skiplist db corruption

[Robert Mueller]

> We've seen corruption caused by 2 diferent sources:
>
> - fail on allocation of memory: linux can negate mmap's with plenty of 
> memory free. There was a patch to use remap instead of unmap / mmap again, 
> but we never tested it;
>
> - there is some sort of race, that shows up in (real) smp boxes (dont 
> count HT in smp);

We haven't seen either of these problems, and we run SMP systems and kernels 
(mostly dual x HT processors, so 4 logical processors). It's usually more 
system crash related, though we have seen it even after "clean" restarts of 
cyrus (eg KILL -15 masterpid). It annoying since the files are supposed to 
be "transaction safe".

Rob