Have to agree with that, it only got worse with RH 9 and happened very
regularly in public clusters. I think that whenever it happened some
applications from previous user were hanging and the only way for a non
privileged user to fix it is to reboot the machine in question (very
annoying). My solution was to patch pam_console code to recognize a new
option that forces takeover of ownership regardless of pam counting.
That way I can force it for gdm logins.
There was already some talk on this list about making sure the logout
shuts down more gnome apps (like gconfd-2 and others), hopefully that
will also help with this problem.
Josko
Alan Cox wrote:
On Tue, Dec 23, 2003 at 08:52:33AM -0500, Michael K. Johnson wrote:
>The ONLY time I've ever seen anything like this it has been by design:
>the second user logs in before the first user logs out (on a different
>terminal that is still considered to be the physical console) and there is
>
>
It happens regularly in the university computer society systems going back
since RH7.x and so far nobody has figured out why. The pam code gets counting
logins wrong in at least one module too which may or may not be related.