Re: [HACKERS] [PATCHES] wal_checksum = on (default) | off - Mailing list pgsql-patches

From Martijn van Oosterhout
Subject Re: [HACKERS] [PATCHES] wal_checksum = on (default) | off
Date
Msg-id 20070112100928.GA6589@svana.org
Whole thread Raw
In response to Re: [HACKERS] [PATCHES] wal_checksum = on (default) | off  ("Simon Riggs" <simon@2ndquadrant.com>)
List pgsql-patches
On Thu, Jan 11, 2007 at 11:10:38PM +0000, Simon Riggs wrote:
> On Thu, 2007-01-11 at 17:06 +0000, Gregory Stark wrote:
> > Having a CRC in WAL but not in the heap seems kind of pointless.
>
> Yes...
>
> > If your
> > hardware is unreliable the corruption could anywhere.
>
> Agreed.

I thought the point was that the WAL protects against unexpected power
failure, that sort of thing. In that situation, the memory is the first
to be corrupted, and an active DMA transfer will thus be corrupted
also. We don't need to worry about the data, because the WAL is known
to be accurate.

The WAL does not protect against random data corruption, in normal
operation it is never read. If we want to detect random corruption,
we'd need checksum everywhere, yes. But that's not the goal here.

Have a nice day,
--
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> From each according to his ability. To each according to his ability to litigate.

Attachment

pgsql-patches by date:

Previous
From: "Guillaume Smet"
Date:
Subject: Re: [PATCHES] Patch to log usage of temporary files
Next
From: Dave Page
Date:
Subject: Allow dumping of roles or tablespaces