Re: Serious Crash last Friday

From: Andrew Sullivan <andrew(at)libertyrms(dot)info>
To: pg <pgsql-general(at)postgresql(dot)org>
Subject: Re: Serious Crash last Friday
Date: 2002-07-10 20:35:47
Message-ID: 20020710163547.W24611@mail.libertyrms.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Wed, Jul 10, 2002 at 02:16:53PM -0600, scott.marlowe wrote:
> This isn't entirely true. According to bad blocks' man page:
>
> -n Use non-destructive read-write mode. By default

Ah, yes, I forgot about that. Takes days ;-)

> Also, another REALLY good test for bad memory is to build postgresql from
> source a couple dozen times, especially with a -j switch set to about 6 or
> so.

I've had extremely good luck with memtestx86, though. Usually, by
the 3rd or 4th test it's spotted something, if something is wrong.

This raises another issue, by the way. Even big, expensive hardware
can have faulty memory. But the big, expensive hardware frequently
has ECC RAM, which saves your hide. I _know_ it's expensive, but
it's worth every penny. If you're building a database server for
production use, and you haven't specified ECC memory, go back and
think again. It might save you hours of work some day.

A

--
----
Andrew Sullivan 87 Mowat Avenue
Liberty RMS Toronto, Ontario Canada
<andrew(at)libertyrms(dot)info> M6K 3E3
+1 416 646 3304 x110

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Arjen van der Meijden 2002-07-10 20:38:45 Re: Linux max on shared buffers?
Previous Message scott.marlowe 2002-07-10 20:16:53 Re: Serious Crash last Friday