xfs
[Top] [All Lists]

Re: noatime

To: linux-xfs@xxxxxxxxxxx
Subject: Re: noatime
From: Robert Brockway <robert@xxxxxxxxxxxxxxxxx>
Date: Mon, 2 Feb 2004 16:30:11 -0500 (EST)
In-reply-to: <1075751733.13701.11.camel@david.internal.NorcrossGroup.com>
References: <115.2e2700bf.2d4bf471@aol.com> <20040130175756.A23646@infradead.org> <Pine.LNX.4.58.0401310332290.1212@zen.canint.timetraveller.org> <1075751733.13701.11.camel@david.internal.NorcrossGroup.com>
Sender: linux-xfs-bounce@xxxxxxxxxxx
On Mon, 2 Feb 2004, Greg Freemyer wrote:

> "Rob created a flat file export of the Customer Database on Dec 15, 03.
> He accessed this flat file at 2pm, Feb 2, 04.  This is 2 hours after he
> was notified that he was being fired, so it is possible that he was
> making an improper copy to use outside the company."

I do know of one such use of atime by a friend of mine while tracking an
errant user.  The problem (as I see it) is that a knowledgable user who
owns the file or has root access (all too common on many boxes) will use
/bin/touch to hide their tracks so I've long believed the usefulness of
atime in this way was limited.  Maybe I'm over-estimating errant users ;)

> I guess what I'm saying is, if you are maintaining valuable info on a
> computer and the possibility of having to litigate about its use exists,
> then having access times available to a computer forensic examiner is a
> good idea.

Fair point.

This reminds me of discussions relating to system optimization (eg, use of
hdparm).  If I really care about a system being rock solid and am not so
worried about performance, I'm going to be much more conservative with
hdparm optimizations.  I suppose this could be said to be similar - being
more conservative with a performance optimization (noatime) because I'd
like the extra auditability.

Cheers,
        Rob

-- 
Robert Brockway B.Sc. email: robert@xxxxxxxxxxxxxxxxx, zzbrock@xxxxxxxxxxxxx
Linux counter project ID #16440 (http://counter.li.org)
"The earth is but one country and mankind its citizens" -Baha'u'llah


<Prev in Thread] Current Thread [Next in Thread>