Message-ID: <4824D928.2050503@lelarge.info>
Date: Sat, 10 May 2008 01:07:20 +0200
From: Guillaume Lelarge <guillaume@lelarge.info>
User-Agent: Thunderbird 2.0.0.14 (X11/20080505)
MIME-Version: 1.0
To: "MARIOTTI, Alessandro \(Alessandro\)" <mariotti@alcatel-lucent.com>
CC: pgsql-admin@postgresql.org
Subject: Re: debug issue on postgresql
References: <85618CB6B1C0F14D888609DB41641AB40127F1EA@DEEXC1U03.de.lucent.com>
	<4824841F.4020104@lelarge.info>
	<85618CB6B1C0F14D888609DB41641AB40127F290@DEEXC1U03.de.lucent.com>
In-Reply-To: 
 <85618CB6B1C0F14D888609DB41641AB40127F290@DEEXC1U03.de.lucent.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

MARIOTTI, Alessandro (Alessandro) a écrit :
> Well I could shortly describe what's happen on these machines ;
> 
> On May 7 2008 at 23:25 an operator performed :
> 
> _a configuration change ;
> 
> _add a new user on a table named MVNO;
> 

None of these could have caused a corruption.

> Customer told us that for  “new user” operator add comma “,” symbol.
> 
> This is something new for their provisioning procedure (it’s seems they 
> never test it )thus as soon as they propagate this updated DB over 
> machines of live net these started to discard all the requests and no 
> more user was able to work or work properly .
> 

> Now they asking us what’s happen so this is the reason why I’m writing 
> here .
> 

You'll need more info if you want to find out what happened.

> Unfortunately I tried to verify the backup db but the one interesting 
> for us is empty :
> 
> ls -l  200805*
> 
> -rw-r--r--    1 root     root       842840 mag  1 04:03 
> 20080501-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842840 mag  2 04:03 
> 20080502-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842840 mag  3 04:03 
> 20080503-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842840 mag  4 04:03 
> 20080504-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842540 mag  5 04:03 
> 20080505-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842613 mag  6 04:03 
> 20080506-0403-cfg_xxxxxxx.sql
> 
> -rw-r--r--    1 root     root       842613 mag  7 04:03 
> 20080507-0403-cfg_xxxxxxx.sql
> 
> *-rw-r--r--    1 root     root            0 mag  8 04:03 
> 20080508-0403-cfg_xxxxxxx.sql*
> 
> -rw-r--r--    1 root     root       843960 mag  9 04:03 
> 20080509-0403-cfg_xxxxxxx.sql
> 

You don't a SQL backup, these won't help you. You need a backup of the 
database after the corruption.

> I‘ll try to find it but but even if I can’t provide this back-up for 
> company policies reason , could I ask you some feed-back to get 
> suggestion to look for on this back-up file?
> 

Nope. You didn't give us enough information to at least get started.


-- 
Guillaume.
  http://www.postgresqlfr.org
  http://dalibo.com