Jim,

I just got your second message while I was working on this so here
is more gory details.

Kris

Our build environment is complex and does some of its own determining
of endian.  To convince myself that I am not mad, I found a very
simple client/server example at:

http://publib.boulder.ibm.com/infocenter/iseries/v5r3/index.jsp?topic=/rzab6/rzab6xconoserver.htm

After a quick cut and past and a very simple makefile:

all: client server

client: client.c
        cc client.c -o client


server: server.c
        cc server.c -o server

I was able to build with a few changes

#define NETDB_MAX_HOST_NAME_LENGTH 256

void main -> int main

And run this sample on i386/Linux.  When I change the
lines for sin_family for htons the code fails
with the same error I mentioned earlier (EAFNOSUPPORT).

Running a fem more test I discovered that no matter
what value I put in for sin_family on the server (bind) it
worked.  The error was on the client connect call,
where it definitely was no happy when htons(AF_INET)
was used.

--
Mr. Kristen J. Webb
Teradactyl LLC.

PHONE: 1-505-242-1091
EMAIL: [EMAIL PROTECTED]
VISIT: http://www.teradactyl.com

        Home of the

 True incremental Backup System

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to