You are dead right of course. 

Disasters don't come on schedule, neatly tied up in a bow. 

A good thing might be a brainstorming session on "what are our implicit 
disaster-related assumptions?" and then questioning those assumptions. 

Charles
Composed on a mobile: please excuse my brevity 

"Staller, Allan" <[email protected]> wrote:

>Although very few shops actually do this, IMO the procedure should be:
>
>Management walks in the room and says " You, you, and you are dead as of 
>"time/date". The rest of you go recover as of that time/date."
>The "dead people" cannot be consulted with during the DR exercise. "You, you, 
>and you" should be different during each iteration of the test.
>After the fact, procedures/documentation are analyzed and updated based on the 
>results.
>
>In too many cases, I have seen "staged" recoveries, whereby the data is all 
>snapshot'ed at the end of a "cycle" and neatly tied up in a package.
>The same "crew" is used repeatedly and becomes very familiar with all of the 
>procedures, and actually tests nothing new.
>All that is proven in this case is your applications can run on other 
>compatible hardware.
>
>I have deliberately ignored the data questions, as your configuration is 
>nothing like mine.
>
>Just my $0.02 USD worth.
>
>HTH,
>
><snip>
>I am looking to see how other shops are currently doing Backups for DR and OR. 
> I think this will be valuable for the archives.
></snip>
>
>----------------------------------------------------------------------
>For IBM-MAIN subscribe / signoff / archive access instructions,
>send email to [email protected] with the message: INFO IBM-MAIN


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to