RE: OT: cluster size=0

[Paul Bunn <Paul UltraBac com>]

Thanks for the elucidation.
We had a customer using our software on Win32 to push a backup across the wire to
be received on an Intel linux system.  Once the software had written to 2GB to a
file it would fail.  Our solution at that time was to allow "subfile" writing
where the backups are broken into smaller files to cater for that limitation.  It
was this (and reading something Alan Cox had written a couple of years ago about
EXT2's limitations) that led me to believe it was the FS itself that was
responsible for the filesize limit.

Regards,

Paul Bunn, UltraBac.com, 425-644-6000
Microsoft MVP - WindowsNT/2000
http://www.ultrabac.com
 

-----Original Message-----
From: Liam Bedford [mailto:lbedford@lbedford.org]
Sent: Sunday, March 25, 2001 8:37 AM
To: 'axp-list@redhat.com'
Subject: Re: OT: cluster size=0


On Sat, Mar 24, 2001 at 12:07:51PM -0800, Paul Bunn came forth with:
> FAT/FAT32 have a 2^32 file-size limitation (4GB).  EXT2 has a 2^31 (2GB)
file-size
> limitation.
bzzt.. wrong. Ext2 doesn't have a 2Gb file-size limitation. The VFS layer on
32-bit archs in 2.2.blah (can't remember which one they fixed it in) does.

it affects more than just ext2, and it's why the alpha doesn't have that
problem..