Go.
Vincent
-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs
On Fri, Feb 13, 2004 at 05:17:34PM +0100, Vincent ROQUETA wrote:
> Well... In my bug, the size of the big file is not 4294967294, that is the
> ... user name and group name (Really stranges too... !!! )
4294967294 == 2^32 - 2 == (uint32_t)(-2). Do a grep for -2 in
fs/nfsd/*.c, and you'll see where that number's coming from.
Nothing to do with the file sizes, in any case.
--Bruce Fields
-------------------------------------------------------
SF.Net is sponsored by: Speed Start Your Linux Apps Now.
Build and deploy apps & Web services for Linux with
a free DVD software kit from IBM. Click Now!
http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs