🚀 go-pugleaf

In article <2hsgm1$f4p@nic.hookup.net>,
Murray S. Kucherawy <mskucher@nic.hookup.net> wrote:
>In article <2hs24t$hag@grasp.insa-lyon.fr>,
>Christophe Wolfhugel <Christophe.Wolfhugel@grasp.insa-lyon.fr> wrote:
>>mskucher@nic.hookup.net (Murray S. Kucherawy):
>>>You're not alone - I've received mail from at least one other person
>>>who's described similar symptoms.  I haven't been able to explain it
>>>yet; somehow it works just fine for me.
>>...
>>Up to know, using the Bind 4.9 (4.8.3 should also be okay) resolver
>>library has proven to provide a stable solution.  Now setting up
>>the 4.9 resolver on system V machines is also a random work.
>
>I'll be putting up my compiled named, named-xfer, and libresolv.a
>for SCO UNIX for FTP sometime soon then, so people can get their
>sendmails working.  Watch for an announcement.

Since my original post, I was able to remove the duplicate modules
from SCO's libsocket.a.  This allowed me to link with my port of 
BIND 4.9.  The only major porting I had to do for BIND 4.9 was to 
change calls of writev() to what I thought were the appropriate
write() calls.  

Sendmail 8.6.5 was not doing MX lookups successfully with my newly
linked configuration with BIND 4.9.  I haven't tried the nslookup
linked with 4.9 yet, under SCO, but I'll let you know if it works.
I was getting errors when I tried to mail to MX records.  No cores
though, which is better than what I got with SCO's resolver.

Was the version of GCC you compiled for SCO easy to do?  If so, then
I would happily try that if it would result in BSD sendmail 8.6.5
working with DNS under ODT 3.0.

-- 
Doug Anderson                I feel like burning something ... he hehe hehe
                             Uuh huh huh ... That would be cool!
dlander@afterlife.ncsc.mil                         -Beavis & Butt-head