On E, 2005-05-23 at 11:42 -0400, Tom Lane wrote:
> Hannu Krosing <hannu@skype.net> writes:
> > I can't think of any other cases where it could matter, as at least the
> > work done inside vacuum_rel() itself seema non-rollbackable.
>
> VACUUM FULL's tuple-moving is definitely roll-back-able, so it might be
> prudent to only do this for lazy VACUUM. But on the other hand, VACUUM
> FULL holds an exclusive lock on the table so no one else is going to see
> its effects concurrently anyway.
I'm not interested in VACUUM FULL at all. This is improvement mainly for
heavy update OLAP databases, where I would not even think of running
VACUUM FULL.
I'll cheks if there's an easy way to exclude VACUUM FULL.
> As I said, it needs more thought than I've been able to spare for it yet
> ...
Ok, thanks for comments this far .
--
Hannu Krosing <hannu@skype.net>