Pete Finnigan's Oracle Security Forum


    
      
        Pete Finnigan's Oracle Security Forum
        (http://www.petefinnigan.com/forum/yabb/YaBB.cgi)
      

        Oracle Security >> Oracle Security >> New book: HOWTO Secure and Audit Oracle 10g and 11
        
(Message started by: Pete Finnigan on Feb 13^th, 2009, 2:12pm)

Title: New book: HOWTO Secure and Audit Oracle 10g and 11
Post by Pete Finnigan on Feb 13^th, 2009, 2:12pm

Today Amazon brought to my attention that there is a new Oracle security book:

"HOWTO Secure and Audit Oracle 10g and 11g" by Ron Ben-Natan (http://www.amazon.com/HOWTO-Secure-Audit-Oracle-10g/dp/1420084127/ref=sr_1_1?ie=UTF8&s=books&qid=1234530366&sr=8-1).

Thought you might be interested.

Title: Re: New book: HOWTO Secure and Audit Oracle 10g an
Post by Pete Finnigan on Feb 16^th, 2009, 9:18am

Thanks for the link Marcel-Jan, take a close look at the jpeg of the books front cover..:-)

Title: Re: New book: HOWTO Secure and Audit Oracle 10g an
Post by Pete Finnigan on Feb 17^th, 2009, 8:40am

I hadn't notice it. :)

So you probably read some of it?

Title: Re: New book: HOWTO Secure and Audit Oracle 10g an
Post by Pete Finnigan on Feb 20^th, 2009, 12:00pm

Hi Marcel-Jan,

Yes I have, I have seen a number of the chapters for review before writing the forward. Its an excellent book.

cheers

Pete

Title: Re: New book: HOWTO Secure and Audit Oracle 10g an
Post by Pete Finnigan on Jul 27^th, 2009, 6:52pm

That's a good book of the concentrated place of relevant knowledge. Simply reading the Oracle manuals is sometime misleading.
However, this book and others are missing a few small topics that are asked every time an issue of auditing pops up:
- performance impact differences between DB and OS audit trails (assuming the rest of the audit set-up is the same);
- methods of seamless audit extract without loosing the audit data (sys.aud$ is a particular issue, but OS-level audit files are vulnerable too - Oracle does not mark them "complete", and it's a set of system related tricks to do to decide if the file is done. All those tricks rely on UNIX admin decision to have them ON. Instead, Oracle would do a favour to simply rename the "ready" files, etc.