| From: | David Gould <daveg(at)sonic(dot)net> |
|---|---|
| To: | Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> |
| Cc: | pgsql-bugs(at)lists(dot)postgresql(dot)org, Alina Alexeeva <alexeeva(at)adobe(dot)com>, Rajesh Bhatia <rajbhati(at)adobe(dot)com>, Arvind Kota <akota(at)adobe(dot)com>, Jothikanth R <jothikanth(dot)r(at)efrontier(dot)com>, Ullas Lakkur Raghavendra <lakkurra(at)adobe(dot)com> |
| Subject: | Re: BUG #15005: ANALYZE can make pg_class.reltuples inaccurate. |
| Date: | 2018-01-23 07:52:24 |
| Message-ID: | 20180122235224.22bf654d@engels |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
On Thu, 11 Jan 2018 11:39:22 -0300
Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> wrote:
> David Gould wrote:
>
> > I've tracked down how this happens and created a reproduction script and a
> > patch. Attached:
> >
> > - vac_estimate_reltuples.patch Patch against master
> > - README.txt Instructions for testing
> > - reltuples_analyze_bug.sql Reproduction script
> > - analyze_counts.awk Helper for viewing results of test
> > - test_standard.txt Test output for unpatched postgresql 10.1
> > - test_patched.txt Test output with patch
>
> Great work. Please make sure to register this as a bug fix in the
> next commitfest.
>
Done last week. See:
https://www.postgresql.org/message-id/flat/20180117164916(dot)3fdcf2e9(at)engels/
I think this should be considered for back-patching, the bug exists in all
versions back to 9.4 (and probably earlier, but I did not test those).
-dg
--
David Gould daveg(at)sonic(dot)net
If simplicity worked, the world would be overrun with insects.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kyotaro HORIGUCHI | 2018-01-23 08:05:11 | Re: Bug in Physical Replication Slots (at least 9.5)? |
| Previous Message | Amit Kapila | 2018-01-23 02:09:22 | Re: Problem running post-install step. |