> "Hiroshi Inoue" <Inoue@tpf.co.jp> writes:
> >>>> I vacuumed here and it worked. I did not use my "old"
> pg_log file - what
> >>>> did I lose?
> >>
> >> Hard to tell. Any tuples that weren't already marked on
> disk as "known
> >> committed" have probably gone missing, because their originating
> >> transaction IDs likely won't be shown as committed in the
> new pg_log.
> >> So I'd look for missing tuples from recent transactions in
> the old DB.
> >>
>
> > Hmm,this may be more serious.
> > MVCC doesn't see committed(marked ) but
> > not yet committed(t_xmin > CurrentTransactionId) tuples.
> > He will see them in the future.
Yes, good point. Is there a way to set CurrentTransactionId to a value
greater that the smallest t_xmin ?
>
> But he did a vacuum --- won't that get rid of any tuples that aren't
> currently considered committed?
He said that the vacuum was blocking and he thus killed it.
The vacuum was thus only partway done.
Andreas
