buffindexed, a racing problem?

[Kiernan, Alex]

> >> Could this be fcntl(F_SETLKW) returning with EINTR? It 
> looks to me like
> >> inn_lock_range will return in this instance without 
> acquiring the write
> >> lock - I've spent an hour or so staring a ovblocknew() to see if I
> >> could see any holes and this is all I can come up with.
> 
> > Good remark.
> 
> Sees like inn_lock_range should retry until it doesn't get EINTR.
> 

Just committed a change for that.

I'd really like this to fix it as I'd really like to be able to switch to
buffindexed.

-- 
Alex Kiernan, Principal Engineer, Development, THUS plc