MIME-Version: 1.0
In-Reply-To: <13706.1510098347@sss.pgh.pa.us>
References: 
 <CABZYQR+Ku+iLFhqwY89QrrnKG9wKxckmssDG2rYKESojiohRgQ@mail.gmail.com>
 <20171107151108.3jtmta4xrgmuyrqf@alap3.anarazel.de>
 <CABZYQR+GE7mV7qiTFGO3EqxdqV_TmSQyzY3ghSE1OfHabZcG7g@mail.gmail.com>
 <20171107194517.pabhi4czbtiasb2f@alap3.anarazel.de>
 <CABZYQR+hepKxvDeMReZLcrYYmrhwFU3aX5qGG2QvDYT4V0wHDQ@mail.gmail.com>
 <CAOR=d=3t=y4A+5+iuSaPWj9zgKNin=9rf9+jvR4TAB+g_tXMVA@mail.gmail.com>
 <CABZYQRKGG6AohYi8U9aX+=iYidyWxCn8ratkiiN7YytmFJESNw@mail.gmail.com>
 <13706.1510098347@sss.pgh.pa.us>
From: =?UTF-8?Q?Ulf_Lohbr=C3=BCgge?= <ulf.lohbruegge@gmail.com>
Date: Wed, 8 Nov 2017 10:31:42 +0100
Message-ID: 
 <CABZYQRJHhYEGXP6W19GoXajC6Ne3N+eUsn3sn_iC3a91xC=QvA@mail.gmail.com>
Subject: Re: Slow execution of SET ROLE, SET search_path and RESET ROLE
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Scott Marlowe <scott.marlowe@gmail.com>,
 Andres Freund <andres@anarazel.de>,
	"pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Content-Type: multipart/alternative; boundary="94eb2c184a660a2f57055d755b5d"
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

--94eb2c184a660a2f57055d755b5d
Content-Type: text/plain; charset="UTF-8"

2017-11-08 0:45 GMT+01:00 Tom Lane <tgl@sss.pgh.pa.us>:

> =?UTF-8?Q?Ulf_Lohbr=C3=BCgge?= <ulf.lohbruegge@gmail.com> writes:
> > I just ran "check_postgres.pl --action=bloat" and got the following
> output:
> > ...
> > Looks fine, doesn't it?
>
> A possible explanation is that something is taking an exclusive lock
> on some system catalog and holding it for a second or two.  If so,
> turning on log_lock_waits might provide some useful info.
>
>                         regards, tom lane
>

I just checked my configuration and found out that "log_lock_waits" was
already enabled.

Unfortunately there is no log output of locks when those long running "SET
ROLE" statements occur.

Regards,
Ulf

--94eb2c184a660a2f57055d755b5d
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote">2017=
-11-08 0:45 GMT+01:00 Tom Lane <span dir=3D"ltr">&lt;<a href=3D"mailto:tgl@=
sss.pgh.pa.us" target=3D"_blank">tgl@sss.pgh.pa.us</a>&gt;</span>:<br><bloc=
kquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #cc=
c solid;padding-left:1ex"><span class=3D"">=3D?UTF-8?Q?Ulf_Lohbr=3DC3=3DBCg=
ge?=3D &lt;<a href=3D"mailto:ulf.lohbruegge@gmail.com">ulf.lohbruegge@gmail=
.com</a>&gt; writes:<br>
&gt; I just ran &quot;<a href=3D"http://check_postgres.pl" rel=3D"noreferre=
r" target=3D"_blank">check_postgres.pl</a> --action=3Dbloat&quot; and got t=
he following output:<br>
</span>&gt; ...<br>
<span class=3D"">&gt; Looks fine, doesn&#39;t it?<br>
<br>
</span>A possible explanation is that something is taking an exclusive lock=
<br>
on some system catalog and holding it for a second or two.=C2=A0 If so,<br>
turning on log_lock_waits might provide some useful info.<br>
<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 =C2=A0 regards, tom lane<br></blockquote><div><br></div><div>I just che=
cked my configuration and found out that &quot;log_lock_waits&quot; was alr=
eady enabled.</div><div><br></div><div>Unfortunately there is no log output=
 of locks when those long running &quot;SET ROLE&quot; statements occur.</d=
iv><div><br></div><div>Regards,</div><div>Ulf</div></div><br></div></div>

--94eb2c184a660a2f57055d755b5d--