Code review comment for lp:~andrzejtp2010/lightdm/lightdm-trunk-xephyr-multiseat

Hello Robert,

thank you for addressing the issue,

W dniu 16.04.2013 04:26, Robert Ancell pisze:
> OK, so the two issues are:
>
>
> 1. LightDM doesn't handle stale X server locks. I've fixed this in lp:~robert-ancell/lightdm/stale-xserver-lock
>
> 2. I'm still not understanding the requirement to have all three X servers running on the same VT. Could you open a bug with this requirement and attach the logs from /var/log/lightdm? I'm not sure if LightDM can properly handle the case you're describing - you have Seat:0 being the main X server and Seat:1 and Seat:2 running inside it. There's nothing to stop Seat:1 starting before Seat:0 and if Seat:0 and Seat:1 are Xephyr servers then they shouldn't have VTs?!

Here is the bug report:

https://bugs.launchpad.net/lightdm/+bug/1169724

I confirm that without the common-vt patch the solution does not work
but with the patch it works well. The difference in the log files that I
was able to spot is that in the successful run all the X servers report
using and activating the VT7, while in the unsuccessful run each X
server has its own VT number, e.g. VT7, VT8, VT9 and the computer locks
to display a blank console screen and a blinking cursor. No combinations
like Ctrl+Alt+F<X> work, nor Alt+F<x> to switch out of this locked
state; it is possible to ssh into the machine, though.

Thanks,

Andrzej