Re: mc 4.5.30 crashes X session

[jg pa dec com (Jim Gettys)]

> From: "Brandon S. Allbery" <allbery@ece.cmu.edu>
> Resent-From: gnome-list@gnome.org
> Date: Sat, 17 Apr 1999 08:17:44 -0400
> To: Devin Butterfield <dbutter@ix.netcom.com>
> Cc: gnome-list@gnome.org
> Subject: Re: mc 4.5.30 crashes X session
> -----
> In message <37183C06.2DA4A197@ix.netcom.com>, Devin Butterfield writes:
> +-----
> | Well........Hmmm....are you suggesting that this problem "x io error
> | occurred" is NOT gnome related???
> |
> | If so, then why is it that I have NEVER...
> +--->8
> 
> Because GNOME does something differently.  That doesn't change the fact that
> it should NEVER be possible for a bug in an X client to kill the X server.
> This isn't Windows; we don't just shrug off bugs in the basic system as
> something to be worked around in the application.
> 
> That doesn't mean that there's no point in looking for workarounds, but it
> *does* mean that the real problem should be found and a fix sent to the
> XFree86 folks before implementing the workaround.
> 

Yup.  On various workstations I've used in the past, I've had X servers
where I've been logged in for many months at a time without failure or
logout.  Alot of current PC servers are not as stable as this, unfortunately.

The core X server is VERY well debugged; almost all bugs will be
in device dependent code for your display board.

Say after me:

If your X server crashes, it is ALWAYS a bug in the X server.
If your X server crashes, it is ALWAYS a bug in the X server.
If your X server crashes, it is ALWAYS a bug in the X server.

			- Jim Gettys