[comp.unix.i386] AIX vs modems -- epilogue

kevin@msa3b.UUCP (Kevin P. Kleinfelter) (07/21/89)

One final note in my AIX modem battles:

I got a callback from IBM yesterday.  They said that I should just set
"prototol" to "DTR" instead of "DC".

This causes the proper minor device number to be set.  When I pointed
out that this parameter (in devices) is completely undocumented,
and that the printed example devices screens do NOT include this parameter,
they said "oh."  I asked them to find out what this parameter means,
AND what the extra bit in the minor device number which it varies
means.  They said they'd have to call back.

When they did, the only explanation they could offer was
	You set it to DC for direct-attached terminals.
	You set it to DTR for modem-attached terminals.

I just love undocumented features.  Also, there seems to be a problem
when you call IBM in talking to someone who understands the questions
(and also you can end up talking to someone who doesn't understand the
ANSWERS!)  The doc for AIX PS/2 has several errors (such as  an incorrect
index, omitted parameters, etc).

Also, a "cute" aside:
   When I installed the system, I ran "devices" and set all the documented
   parameters.  When I ran into "protocol", I did not know what "dtr" and
   "dc" meant in this context.  (I assumed "dc" = "data carrier  detect".)
   However, since it was only one parameter, I figgured that if it did not
   work one way, I could try the other.  Unfortunately I tried the wrong
   one first.  If you use dc first, it creates minor device 96.  If you
   interrupt an open on a 96, which is waiting (forever) on CD, the
   port is hosed until you either reboot, or run an open on the 96  which
   succeeds because CD was high before the open.  I just tried one value
   and then the other, and neither worked.

Now I can proceed to real work:  I'll post my opinions of AIX PS/2
after I learn more. (I'll try to limit further postings to 
comp.unix.i386.)

Thanks to all who offered helpful suggestions.
-- 
Kevin Kleinfelter @ Management Science America, Inc (404) 239-2347
gatech!nanovx!msa3b!kevin