From: "Anibal David Acosta" <aa@devshock.com>
To: "'Sergey Konoplev'" <gray.ru@gmail.com>,
	"'Richard Huxton'" <dev@archonet.com>
Cc: "'Stephen Frost'" <sfrost@snowman.net>, <pgsql-performance@postgresql.org>
References: <005501cca0ad$b9478440$2bd68cc0$@devshock.com>
	<20111111201822.GN24234@tamriel.snowman.net>
	<CAL_0b1t02q77ft2JWdjMv8q4vaGm=x__K0rzVkpNSvSy2-SjzA@mail.gmail.com>
	<4EC0D827.8090601@archonet.com>
	<CAL_0b1tuxi9YVhGKsNUSTZCUBXgOmppan7w+KkRZARbQ=u4P5w@mail.gmail.com>
	<4EC0EAD6.9010302@archonet.com>
	<CAL_0b1sSyTgMUxcRi+2ibTHHV4-qH3fvacsQUCbbeBsu0jYGAQ@mail.gmail.com>
In-Reply-To: 
 <CAL_0b1sSyTgMUxcRi+2ibTHHV4-qH3fvacsQUCbbeBsu0jYGAQ@mail.gmail.com>
Subject: Re: unlogged tables
Date: Mon, 14 Nov 2011 14:05:51 -0300
Message-ID: <00ea01cca2ef$abcf3a30$036dae90$@devshock.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Thread-Index: 
 AQG0h8c2HxwgVHVwA7Oudni7xLo/1gKNzE9QAZGzEw4Bc7WZpAHcxNALAamzBiMCHqGblZWDDJWA
Content-Language: es-py

Maybe an option like "Recover from file " will be useful
So, for example, daily some process do a COPY of entire table to a file=20

In case of crash postgres recover content from the file.

:)


-----Mensaje original-----
De: Sergey Konoplev [mailto:gray.ru@gmail.com]=20
Enviado el: lunes, 14 de noviembre de 2011 07:39 a.m.
Para: Richard Huxton
CC: Stephen Frost; Anibal David Acosta; pgsql-performance@postgresql.org
Asunto: Re: [PERFORM] unlogged tables

On 14 November 2011 14:17, Richard Huxton <dev@archonet.com> wrote:
> On 14/11/11 10:08, Sergey Konoplev wrote:
>>
>> On 14 November 2011 12:58, Richard Huxton<dev@archonet.com> =A0wrote:
> Let's say you were doing something like "UPDATE unlogged_table SET =
x=3D1=20
> WHERE y=3D2". If a crash occurs during this command, there's no=20
> guarantee that the affected disk pages were all updated. Worse, a=20
> single page might be partially updated or even have rubbish in it=20
> (depending on the nature of the crash).
>
> Without the WAL there's no way to check whether the table is good or=20
> not, or even to know what the last updates were. So - the only safe=20
> thing to do is truncate the unlogged tables.
>
> In the event of a normal shutdown, we can flush all the writes to disk =

> so we know all the data has been written, so there is no need to =
truncate.

Thank you for the explanation. Now I understand it.

>
> --
> =A0Richard Huxton
> =A0Archonet Ltd
>


--
Sergey Konoplev

Blog: http://gray-hemp.blogspot.com
LinkedIn: http://ru.linkedin.com/in/grayhemp
JID/GTalk: gray.ru@gmail.com Skype: gray-hemp