On Wed, 2007-02-07 at 08:55 -0800, Dan Nicholson wrote:
> On 2/7/07, Andrew Beverley <[EMAIL PROTECTED]> wrote:
> > On Wed, 2007-02-07 at 08:32 -0800, Dan Nicholson wrote:
> > > On 2/7/07, Andrew Beverley <[EMAIL PROTECTED]> wrote:
> > > >
> > > > Sorry, I don't mean modules as such. I mean extensions. Each
> > > > target/match for iptables has its own extension.
> > >
> > > Yeah, you're right. Could you show the exact errors? I'm guessing that
> > > our kernel headers are just too old in LFS-6.2.
> >
> > I don't think the kernel headers are too old, but there's not an error
> > as such for the following reasons:
> >
> > The extensions that require the headers all have a hidden test script
> > located in the extensions directory.
> >
> > For example the connbytes extension has a test script
> > iptables-1.3.x/extensions/.connbytes-test
> 
> Thanks for the explanation. Well, I don't think anything's changed,
> then. I don't have a libipt_connbytes.so for iptables-1.3.5, either.
> For BLFS, I think you just get the extensions that build against the
> sanitized kernel headers. This has been working fine for people for a
> long time.
> 
> I don't envision us changing that since it would require us to always
> build against the raw headers, which can change constantly with kernel
> version. Not that it's really important, but Fedora builds against the
> sanitized headers, too. I suppose we could add a note that says you
> can build more extensions against the raw kernel sources at your own
> risk.

That would be appreciated; it will at least save other people pulling
their hair out!

Regards,

Andy Beverley


-- 
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to