OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
Subject: Re: Redhat Linux, Mailman and Sendmail
From: Matthew Hawkins (matthewtopic.com.au)
Date: Tue Apr 04 2000 - 14:21:30 CDT

On 2000-04-04 20:29:26 +0200, Brad Knowles wrote:
> Of course, Linux casually ignores this advice, and if a mail
> administrator does not make changes to their configuration so as to
> either mount the entire filesystem that /var/spool/mqueue is on with
> synchronous meta-data writes, or use chattr +S on /var/spool/mqueue
> to ensure that all writes to this directory are done with synchronous
> meta-data updates, then they are in violation of RFC 1123, section
> 5.3.3.

Note that the section explicitly says "must not lose" - not "must not
make it theoretically, though remotely, possible to lose".

If your ext2 filesystem with async writes enabled does not lose data,
you are not in violation of RFC 1123 section 5.3.3 no matter what any
BSD bigot says.

Also note that the whole argument of sync versus async writes is
pointless when most drives come with 2Mb of cache anyway - should the
power go out it is irrelevent if the data was written to this cache
synchronously or asynchronously: you're now violating RFC 1123 ;-)

> In fact, using chattr +S really isn't enough, since writes to
> other parts of /var or /var/spool could cause the entire filesystem
> to be toasted, and thus taking /var/spool/mqueue along with it.

*chuckles politely* Maybe FreeBSD will corrupt its own filesystem in
this manner, Linux ext2 certainly does not.

dd, rm, fdisk, and a whole plethora of other utilities will toast
filesystems too. The _only_ thing that can save you somewhat is regular
backups of your important data. ANY filesystem on any sort of magnetic
media will eventually corrupt.

Other things than disk media can also cause corruption in violation of
RFC 1123 - faulty network interface devices or cabling for example. 

-- 
Matt