On Mon, Jan 11, 2010 at 12:46, Martin Langhoff <martin.langh...@gmail.com> wrote: > With the help of Mikus, a specific shutdown bug has been narrowed > down, and I think it's a race with the X.org server needing a little > bit longer to shut down cleanly. > > More details on the bug itself at http://dev.laptop.org/ticket/9943 > ("Shutdown process stops in text console when USB keyboard in use"). > > But it is not clear to me who is killing X.org in these builds. I am a > bit old fashioned (ie: I first look at rc6.d) but probably hal, udev, > hot/cold plug et al are the murderours characters. > > (It is a low priority bug as it's fairly specific in the steps to > trigger it -- but fixing bugs is what we're here for... help welcome)
In case you end up trying to trace this yourself, this systemtap script has been useful for me in a similar situation: http://sourceware.org/systemtap/examples/process/sigkill.stp Regards, Tomeu -- «Sugar Labs is anyone who participates in improving and using Sugar. What Sugar Labs does is determined by the participants.» - David Farning _______________________________________________ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel