| From: | Fujii Masao <masao(dot)fujii(at)gmail(dot)com> |
|---|---|
| To: | PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | vacuumdb --missing-stats-only and permission issue |
| Date: | 2025-08-21 03:06:32 |
| Message-ID: | CAHGQGwHh43suEfss1wvBsk7vqiou=UY0zcy8HGyE5hBp+HZ7SQ@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
While following the discussion on vacuumdb --missing-stats-only [1],
I noticed that this option queries pg_statistic and pg_statistic_ext_data.
As a result, non-superusers like pg_maintain cannot use it because
by default they lack permission to access those catalogs.
I'm not sure whether --missing-stats-only was intended to work for
non-superusers, but if so, this restriction is inconvenient. Would it
make sense to use the views pg_stats and pg_stats_ext instead?
Since the catalogs are only consulted to check whether statistics exist,
the views should be sufficient. Thought?
Regards,
[1] https://postgr.es/m/20250820104226.8ba51e43164cd590b863ce41@sraoss.co.jp
--
Fujii Masao
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Zhijie Hou (Fujitsu) | 2025-08-21 03:41:30 | RE: Fix replica identity checks for MERGE command on published table. |
| Previous Message | Michael Paquier | 2025-08-21 02:57:58 | Re: Sequence Access Methods, round two |