[comp.unix.aix] Serial Port Access

ken@pyr.gatech.EDU (Ken Hall) (09/21/89)

I am having trouble logging into my aix system through the serial port on 
my ps/2.  The serial connection runs 7,e,1 at 9600 baud, through an     
ungermann-bass niu-180.  There shouldn't be a problem.  But I am new at
all this, so figure my ignorance into this equation.

I haven't done anything to configure the port except the usual software load
and creation of ttyp's and ttyn's.  I know there's a simple solution, but I
seem to be missing it in the documentation.

Any suggestions from the wise old network?

Ken

kevin@msa3b.UUCP (Kevin P. Kleinfelter) (09/22/89)

ken@pyr.gatech.EDU (Ken Hall) writes:

>I am having trouble logging into my aix system through the serial port on 
>my ps/2.  The serial connection runs 7,e,1 at 9600 baud, through an     

The "devices" command is documented somewhat poorly.  (Be sure to use
devices, even though I recommended against it a month or so ago!)

The critical values which are often set wrong are
DVAM and PRO.  DVAM=1 for modem (or modem-like) connection (i.e. using
CD, DTR, etc).  Use PRO=dtr for normal situations.  There are only about
6 different combinations, so sooner or later you should get what you
need for your connection.  One important note:  if you get them wrong
and a "getty" gets "hung", you will have to reboot after you change the
values, because changeing the values in devices does not seem to change
the real values until after a reboot sometimes.

-- 
Kevin Kleinfelter @ Management Science America, Inc (404) 239-2347
gatech!nanovx!msa3b!kevin

garnett@rpp386.Dallas.TX.US (John Garnett) (09/23/89)

In article <9195@pyr.gatech.EDU>, ken@pyr.gatech.EDU (Ken Hall) writes:
> I am having trouble logging into my aix system through the serial port on 
> my ps/2....
> 
> I haven't done anything to configure the port except the usual software load
> and creation of ttyp's and ttyn's....

Unless you are using AIX Access for DOS User's (AADU) over token ring or
ethernet, you shouldn't need to define any ttyn's [ptydev] (if anyone
has info to the contrary, let me know).

If you attaching a regular Ascii terminal or PC running a standard terminal
emulation package, then add tty's [ttydev] instead.

John Garnett
garnett@rpp386.cactus.org

oleg@gryphon.COM (Oleg Kiselev) (09/24/89)

In article <1129@msa3b.UUCP> kevin@msa3b.UUCP (Kevin P. Kleinfelter) writes:
>The "devices" command is documented somewhat poorly.

An understatement, to be sure.

>you will have to reboot after you change the
>values, because changeing the values in devices does not seem to change
>the real values until after a reboot sometimes.

You may need to kill getty you have running on the port so the new getty
picks up the new information (you can also kick init with "init -Q" to
spawn getty on a newly configured port if devices did not do it already).
-- 
			"No regrets, no apologies"   Ronald Reagan

Oleg Kiselev            ARPA: lcc.oleg@seas.ucla.edu, oleg@gryphon.COM
(213)337-5230           UUCP: [world]!{ucla-se|gryphon}!lcc!oleg

kevin@msa3b.UUCP (Kevin P. Kleinfelter) (09/26/89)

oleg@gryphon.COM (Oleg Kiselev) writes:

>In article <1129@msa3b.UUCP> kevin@msa3b.UUCP (Kevin P. Kleinfelter) writes:
>>The "devices" command is documented somewhat poorly.

>An understatement, to be sure.

>>you will have to reboot after you change the
>>values, because changeing the values in devices does not seem to change
>>the real values until after a reboot sometimes.

>You may need to kill getty you have running on the port so the new getty
>picks up the new information (you can also kick init with "init -Q" to
>spawn getty on a newly configured port if devices did not do it already).
>-- 
I "init -q"-ed and killed left and right.  When I had set-up
"pro=dc" (thinking "DCD") I found that killing getty did not seem to
help.  The next getty was hung too (even though I had changed pro to
dtr).  The only way I found to clear the situation (other than rebooting)
was to raise CD (via a breakout box) and kill the existing getty, and let
the NEXT getty SUCCESSFULLY open the port.  My conclusion was that
"asy" was getting confused. (Of course, I was pretty confused too!)

-- 
Kevin Kleinfelter @ Management Science America, Inc (404) 239-2347
gatech!nanovx!msa3b!kevin

jackv@turnkey.gryphon.COM (Jack F. Vogel) (09/27/89)

In article <1134@msa3b.UUCP> kevin@msa3b.UUCP (Kevin P. Kleinfelter) writes:
> [ Oleg and Kevins original exchange deleted..]
 
>I "init -q"-ed and killed left and right.  When I had set-up
>"pro=dc" (thinking "DCD") I found that killing getty did not seem to
>help.  The next getty was hung too (even though I had changed pro to
>dtr).  The only way I found to clear the situation (other than rebooting)
>was to raise CD (via a breakout box) and kill the existing getty, and let
>the NEXT getty SUCCESSFULLY open the port.  My conclusion was that
>"asy" was getting confused. (Of course, I was pretty confused too!)
 
Kevin, just to insert my two cents into things here...what does the node
that you are using look like? There is both a modem controlled and non-
modem controlled entry to the async device driver. There were also some changes
in the kernel between 1.x and what we work on now, so I am not sure, but
I believe there is a chance this could be causing the problem. If your
getty is running on the device with minor 0, make a device node with minor
128 ( or x+128 ) , or vice versa, and try running the getty on that. 

Just a wild stab since I really do not do standalone support, but figured
it would be worth a try.

Good luck!

Disclaimer: IMHO only!

-- 
Jack F. Vogel			jackv@seas.ucla.edu
AIX Technical Support	              - or -
Locus Computing Corp.		jackv@ifs.umich.edu