Re: Faulty HEAP_XMAX_LOCK_ONLY & HEAP_KEYS_UPDATED hintbit combination

On Mon, Jan 25, 2021 at 12:01 AM Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
>
> On 2021-Jan-24, Julien Rouhaud wrote:
>
> > + /*
> > + * Do not allow tuples with invalid combinations of hint bits to be placed
> > + * on a page. These combinations are detected as corruption by the
> > + * contrib/amcheck logic, so if you disable one or both of these
> > + * assertions, make corresponding changes there.
> > + */
> > + Assert(!((tuple->t_data->t_infomask & HEAP_XMAX_LOCK_ONLY) &&
> > + (tuple->t_data->t_infomask2 & HEAP_KEYS_UPDATED)));
> >
> >
> > I attach a simple self contained script to reproduce the problem, the last
> > UPDATE triggering the Assert.
> >
> > I'm not really familiar with this part of the code, so it's not exactly clear
> > to me if some logic is missing in compute_new_xmax_infomask() /
> > heap_prepare_insert(), or if this should actually be an allowed combination of
> > hint bit.
>
> Hmm, it's probably a bug in compute_new_xmax_infomask. I don't think
> the combination is sensible.

Yeah, the combination clearly doesn't make sense, but I'm wondering
what to do about existing data? Amcheck.verify_am will report
corruption for those, and at least all servers where powa-archivist
extension is installed will be impacted.