MIME-Version: 1.0
References: <Z73IsKBceoVd4t55@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z762lusJXp_wNF_z@paquier.xyz>
 <Z77jgvhwOu9S0a5r@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z8FMjlyNpNicucGa@paquier.xyz>
 <Z8GSY8mk+0qVHzsW@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z8ULFzNY24_SMCdu@paquier.xyz>
 <Z8WYf1jyy4MwOveQ@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z84xFrbmnW7Knb1Q@paquier.xyz>
 <Z878uhEaxYxRoXj7@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z8/T2DU208vq3rFc@ip-10-97-1-34.eu-west-3.compute.internal>
 <Z9Kl8UfC/GNExHDN@ip-10-97-1-34.eu-west-3.compute.internal>
 <CABPTF7VRN_-DfrPFjVtRYuN0E8mWETtVp6tSXNsXDJPDAuHHxA@mail.gmail.com>
In-Reply-To: 
 <CABPTF7VRN_-DfrPFjVtRYuN0E8mWETtVp6tSXNsXDJPDAuHHxA@mail.gmail.com>
From: Xuneng Zhou <xunengzhou@gmail.com>
Date: Thu, 13 Mar 2025 19:35:24 +0800
Message-ID: 
 <CABPTF7Vf_j7Eb1F3wuaFHr6r6m2_4imVzSKCYPkciN+Ti2eR2w@mail.gmail.com>
Subject: Fwd: [BUG]: the walsender does not update its IO statistics until it
 exits
To: PostgreSQL-development <pgsql-hackers@postgresql.org>
Content-Type: multipart/alternative; boundary="00000000000071b04b063037b863"
Archived-At: 
 <https://www.postgresql.org/message-id/CABPTF7Vf_j7Eb1F3wuaFHr6r6m2_4imVzSKCYPkciN%2BTi2eR2w%40mail.gmail.com>
Precedence: bulk

--00000000000071b04b063037b863
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Sorry, forgot to cc this one too.

---------- Forwarded message ---------
=E5=8F=91=E4=BB=B6=E4=BA=BA=EF=BC=9A Xuneng Zhou <xunengzhou@gmail.com>
Date: 2025=E5=B9=B43=E6=9C=8813=E6=97=A5=E5=91=A8=E5=9B=9B 19:29
Subject: Re: [BUG]: the walsender does not update its IO statistics until
it exits
To: Bertrand Drouvot <bertranddrouvot.pg@gmail.com>


Hi,
Given that PGSTAT_MIN_INTERVAL is the go-to setting for stats flushes
 * Unless called with 'force', pending stats updates are flushed happen onc=
e
 * per PGSTAT_MIN_INTERVAL (1000ms). When not forced, stats flushes do not
 * block on lock acquisition, except if stats updates have been pending for
 * longer than PGSTAT_MAX_INTERVAL (60000ms).
 *
 * Whenever pending stats updates remain at the end of pgstat_report_stat()
a
 * suggested idle timeout is returned. Currently this is always
 * PGSTAT_IDLE_INTERVAL (10000ms). Callers can use the returned time to set
up
 * a timeout after which to call pgstat_report_stat(true), but are not
 * required to do so.

 I am curious about the reason for this:


> 1. maybe relying on PGSTAT_IDLE_INTERVAL would make more sense? In both
> case
> PGSTAT_MIN_INTERVAL or PGSTAT_MIN_INTERVAL, I'm not sure there is a need =
to
> update the related doc.
>
>
PGSTAT_IDLE_INTERVAL seems to reduce the frequency even more.

>
> With the attached in place, the number of times the stats are being flush=
ed
> reduce drastically but still seems enough to get an accurate idea of the
> walsender
> IO activity. Thoughts?
>
>
>

--00000000000071b04b063037b863
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Sorry, forgot to cc this one too.<br><br><div class=3D"gma=
il_quote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">-----=
----- Forwarded message ---------<br>=E5=8F=91=E4=BB=B6=E4=BA=BA=EF=BC=9A <=
strong class=3D"gmail_sendername" dir=3D"auto">Xuneng Zhou</strong> <span d=
ir=3D"auto">&lt;<a href=3D"mailto:xunengzhou@gmail.com">xunengzhou@gmail.co=
m</a>&gt;</span><br>Date: 2025=E5=B9=B43=E6=9C=8813=E6=97=A5=E5=91=A8=E5=9B=
=9B 19:29<br>Subject: Re: [BUG]: the walsender does not update its IO stati=
stics until it exits<br>To: Bertrand Drouvot &lt;<a href=3D"mailto:bertrand=
drouvot.pg@gmail.com">bertranddrouvot.pg@gmail.com</a>&gt;<br></div><br><br=
><div dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr">Hi,<br>Given that PGSTA=
T_MIN_INTERVAL is the go-to setting for stats flushes<br>=C2=A0* Unless cal=
led with &#39;force&#39;, pending stats updates are flushed happen once<br>=
=C2=A0* per PGSTAT_MIN_INTERVAL (1000ms). When not forced, stats flushes do=
 not<br>=C2=A0* block on lock acquisition, except if stats updates have bee=
n pending for<br>=C2=A0* longer than PGSTAT_MAX_INTERVAL (60000ms).<br>=C2=
=A0*<br>=C2=A0* Whenever pending stats updates remain at the end of pgstat_=
report_stat() a<br>=C2=A0* suggested idle timeout is returned. Currently th=
is is always<br>=C2=A0* PGSTAT_IDLE_INTERVAL (10000ms). Callers can use the=
 returned time to set up<br>=C2=A0* a timeout after which to call pgstat_re=
port_stat(true), but are not<br>=C2=A0* required to do so.</div><div dir=3D=
"ltr"><br>=C2=A0I am curious about the reason for this:<br><br></div><div c=
lass=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margin:0px =
0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><br>=
1. maybe relying on PGSTAT_IDLE_INTERVAL would make more sense? In both cas=
e<br>PGSTAT_MIN_INTERVAL or PGSTAT_MIN_INTERVAL, I&#39;m not sure there is =
a need to<br>update the related doc.<br><br></blockquote><div><br></div><di=
v>PGSTAT_IDLE_INTERVAL seems to reduce the frequency even more.</div></div>=
</div><div class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D=
"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-le=
ft:1ex"><br>
With the attached in place, the number of times the stats are being flushed=
<br>
reduce drastically but still seems enough to get an accurate idea of the wa=
lsender<br>
IO activity. Thoughts?<br>
<br><br>
</blockquote></div></div>
</div></div>

--00000000000071b04b063037b863--