Postgres Pro
Downloads
Home   >   mailing lists

Re: Data Corruption in case of abrupt failure - Mailing list pgsql-general

From Tom Lane
Subject Re: Data Corruption in case of abrupt failure
Date
Msg-id 20893.1079416598@sss.pgh.pa.us
Whole thread Raw
In response to Re: Data Corruption in case of abrupt failure  (Stephen Robert Norris <srn@commsecure.com.au>)
List pgsql-general
Tree view 
Stephen Robert Norris <srn@commsecure.com.au> writes:
> On Tue, 2004-03-16 at 16:44, Tom Lane wrote:
>> My money is on IDE drives with write cache enabled.

> I was wondering if he'd turned off fsync(), as seems common for some
> reason...

That would be bad too.  But the nasty thing about IDE write caches is
that even when you think you've fsync'd, you haven't :-(

            regards, tom lane

pgsql-general by date:

Previous
From: Stephen Robert Norris
Date:
Subject: Re: Data Corruption in case of abrupt failure
Next
From: "Keith C. Perry"
Date:
Subject: Re: Data Corruption in case of abrupt failure