Feedback-ID: i0fe9450f:Fastmail
Date: Wed, 1 Mar 2023 09:09:57 +0900
From: Michael Paquier <michael@paquier.xyz>
To: "Imseih (AWS), Sami" <simseih@amazon.com>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Subject: Re: Doc update for pg_stat_statements normalization
Message-ID: <Y/6X1WdVN+EOi23L@paquier.xyz>
References: <9CFF3512-355B-4676-8CCC-6CF622F4DC1A@amazon.com>
 <Y/2JecAVDcD/iIKQ@paquier.xyz>
 <8042E3F8-0BC8-4194-9F4C-32071389552A@amazon.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
	protocol="application/pgp-signature"; boundary="EGhBHrxkwPYXz5xM"
Content-Disposition: inline
In-Reply-To: <8042E3F8-0BC8-4194-9F4C-32071389552A@amazon.com>
Archived-At: <https://www.postgresql.org/message-id/Y/6X1WdVN%2BEOi23L%40paquier.xyz>
Precedence: bulk


--EGhBHrxkwPYXz5xM
Content-Type: multipart/mixed; boundary="FRtDZb5Zt4mPHzkc"
Content-Disposition: inline


--FRtDZb5Zt4mPHzkc
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Feb 28, 2023 at 11:11:30PM +0000, Imseih (AWS), Sami wrote:
> I agree. We introduce the concept of a plannable statement in a=20
> previous section and we can then make this distinction in the new
> paragraph.=20
>=20
> I also added a link to pg_stat_statements_info since that is introduced
> later on int the doc.

I have reworded the paragraph a bit to be more general so as it would
not need an update once more normalization is applied to utility
queries (I am going to fix the part where we mention that we use the
strings for utilities, which is not the case anymore now):
+  <para>
+   Queries on which normalization can be applied may be observed with cons=
tant
+   values in <structname>pg_stat_statements</structname>, especially when =
there
+   is a high rate of entry deallocations. To reduce the likelihood of this
+   happening, consider increasing <varname>pg_stat_statements.max</varname=
>.
+   The <structname>pg_stat_statements_info</structname> view, discussed be=
low
+   in <xref linkend=3D"pgstatstatements-pg-stat-statements-info"/>,
+   provides statistics about entry deallocations.
+  </para>

Are you OK with this text?
--
Michael

--FRtDZb5Zt4mPHzkc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v3-0001-doc-update-regarding-pg_stat_statements-normaliza.patch"
Content-Transfer-Encoding: quoted-printable

=46rom dd8938e4ba1b1f29d14b3fa2dc76301f42592cad Mon Sep 17 00:00:00 2001
=46rom: Michael Paquier <michael@paquier.xyz>
Date: Wed, 1 Mar 2023 09:05:08 +0900
Subject: [PATCH v3] doc update regarding pg_stat_statements normalization.

It is quite possible that a query text willl not normalize
(replace constants with $1, $2..) when there is a high rate
pgss_hash deallocation. This commit calls this out in docs.
---
 doc/src/sgml/pgstatstatements.sgml | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/doc/src/sgml/pgstatstatements.sgml b/doc/src/sgml/pgstatstatem=
ents.sgml
index efc36da602..f1ba78c8cb 100644
--- a/doc/src/sgml/pgstatstatements.sgml
+++ b/doc/src/sgml/pgstatstatements.sgml
@@ -516,6 +516,16 @@
    <structname>pg_stat_statements</structname> entry.
   </para>
=20
+  <para>
+   Queries on which normalization can be applied may be observed with cons=
tant
+   values in <structname>pg_stat_statements</structname>, especially when =
there
+   is a high rate of entry deallocations. To reduce the likelihood of this
+   happening, consider increasing <varname>pg_stat_statements.max</varname=
>.
+   The <structname>pg_stat_statements_info</structname> view, discussed be=
low
+   in <xref linkend=3D"pgstatstatements-pg-stat-statements-info"/>,
+   provides statistics about entry deallocations.
+  </para>
+
   <para>
    In some cases, queries with visibly different texts might get merged in=
to a
    single <structname>pg_stat_statements</structname> entry.  Normally thi=
s will happen
--=20
2.39.2


--FRtDZb5Zt4mPHzkc--

--EGhBHrxkwPYXz5xM
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----
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=rvYS
-----END PGP SIGNATURE-----

--EGhBHrxkwPYXz5xM--