Problems building argus on non-i386 systems.

[Carter Bullard]

The private header files have solved a lot of problems
in the past, much much much more than the problems that
are just now showing themselves.  Argus has been ported
to 20+ Unixes over the last 5 years, but I haven't had
the time to make all the right decisions on some of these
ports, so I can understand it looking abit patched together.

But this is the type of work that makes it better, and
we definitely can make the changes to make it work.

Carter

Carter Bullard
QoSient, LLC
300 E. 56th Street, Suite 18K
New York, New York  10022

carter at qosient.com
Phone +1 212 588-9133
Fax   +1 212 588-9134
http://qosient.com 

-----Original Message-----
From: owner-argus-info at lists.andrew.cmu.edu
[mailto:owner-argus-info at lists.andrew.cmu.edu] On Behalf Of Yotam Rubin
Sent: Monday, June 04, 2001 10:53 PM
To: argus-info at lists.andrew.cmu.edu
Subject: Re: Problems building argus on non-i386 systems.


Greetings,

	As I see it, using private instances of header files can lead to

porting problems. If a user does not have the relevant headers then he
can just fetch them. The rationale for the current configuration is to
lower dependencies, but argus already has many dependencies anyway, so
the 
disadvantages out-weigh the merits. This is not the first time that
problems have occurred as a result of dated headers, I recall that an
old pcap.h 
cause some trouble.

	Regards, Yotam Rubin

On Mon, Jun 04, 2001 at 10:38:55PM -0400, Carter Bullard wrote:
> Hey Yotam,
>    The headers are there for when there are no system
> headers at all.  There maybe a problem with ./configure where
> the include flag " -I ../include/linux-include" is used
> when it is not needed.  The various include files have been
> used for many years, so they may be dated, or from an old
> BSD system, ...
> 
>    Thanks for looking into this.  If there is anything I
> can do to help, just holler!
> 
> Carter
> 
> Carter Bullard
> QoSient, LLC
> 300 E. 56th Street, Suite 18K
> New York, New York  10022
> 
> carter at qosient.com
> Phone +1 212 588-9133
> Fax   +1 212 588-9134
> http://qosient.com
> 
> -----Original Message-----
> From: owner-argus-info at lists.andrew.cmu.edu
> [mailto:owner-argus-info at lists.andrew.cmu.edu] On Behalf Of Yotam 
> Rubin
> Sent: Monday, June 04, 2001 10:22 PM
> To: argus-info at lists.andrew.cmu.edu
> Subject: Re: Problems building argus on non-i386 systems.
> 
> 
> on Sun, Jun 03, 2001 at 09:53:14AM -0400, Carter Bullard wrote:
> > Hey Yotam,
> >    Yes argus compiles fine on Solaris.  Somehow its finding a 
> > <netinet/ip.h> that has a non-BSD type of "struct ip" defined. These
> 
> > are Debian machines?  What does the standard Debian
> > /usr/include/netinet/ip.h look like?
> 
> 
> Hmmm, apparently argus has its own versions of certain header files, 
> including ip.h, the following bit is the problematic one: " #if
> defined(vax) || defined(i386)
>         u_char  ip_hl:4,                /* header length */
>                 ip_v:4;                 /* version */
> #endif
> #if defined(mc68000) || defined(sparc)
>         u_char  ip_v:4,                 /* version */
>                 ip_hl:4;                /* header length */
> #endif"
> 
> It should be:
> struct ip
>   {
> #if __BYTE_ORDER == __LITTLE_ENDIAN
>     unsigned int ip_hl:4;               /* header length */
>     unsigned int ip_v:4;                /* version */
> #endif
> #if __BYTE_ORDER == __BIG_ENDIAN
>     unsigned int ip_v:4;                /* version */
>     unsigned int ip_hl:4;               /* header length */
> #endif
> 
> 
> Is there a particular reason for duplicating these header files?
> I believe we should use the official header files whenever possible.
> Anyway, I'll patch the existing ip.h and you decide what to do for the

> release.
> 
> 	Regards, Yotam Rubin
>