linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* 64bit inodes and IA32 apps over NFS
@ 2010-08-13 14:47 Lukas Hejtmanek
  2010-08-15  0:58 ` Trond Myklebust
  0 siblings, 1 reply; 2+ messages in thread
From: Lukas Hejtmanek @ 2010-08-13 14:47 UTC (permalink / raw)
  To: linux-nfs

Hello,

is there any workaround for NFSv4 when exporting XFS with 64bit inodes?

IA32 apps receives:
getdents(3, 0xb7571008, 1000)           =3D -1 EOVERFLOW (Value too lar=
ge for defined data type)

when compiling on x86_64, I got not error.

=46rom the XFS man page:
By default, with 32bit inodes, XFS places inodes only in the first 1TB =
of
a disk. If you have a disk with 100TB, all inodes will be stuck in the =
first
TB. This can lead to strange things like "disk full" when you still hav=
e
plenty space free, but there's no more place in the first TB to create =
a new
inode. Also, performance sucks.

To come around this, use the inode64 mount options for filesystems >1TB=
=2E
Inodes will then be placed in the location where their data is, minimiz=
ing
disk seeks.

Beware that some old programs might have problems reading 64bit inodes,
especially over NFS.


So, is there any workaround in NFS for IA32 apps?

--=20
Luk=E1=B9 Hejtm=E1nek

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: 64bit inodes and IA32 apps over NFS
  2010-08-13 14:47 64bit inodes and IA32 apps over NFS Lukas Hejtmanek
@ 2010-08-15  0:58 ` Trond Myklebust
  0 siblings, 0 replies; 2+ messages in thread
From: Trond Myklebust @ 2010-08-15  0:58 UTC (permalink / raw)
  To: Lukas Hejtmanek; +Cc: linux-nfs

On Fri, 2010-08-13 at 16:47 +0200, Lukas Hejtmanek wrote:
> Hello,
> 
> is there any workaround for NFSv4 when exporting XFS with 64bit inodes?
> 
> IA32 apps receives:
> getdents(3, 0xb7571008, 1000)           = -1 EOVERFLOW (Value too large for defined data type)
> 
> when compiling on x86_64, I got not error.
> 
> From the XFS man page:
> By default, with 32bit inodes, XFS places inodes only in the first 1TB of
> a disk. If you have a disk with 100TB, all inodes will be stuck in the first
> TB. This can lead to strange things like "disk full" when you still have
> plenty space free, but there's no more place in the first TB to create a new
> inode. Also, performance sucks.
> 
> To come around this, use the inode64 mount options for filesystems >1TB.
> Inodes will then be placed in the location where their data is, minimizing
> disk seeks.
> 
> Beware that some old programs might have problems reading 64bit inodes,
> especially over NFS.
> 
> 
> So, is there any workaround in NFS for IA32 apps?
> 

[trondmy@heimdal linux_nfs-2.6]$ less Documentation/kernel-parameters.txt

        nfs.enable_ino64=
                        [NFS] enable 64-bit inode numbers.
                        If zero, the NFS client will fake up a 32-bit inode
                        number for the readdir() and stat() syscalls instead
                        of returning the full 64-bit number.
                        The default is to return 64-bit inode numbers.

Trond

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2010-08-15  0:58 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-08-13 14:47 64bit inodes and IA32 apps over NFS Lukas Hejtmanek
2010-08-15  0:58 ` Trond Myklebust

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).