strict allocate

[Björn Jacke]

Hi,
On 2009-12-01 at 09:33 -0800 Jeremy Allison sent off:
> On Tue, Dec 01, 2009 at 05:56:59PM +0100, Björn JACKE wrote:
> > It might be a good idea to change the default of strict allocate to "yes". This
> > makes out-of-quota situation less error prone and scenarios like in that article
> > will improve, too.
> > 
> > Has anyone strong feelings against that change of defaults?
> 
> Yes, don't do this. It's improves performance for some
> cases, but kills performance for others.

and it avoids samba's misbehaviour when quota limits are reached.


> It's an ext3-only
> fix as well, which doesn't work for ext4.

it fixes ext3 - and with ext4 we'll see no performance penalty at all when
posix_fallocate/fallocate would be used.

> Admittedly ext3
> is our most common filesystem for now, but that won't always
> be the case.

see above. With filesystems like ext4 or xfs we have no overhead for
preallocation of space.


> > In addition to that: When available, we could do allocation with
> > posix_fallocate(), which has the additional benefit that with recent
> > Linux
> > kernels and advanced filesystems, allocation is being done at
> > kernel/filesystem
> > layer via fallocate with quasi zero overhead.
> 
> Look at https://bugzilla.samba.org/show_bug.cgi?id=6942
> where there is a patch I did a while ago (modified since
> then I think) that changes strict allocate to a tri-state
> where the setting "partial" does most of what you want.

correct me if I'm wrong but it's mostly a workaround for the extreme
framenatations problem. it doesn't fix the we-run-out-of-quota situation and
the fallocate for ext4 and xfs is also not yet there.

Cheers
Björn