[net.micro] Darn Xenix

farber@UDEL-HUEY.ARPA (Dave Farber) (05/27/85)

Is there anyone out there who has dug into AT Xenix's uucico
to tell me how to patch a decent DN timeout. The way those
yo yos set it, you CANNOT make a cross country dial pulse
call and make the timing.

If no one has found the patch, can someone point me at
" an alternative uucico that can be recompiled."

BTW , on Compuserve Chuck Forsberg of Omen Technology Inc
has documented some 58 known bugs in At xenix 1.0.

I will get it on the net Tuesday and send a copy to anyone
who wants it (58000 bytes).

Dave

lauren@RAND-UNIX.ARPA (05/27/85)

Dave,

You don't need to get anything else, the uucico you have with
your Xenix will work just fine.  All you have to do is set up
a separate little program (it takes about 10 lines of C code)
to do the actual dialing of your modem, then let that program
exec uucico (configure so that uucico doesn't try to do any dialing
on its own).  This way, your little program can do whatever is necessary
for dialing your particular modem, then hand over control to uucico.
Works like a charm on every system I've seen.

--Lauren--

kds@intelca.UUCP (Ken Shoemaker) (05/29/85)

another possible way to fool it is to put the dialing commands in the
L.sys file, and make it think you are a direct connect line.  Then,
if you timeout in the send/expect sequence, you can instead of putting
the modem in a "wait for connect" mode, you keep it in command mode,
echoing characters sent.  You then put in a few "A A" (send A, expect A)
in your L.sys file, then at the end, put the modem back on line, just
in time for your long distance connection to be made!  Don't laugh, it
works (at least for a Hayes Smartmodem, I actually used it here for a while!).
-- 
It looks so easy, but looks sometimes deceive...

Ken Shoemaker, Intel, Santa Clara, Ca.
{pur-ee,hplabs,amd,scgvaxd,dual,omovax}!intelca!kds
	
---the above views are personal.  They may not represent those of Intel.

sewilco@mecc.UUCP (Scot E. Wilcoxon) (06/03/85)

In <592@intelca.UUCP> From: kds@intelca.UUCP (Ken Shoemaker)
>another possible way to fool it is to put the dialing commands in the
>L.sys file, and make it think you are a direct connect line.  Then,
>if you timeout in the send/expect sequence, you can instead of putting
>the modem in a "wait for connect" mode, you keep it in command mode,
>echoing characters sent.  You then put in a few "A A" (send A, expect A)
>in your L.sys file, then at the end, put the modem back on line, just
>in time for your long distance connection to be made!  Don't laugh, it
>works (at least for a Hayes Smartmodem, I actually used it here for a while!).

Under Xenix 3.0 Lisa w/Priam and a Racal-Vadic VA3451 (ASCII-controlled
modem), the following works OK.  IT IS NOT HAYES COMPATIBLE, but the
method I use only requires that the modem not abort dialing if input
is received.  A VA3451 pulse dials, so uucico was timing out even on
local calls.  Of course, the entire script should be in L.sys as one
line without comments and a ^M is really the single character control-M.


somesys Any tty0a 1200 0	/* for machine somesys, Any time, use port
				tty0a at 1200 baud, direct connect */
*-^E-*-I-*-^E-* D MBER? 1234567 4567 ^M		/* send control-E until the
				* indicating command mode comes along, then
				give number, expect it to be echoed back,
				then give control-M to make dialing begin */
LINE--LINE--LINE--LINE		/* Here's the part you want.  VA3451 reports
				ON LINE when carrier detected.  So just keep
				sending a RETURN until we are ON LINE.  If
				your uucico goes through three timeouts
				before carrier detect, put more --LINE on
				here.  You can't do this if your modem
				aborts dialing when you send it a RETURN. */
^M ogin:--ogin:-BREAK-ogin:--ogin: mecc 	/* Now the easy part. */
sword:--sword: damocles

Scot E. Wilcoxon	Minn. Ed. Comp. Corp.     circadia!mecc!sewilco
93W15',45N03'	(612)481-3507 {ihnp4,mgnetp,uwvax}!dicomed!mecc!sewilco