This leads me to recommend what every company should have:

A mirror testing environment of your production machines,
which represent as closely as possible your running environment. When you patch and test here, work out your backout strategy if/when things go pointy side up, ensure your backups of important files are current.

Then and only then are you ready to patch your production systems. Even if everything goes horribly wrong you should
have a means of recovering quickly and causing minimal downtime.

If your company does not have such an environment and won't implement it because of cost, find another company, as they
obviously do not take their production side seriously.

BTW: I work in the Unix realm, but the same approach applies
to any O/S or hardware setup.

Thanks.


[ reply ]

Link to this comment: http://www.securityfocus.com/comments/columns/74/12243#12243