Horrible performance of samba-3.0pre expected?

[Jeremy Allison]

Matt Zinkevicius wrote:
> 
> Howdy all,
> We have been getting miserable performace out of samba 3.0pre snapshots for
> the last month. Is this to be expected since this is still beta code? Are
> there any debug switches, etc that I could disable. Our netbench results
> show samba 3.0pre (May 2nd snapshot of MAIN branch) to be roughly 40% slower
> than 2.0.7. I know this is a development branch, and we shouldn't expect
> much from it, but this seems like extreme behavior. Anyone else have
> similar/different benchmark results?

This is good to know. There are many changes that
could affect this, probably the addition of the two
byte range lock databases to give the completely correct
byte range and POSIX lock maps.

Can you add the line "posix locking=False" and let me
know if this makes a difference ? I'm going to be spending
some time in the SGI lab next week to look at exactly
this issue, so it would be good to know what to look at.

Any chance of a profile output ?

Regards,

	Jeremy Allison,
	Samba Team.

-- 
--------------------------------------------------------
Buying an operating system without source is like buying
a self-assembly Space Shuttle with no instructions.
--------------------------------------------------------