[comp.unix.ultrix] Modems Under Ultrix 2.0

bwong@mickey.UUCP (01/12/88)

I have some multitech modems that I am trying to get to dial out under
Ultrix 2.0.  I have successfully dialed out but when the call is completed
and It hangs up, the modem goes into a race condition for awhile and then
the machine hangs (or goes slow enough to make be believe this).  Does
anyone know what I am doing wrong?

/b

sdowdy@charon.unm.edu (Stephen Dowdy) (01/22/88)

In article <18456@felix.UUCP> bwong@mickey.UUCP writes:
>I have some multitech modems that I am trying to get to dial out under
>Ultrix 2.0.  I have successfully dialed out but when the call is completed
>and It hangs up, the modem goes into a race condition for awhile and then
>the machine hangs (or goes slow enough to make be believe this).  Does
>anyone know what I am doing wrong?

I recently setup some multitechs on a vms microvax, and encountered this
problem.  (well, the machine didn't exactly hang, but slowed down due to
the large number of process creations started by the false login detections.)
After any successful incoming or outgoing call was terminated
the terminal port would stay in a state indicating a <login> session was
being established there.

What was happening was that when the host would disconnect from the modem
at the end of a session, the modem would attempt to send "disconnect"
status to the host.  This would bring RING high, and the host would be
fooled into believing that a login session was being initiated.  After
the login session would timeout, the same process would be repeated,
ad infinitum.

The solution?  set the status echo dip switch off.  (i forget the actual
name of this switch)  This will disable the modem from sending the status
message (or the digit code, depending on the current status send mode)
I modified the image DTE_DF03.EXE to DTE_HAYES.EXE and included an
initializer string to be sent to the modem which would enable the status
digit codes to be sent back via software, so i could detect whether the
modem was receiving a dialtone, a busy signal..., to take appropriate
action.  For some odd reason, even though i never set the modem state
back, the problem mentioned above never returned.  (well, some
occasional flakiness somewhere causes the modem to enter this state,
but this is relatively rare)  I suppose on disconnect, the modem resets
to default state (which is determined by the switch settings and the
internal default settings prom.)

--stephen
(p.s. anyone wanting a copy of DTE_HAYES.EXE, let me know, i'll be happy
to send it out.  Oh, this is unix land? sorry)
-- 
$!#######################################################################
$! stephen dowdy (UNM CIRT) Albuquerque, New Mexico, 87131 (505) 277-8044
$! Usenet:   {convex,ucbvax,gatech,csu-cs,anl-mcs}!unmvax!charon!sdowdy
$! BITNET:   sdowdy@unmb
$! Internet: sdowdy@charon.UNM.EDU
$!      Team SPAM in '87!            SPAAAAAAAAAAAAAAAAAAAAMMMMMMM!
$!#######################################################################