[Veritas-bu] Backup Audit script...
tim.berger at gmail.com
Wed Mar 30 12:07:37 CST 2005
On Wed, 30 Mar 2005 12:14:25 -0500, Paul Keating
<pkeating at bank-banque-canada.ca> wrote:
> > -----Original Message-----
> > From: Ungaro, Matt [mailto:mjungaro at capitolindemnity.com]
> > Sent: March 30, 2005 11:10 AM
> > To: Paul Keating; veritas-bu at mailman.eng.auburn.edu
> > Subject: RE: [Veritas-bu] Backup Audit script...
> > Paul,
> > Even that's not enough. Sadly. The best test of completeness of
> > backups is to perform recovery on randomized sets of Data.
> That's already done.
> But random isn't sufficient. :o|
> > on a quarterly basis attempt to do a full system
> > restore from a randomly picked set of backups. It's a total
> > pain, but it's
> > really the only way to find out if your backups are usable.
How come they're not demanding md5sum discrepancy reports? ;-)
What auditors demand is highly dependent on who comes in and how
competent they are. They ought to be following published standards
like COBIT that define required sample frequencies and such. If no
standard is followed, the sky is the limit and auditing demands become
If you can demonstrate that your backup system is functional, that you
perform restores on a frequent basis, and keep logs of all restores
and their status that ought to be good enough.
You may also need to keep the backup logs around, and keep a log
showing that you review them on a regular basis.
It's unfortunate that we need to build our own reporting system or buy
something to get useful reporting.
> I agree that is necessary....we do that on a formal basis, and more
> often on an informal basis, but this "issue" we had got blown a little
> out of proportion, so we're being asked to provide MORE assurance.
> Veritas-bu maillist - Veritas-bu at mailman.eng.auburn.edu
More information about the Veritas-bu