On Thu, Jun 12, 2025 at 10:28 PM Fujii Masao
<masao.fujii@oss.nttdata.com> wrote:
> On 2025/06/07 0:13, Robert Treat wrote:
> > On Fri, Jun 6, 2025 at 9:57 AM David G. Johnston
> > <david.g.johnston@gmail.com> wrote:
> >> On Friday, June 6, 2025, Fujii Masao <masao.fujii@oss.nttdata.com> wrote:
> >>>
> >>> Hi,
> >>>
> >>> Since last_vacuum and vacuum_count in pg_stat_all_tables explicitly mention
> >>> that they don't include VACUUM FULL ("not counting VACUUM FULL"), I think
> >>> we should add the same clarification to the description of total_vacuum_time.
> >>> This field also excludes VACUUM FULL, and without this note, users might
> >>> mistakenly think the time spent on VACUUM FULL is included. Thought?
> >>>
> >>> <structfield>total_vacuum_time</structfield> <type>double precision</type>
> >>> </para>
> >>> <para>
> >>> - Total time this table has been manually vacuumed, in milliseconds.
> >>> + Total time this table has been manually vacuumed, in milliseconds
> >>> + (not counting <command>VACUUM FULL</command>).
> >>> (This includes the time spent sleeping due to cost-based delays.)
> >>> </para></entry>
> >>> </row>
> >>
> >>
> >> Makes sense. Our naming this table rewrite vacuum full does confuse people into thinking it is related to
vacuum.
> >>
> >
> > +1 for this change,
>
> Thanks both for the review!
>
>
> > but I think we should also update
> > n_ins_since_vacuum as well, no?
>
> I didn't update n_ins_since_vacuum since it's mainly used by autovacuum rather
> than end users, and there haven't been any complaints about the current
> description so far. That said, I don't have a strong opinion either way,
> so I'm fine with making the change if others think it's worthwhile.
>
Well, I admit I mostly mentioned it because when I noticed this one
wasn't documented the same way the other ones were, I second-guessed
myself about if I knew how it really behaved and did a quick test to
confirm :-)
I suspect others might have similar confusion.
Robert Treat
https://xzilla.net
