• Please review our updated Terms and Rules here

Help with Xenix

update: logging in from hyperterm is really easy.

On the xenix machine:
enable /dev/tty1a
Reboot just to make sure, wait for the "login:" screen to appear.

Now, on the XP box.
Set VT100 emulation on hyperterm, connect to the right serial port with: 9600, 8 bit, no parity, 1 stop bit, no handshaking.
Now press backspace until you see

xenix1!login:

now you can login remotely. How to transfer file is still a mistery though, since you're only working remotely on the machine... you need a terminal program on the xenix box to send files, from what I can understand

By the way, which version of xenix are you using?
 
Last edited:
Well I am about 400 km from Rijeka I would guess...

As for xenix - its Release 2.3.4 of SysV apparently.

edit: Whats with hitting backspace by the way? What is that actually doing?
 
Last edited:
edit: Whats with hitting backspace by the way? What is that actually doing?

Uhm apparently nothing that the n or f or enter or whatever key couldn't do.
It just "refreshes the screen" if you want... I think simply it tells the other machine that someone is there, so it can send again the login prompt.

I saw a few tutorials that mentioned backspace, so I guess that

1) those howtos were written by the same person, who likes to use backspace so much... we all have a favourite key, right?

2) it's a commond rule of the good ol' days... when terminals were terminals and men were men.

3) Maybe it's just because sending backspace won't be recognized as the first character of the login string, but simply ignored
 
: )

Speaking of hhgttg refs, I was somewhat amused to see that the first page of the uupc/extended manual tells me "Don't Panic! (Much)", although I don't find the letters they used to be particularly big or friendly. At least they tried I guess.
 
Lol, you're right.

I'll see if it's possible to get it working somehow... the manual was written by a monkey... :rolleyes:
 
Last edited:
update: logging in from hyperterm is really easy.

On the xenix machine:
enable /dev/tty1a
Reboot just to make sure, wait for the "login:" screen to appear.

Now, on the XP box.
Set VT100 emulation on hyperterm, connect to the right serial port with: 9600, 8 bit, no parity, 1 stop bit, no handshaking.
Now press backspace until you see

xenix1!login:

I tried this last night, but having pressed backspace at least 40 or 50 times (getting desperate) nothing seemed to happen. So I am beginning to suspect that the guy in the computer shop who sold me the serial cable had absolutely no idea what he was talking about when he assured me (I asked at least 3 times damnit) it was cross-linked. I will try a different guy, shop and cable asap, although that will probably turn out to be friday at earliest.

/Majority of the guys in the shops have never even heard of a null modem, which I thought was a bit strange, they are not exactly arcane or esoteric, not like floppy drives...
 
I can assure you that about 5/6 years ago I had trouble to find a cross-linked serial cable... so don't be so surprised. Most of the shops won't sell you a cd-rom unit either. Anything older than 2 years is simply discarded... from shops and from memory. Try in a shop that sells also electronic components, usually they are more experienced.

Btw, backspace is needed only 1-2 times :mrgreen:

Here's what you could do to check your cable:

Check it phisically:
Have you got a multimeter? check the cable with it. Use the resistance-meter if there is no continuity test, if it's = 0 then you are touching the same cable on the 2 connectors, if it's a BIG number, then the pins are not of the same cable... Hope you understand. My english is getting worse every day :mrgreen:

Check it with hyperterm and xenix:
disable /dev/tty1a
echo ruthere > /dev/tty1a

(maybe you should try also tty2a, just to be sure)
If you can see in hyperterminal what you echoed, then the cable is right and something else is wrong
 
Last edited:
So I tried what you suggested re: testing the cable, twas a bit frustrating, but after trying tty1A and tty1a and tty2a, I finally got some response from tty2A.
But not very much, just !P. Oops, then realised baud rate was wrong - I set hyperterminal to 9600, and now it seems to work:

In root of xenix box:
Code:
echo pleasework > /dev/tty2A
echo thanks > /dev/tty2A

On HyperTerminal:
Code:
pleasework
          thanks

So, can I now be reasonably sure the cable is indeed cross-linked?
Meanwhile, no success re logging in with backspace, no real joy with uuinstall. Am persevering.
 
Last edited:
Yes, we are resonably sure that the guy in the computer shop was somewhat competent (and, consequently, that the cable is cross-linked) :D

Now, before getting desperate, you could try to do this:

cat /etc/systemid
cat /usr/lib/uucp/Devices
cat /usr/lib/uucp/Systems

those are the files you configure with uuinstall, post them and we will see if something is wrong with it, maybe. Uh, I don't expect you to trascribe them line per line... you can do a
cat /etc/systemid > /dev/tty2a

and then copy the output from hyperterm. This will be a bit screwed up (no carriage return, probably), so you'll have to fix this in notepad or msdos edit or whatever you want.

By the way, don't post them here (it will take a lot of space!), use http://rafb.net/paste or a similar site instead.

Last but not least: are you sure that /dev/tty2A works and /dev/tty2a doesn't? I just noted the capital A and I'm a bit confused about that :confused:
I don't remember very well what's the difference, I know that the one with the capital letter is used for modems.

update (Thanks for this go to a friend that is helping me a lot with xenix :) ):
From ckubwr.txt in the kermit source:
"SCO systems tend to use different names (i.e. drivers) for the same
device. Typically /dev/tty1a refers to a terminal device that has no
modem control; open, read, write, and close operations do not depend
on carrier. On the other hand, /dev/tty1A (same name, but with final
letter upper case), is the same device with modem control, in which
carrier is required (the SET LINE command does not complete until
carrier appears, read/write operations fail if there is no carrier, etc)."
 
Last edited:
right, but I don't seem to get it working :confused:

I simply gave up on that, but if you know how that's possible please explain...
 
:-/ Cleanup proved a bit tricky...
cat of Devices is here
cat of Systems is here

Second in particular seems to be garbage... <erm>

Back in uuinstall:

In devices, was trying to set up device at tty2a, but it seems that there is
Code:
>>> No such character device in /dev
which I don't really understand...

Incidentally, seems it doesn't matter if I use /dev/tty2A or /dev/tty2a for my echos and cats, it arrives in a mess at Hyperterminal anyway...
 
First of all, /dev/tty2a and /dev/tty2A are the same device, only that the second supports hardware handshaking. If the pc is not sending CR/LF (if THAT is the problem) that's not a port problem.

I cannot understand the second file, but before that we can set up the first:
Code:
Direct tty1A - 9600 direct
This says that tty1A is suitable for a direct connection at 9600 baud.
You need to add a
Code:
Direct tty2a - 9600 direct
line (or just uncomment it from the file. Hope you know vim, lol. Otherwise, to delete a character, just press x when the cursor is on the character. To save and quit, press :wq but I guess reading the man page will help you more than me).

in the Systems file do you have only the sosco entry? Then we are missin something... you have to let uupc know who can call and be called.
I will answer to this later, now I have to finish my breakfast and run to the university otherwise I'll miss the lesson :mrgreen:
 
added that line to Devices.
I have an entry imaginatively called malcolm in my Systems file, as well as the sosco entry, not sure why couldn't see it properly, will try again...

Meanwhile, I just realised that you have already given me a temporary solution to my data manipulation problems:

Short term solution:
I can get the instrument software on the xenix box to output the data files as ASCII based report files. I can then set up HyperTerminal to capture text in FileIWant2See.txt, and use
Code:
cat FileIWant2See > tty2a
to copy all my data (which is nicely comma delimited by the instrument software) across to that file, and import it into excel.

Mucking around with ascii report files is pretty much what I did before anyway, its just a bit more awkward this way.

The only problem is that I don't think I can backup the raw data files in this manner, as they are allegedly binaries.

I still haven't managed to get anything to work in opposite direction either - not managed to send anything from PC to xenix box.

Still hoping to:
Get this uucp uupc/extended thing sorted out, so I can actually copy files across properly from either computer whenever I need to.
 
you could probably write a little program that access the parallel port and captures the raw data, deleting the spaces and adding CR/LF.

When i get home I'll try to send a file between two xenix machines with cu and %take, if that works I'll send you a xenix vm with kermit on it, so you can transfer it on your machine and then use it to copy files (it's way better than uucp, trust me. And you only need hyperterm to transfer files then).

update: you are terribly lucky my friend. Com0com has stopped working, so every time I get to the login prompt, qemu hangs. I tried reinstalling it but no way. Now i'm downloading a mini xp to install in vpc, and I'll try it there (then I'll have an emulated xp with two emulated xenix on it... all running on my poor 1.6 ghz atom processor... omg)
 
Last edited:
Sorry, it didn't work. I cannot get the two boxes to communicate, or even to use hyperterm from a box to another. If you have some patience I will try on my real xenix box (a speedy 486!) with a real nullmodem cable and a software terminal (if someone wants to donate a real vt100 to me... :rolleyes:)

Btw, tomorrow 2 of my wisdom teeth will be extracted, so maybe I won't feel very well. Don't expect me to move computers around the house and configure them in the next few days. :rolleyes:
My personal feelings are perfectly described by "The scream" (Munch's painting) at the moment.
 
Ouch. Take it easy dude, the cat file > /dev/xxx work-around may not be terribly elegant, but its working fine for the moment - once I have transferred and number crunched the hundreds of samples I have currently waiting on the xenix box, I will get back to playing around with the rest of it, so really, no need to rush job sorting the rest of my problems : ).
 
Ok then. When you need it drop a line here and we will start working on this again :)
By the way I'm feeling well already, it didn't hurt too much.
 
Back
Top