Content-Type: text/plain; charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\))
Subject: Re: Disk filled-up issue after a lot of inserts and drop schema
From: Pietro Pugni <pietro.pugni@gmail.com>
In-Reply-To: <16009.1473875751@sss.pgh.pa.us>
Date: Wed, 14 Sep 2016 20:00:59 +0200
Cc: Rick Otten <rottenwindfish@gmail.com>,
 pgsql-performance@postgresql.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <F4AC0CD0-0BCA-493F-87B0-E6AD169B4EB5@gmail.com>
References: <AEA46136-0A29-4F3C-8966-90178EE44B4B@gmail.com>
 <CAMAYy4K6nwkqqdRcWR4cNeYvCXAH7h4WAo6vq1AkX3+MYOezsw@mail.gmail.com>
 <28460.1473864279@sss.pgh.pa.us>
 <FE960B29-7786-4A97-9288-4346F48A21A7@gmail.com>
 <16009.1473875751@sss.pgh.pa.us>
To: Tom Lane <tgl@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

Log rotation was active and set to 5MB or 1 day.
I don=E2=80=99t know if it is a bug, but Postgres was logging even if loggi=
ng_collector was set to =E2=80=9Coff=E2=80=9D.
Also, that big log file wasn=E2=80=99t visible for me, in fact =E2=80=9Cls=
=E2=80=9D and =E2=80=9Cdu=E2=80=9D didn=E2=80=99t detect it.

Thanks again

Best regards,
 Pietro Pugni

> Il giorno 14 set 2016, alle ore 19:55, Tom Lane <tgl@sss.pgh.pa.us> ha sc=
ritto:
>=20
> Pietro Pugni <pietro.pugni@gmail.com> writes:
>> I=E2=80=99ve jsut discovered the issue.. I set "logging_collector=3Doff=
=E2=80=9D in the previous email but didn=E2=80=99t comment the other log* p=
arameters, so Postgres was logging every single INSERT! This was caused the=
 disk to fill up.
>=20
> Ah.
>=20
>> The strange issue is that the log file didn=E2=80=99t exists when the di=
sk filled up. I personally looked for it but it wasn=E2=80=99t where it sho=
uld have been ( /var/log/postgesql/ ), so I can=E2=80=99t exactly confirm t=
hat the issue was the log file getting bigger and bigger.
>=20
> Seems like the log file must have gotten unlinked while still active,
> or at least, *something* had an open reference to it.  It's hard to
> speculate about the cause for that without more info about how you've got
> the logging set up.  (Are you using the log collector?  Are you rotating
> logs?)  But I seriously doubt it represents a Postgres bug.  Unlike the
> situation with data files, it's very hard to see how PG could be holding
> onto a reference to an unused log file.  It only ever writes to one log
> file at a time.
>=20
> 			regards, tom lane


--=20
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance