Subject: Re: one more xdm problem
To: maximum entropy <>
From: Bill Studenmund <skippy@macro.Stanford.EDU>
List: port-pmax
Date: 01/29/1999 21:21:24
On Fri, 29 Jan 1999, maximum entropy wrote:

> >Date: Fri, 29 Jan 1999 20:05:15 -0800 (PST)
> >From: Bill Studenmund <skippy@macro.Stanford.EDU>
> >
> >On Fri, 29 Jan 1999, maximum entropy wrote:
> >> 
> Can you actually reproduce this problem on NetBSD/pmax, or are you
> guessing?

Guessing. From experience w/ other X systems in this situation.

> Looking at the /etc/rc script, I see that xdm is not started
> asynchronously.  If it so happens that xdm opens the console before
> daemonizing itself, it's going to win the race every time.  If the
> correct process is guaranteed to win the race every time, it isn't a
> race condition.

It may be very deterministic, but it's still a race. If something happens
to xdm, then the getty fires up & you've lost the console.

> Additionally, xdm does not exit when you log out of the X session, so
> the "race" would stay won even after logging out.

Most of the time you'll be right. But the race is still there. You
probably would have no problem with this situation - if XDM died for some
reason, you could fix things up.

I'm familiar with the cluster Nick is setting up, and not all the admins
will be that sharp on NetBSD/pmax. With the race still in place, all it
would take would be for an X server to get wedged and one of the other
admins to come along & kill xdm to try to unwedge things. Then getty
starts to win the race. That doesn't sound too safe.

Take care,