From: Greg Smith <greg@2ndquadrant.com>
To: Tony Nelson <tnelson@starpoint.com>
Cc: pgsql-admin@postgresql.org
Subject: Re: Corrupted disk
Date: Sun, 20 Feb 2011 11:52:57 -0500
Message-ID: <4D6146E9.2030302@2ndquadrant.com> (raw)
In-Reply-To: <2B07A451-9F67-4438-8802-B10F46FDAF2E@starpoint.com>
References: <2B07A451-9F67-4438-8802-B10F46FDAF2E@starpoint.com>
Tony Nelson wrote:
> I am ok, because I have a dump from last night, and wal since then.
>
Dumps made with pg_dump are easy to restore from. But they're fixed in
time: there is no applying WALs to them in order to update them.
Anything that's happened to the server since then you can't add.
If you made a filesystem copy of the server using
pg_start_backup/pg_stop_backup, and you save all of the WAL files after
the backup began, that's also useful. You can store from that backup
and apply all the WAL that's happened since then.
> Can I drop/create/restore from this dump? Or should I restore from last nights full and apply the WAL?
>
If last night's backup with a filesystem one done with
pg_start_backup/pg_stop_backup, and you have WAL since then, I would
favor that set as likely to work fine. But it sounds like what you have
might instead be a pg_dump backup and some WAL files; you can't apply
the WAL to such a dump.
Whatever you do, you want to make a full filesystem copy of the server's
data directory--with the server shutdown--before you do anything else.
It's possible to recover from page errors and extract the available data
using the right data recovery techniques, especially if there's a
pg_dump available too; we offer some services in this area. But if any
serious changes are made to the database before we get to it, odds of
successful recovery can drop fast.
--
Greg Smith 2ndQuadrant US greg@2ndQuadrant.com Baltimore, MD
PostgreSQL Training, Services, and 24x7 Support www.2ndQuadrant.us
"PostgreSQL 9.0 High Performance": http://www.2ndQuadrant.com/books
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-admin@postgresql.org
Cc: greg@2ndquadrant.com, tnelson@starpoint.com
Subject: Re: Corrupted disk
In-Reply-To: <4D6146E9.2030302@2ndquadrant.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox