ps bug in ODT 1.0?

[Sean Eric Fagan]

In article <832 at uswnvg.UUCP> cjackso at uswnvg.UUCP (Clay Jackson) writes:
>I'm in the midst of trying to compile and install the korn shell
>on my ODT 1.0 system (and having some "interesting problems) and
>have run into one that is VERY bizzare.  

Do not use job control.  Either undefine it, or, in your .kshrc, put in a
'set +m'.  The problem, once again, is that job control in 3.2.0 and 3.2.1
(which is what ODT 1.0 uses as a kernel) is so completely broken that it has
no chance of working.  Preferably, btw, you would upgrade your OS, at least,
to 3.2v2, which has working job control and includes ksh.

>I got curious, and, from another multiscreen, did a ps -fu <me> to see
>what was running.  The ps display says that the ppid of the 
>scoterm window is 1.  Everything I know about Unix says that process 1
>is init.....

Yes, that's right.  The ppid:  the parent process.  Its parent has died,
so it gets process 1 as its parent.  This is normal for unix.

>While you're at it, I'd sure appreciate
>it if someone from SCO might be able to give me some idea as to the
>difficulties THEY ran into trying to put ksh on ODT 1.0, and if maybe
>I'm attempting something that's just plain impossible?

I did get ksh working for 3.2.0; however, I did that as an attempt to get
job control working, and my kernel never made it very far outside of the
office 8-).  But, if you are set on using ksh with odt1.0, all you should
need to do is the set +m I mentioned above.  (This prevents ksh from doing a
bunch of things, including putting each command set into a different process
group).

-- 
Sean Eric Fagan  | "I made the universe, but please don't blame me for it;
sef at kithrup.COM  |  I had a bellyache at the time."
-----------------+           -- The Turtle (Stephen King, _It_)
Any opinions expressed are my own, and generally unpopular with others.