On 2/23/10 12:07 AM, Oliver Falk wrote:
Hi Jay!
On 02/23/2010 04:09 AM, Jay Hankinson wrote:
[ ... ]
> * Ingres requires that the server executable is setuid (non-root)
>
Actually I don't know any>rules< that speak against this. However, if
it's really needed, then we cannot really do anything against it. And if
there's a rule that speaks against it, there's usually also a 'this is
the way to fix it' guide :-)
rpmlint flagged it, which is really why I mentioned it. Increasingly
looking like it's not going to be an issue though. :-)
> * Current build process uses $ORIGIN for relative RPATH linking.
>
RPath is bad - And really is something that needs to be fixed; No
matter which OS and/or distribution we are talking about!
Regular RPath, yes I agree entirely. $ORIGIN allows relative linking wrt
to the binary location and is (as I understand it) as secure as anything
else. As I said, I can work around it though.
> * Executable stack code
>
And? :-) I'm not too familiar with NX. However, nobody will force you to
compile it with no-exec-stack :-)
> If the best approach is "write the spec, follow the guidelines, create a
> Bugzilla issue and we'll go from there", then that is what I will do. If
> I can gain experience by assisting in with the packaging of other
> software for fedora then I'm happy to do this too.
>
I (personally) really appreciate your wish to include Ingres. I'm not
working with Ingres, but I'm always happy to get in touch with new
databases; Since it's (also) part of my job :-)
And yes, the best way is to craft a specfile and follow the guidelines.
If you need assistance in packaging, please don't hesitate to contact me...
Thank you, that's really encouraging to hear. I've been trying to get
this project underway for over a year. Very glad to be moving forward
with it at last. I will no doubt be sending questions this way on a
regular basis.
Best,
Oliver