On 23 Jun 2005, you wrote in lfs.book:

> [EMAIL PROTECTED] wrote:
>> Version increment (2.6.12.1) - two security fixes in this one.
> 
> Bad. Does this mean that all 2.6.11.x kernels are vulnerable? Given the 


Not necessarily. If the issues (which I haven't looked at yet) have been 
introduced into 2.6.12 through new code\features, 2.6.11.x won't be 
affected. That said, there will not be anymore 2.6.11.x releases, as per 
the guidelines for the 2.6.x.y release series, so any newly disocvered 
security problems that do affect 2.6.11 will *not* be fixed in a new 
2.6.11.x release.

> frequency with which kernel security issues are being found, does it 
> make any sense to refer to any fixed kernel version in the stable book?

We have to pick a point in time for the stable releases - changes in kernel 
internals may result in known working components breaking. THis is true for 
all the software LFS tracks, although the kernel is probably the most 
rapidly changing target.

What we need is an errata system for stable releases, so we can add such 
information after a release is made. One of the issues that will need to be 
discussed if an errata mechanism is followed is how far back errata will be 
tracked (i.e. do we provide errata for much older releases, i.e. LFS 5.0) - 
I would suggest we only provide errata for the current stable release, and 
have a section in the LFS book (perhaps at the end) describing the changing 
software world, and perhaps recommending some resources the user can access 
to track security and functionality issues.

- --
Steve Crosby
-- 
http://linuxfromscratch.org/mailman/listinfo/lfs-book
FAQ: http://www.linuxfromscratch.org/lfs/faq.html
Unsubscribe: See the above information page

Reply via email to