Re: New vacuum option to do only freezing

On Wed, Mar 27, 2019 at 12:12 AM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> wrote:
>
> On Sat, Mar 23, 2019 at 3:25 AM Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
> >
> > On Fri, Mar 8, 2019 at 12:14 AM Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com> wrote:
> > > IIUC we've discussed the field-and-value style vacuum option. I
> > > suggested that since we have already the disable_page_skipping option
> > > the disable_page_skipping option would be more natural style and
> > > consistent. I think "VACUUM (INDEX_CLEANUP false)" seems consistent
> > > with its reloption but not with other vacuum options. So why does only
> > > this option (and probably up-coming new options) need to support new
> > > style? Do we need the same change to the existing options?
> >
> > Well, it's too late to change to change DISABLE_PAGE_SKIPPING to work
> > some other way; it's been released, and we're stuck with it at this
> > point.
>
> Agreed.
>
> > However, I generally believe that it is preferable to phrase
> > options positively then negatively, so that for example one writes
> > EXPLAIN (ANALYZE, TIMING OFF) not EXPLAIN (ANALYZE, NO_TIMING). So
> > I'd like to do it that way for the new options that we're proposing to
> > add.
>
> Agreed with using phrase options positively than negatively. Since
> DISABLE_PAGE_SKIPPING is an option for emergency we might be able to
> rename for consistency in a future release.
>
> Attached updated version patches.

The patch adds the basic functionality to disable index cleanup but
one possible argument could be whether we should always disable it
when anti-wraparound vacuum. As discussed on another thread[1]
anti-wraparound vacuum still could lead the I/O burst problem and take
a long time, especially for append-only large table. Originally the
purpose of this feature is to resolve the problem that vacuum takes a
long time even if the table has just a few dead tuples, which is a
quite common situation of anti-wraparound vacuum. It might be too late
to discuss but if we always disable it when anti-wraparound vacuum
then users don't need to do "VACUUM (INDEX_CLEANUP false)" manually on
PostgreSQL 12. Dose anyone have opinions?

[1] https://www.postgresql.org/message-id/CAC8Q8t%2Bj36G_bLF%3D%2B0iMo6jGNWnLnWb1tujXuJr-%2Bx8ZCCTqoQ%40mail.gmail.com

Regards,

--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center