Re: Incremental Backups in postgres

From: Greg Stark <gsstark(at)mit(dot)edu>
To: Alban Hertroys <dalroi(at)solfertje(dot)student(dot)utwente(dot)nl>
Cc: akp geek <akpgeek(at)gmail(dot)com>, pgsql-general(at)postgresql(dot)org
Subject: Re: Incremental Backups in postgres
Date: 2009-11-10 14:52:15
Message-ID: 407d949e0911100652s627286dbt8cd5d98030c8b873@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tue, Nov 10, 2009 at 11:03 AM, Alban Hertroys
<dalroi(at)solfertje(dot)student(dot)utwente(dot)nl> wrote:
> IMHO The simplest solution is to just write a dump to the same file every
> now and then and have the backup software take care of storing only the
> differences. It does have a few drawbacks; it means you'll have a file about
> as large as your database on your filesystem just for making backups and
> there is a risk that your backup software kicks in before the dump has
> finished writing.
>
> As others mentioned, you can also go with a PITR solution, which is probably
> prettier but is a bit harder to set up.

It's always worth having the dump, even if you also implement PITR.
The dump allows you to restore just specific tables or to restore onto
a different type of system. The PITR backup is a physical
byte-for-byte copy which only works if you restore the whole database
and only on the same type of system.

--
greg

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Richard Broersma 2009-11-10 14:56:17 Re: How can I calculate differences between values
Previous Message Vyacheslav Kalinin 2009-11-10 14:49:27 Re: How can I calculate differences between values