>> [ 12384.156] (II) UnloadModule: "ws" >> [ 12384.156] (II) UnloadModule: "kbd" >> [ 12384.515] (II) Server terminated successfully (0). Closing log >> file. > > It woud be the lines just above those three that show if the X server > got a signal... If not, it means that it is the window manager that > exited (probably because he got the signal...).
The lines right seem to just be startup... [ 11455.338] (**) ws: /dev/wsmouse: YAxisMapping: buttons 4 and 5 [ 11455.338] (II) XINPUT: Adding extended input device "/dev/wsmouse" (type: MOUSE, id 7) [ 11455.406] (**) /dev/wsmouse: (accel) keeping acceleration scheme 1 [ 11455.406] (**) /dev/wsmouse: (accel) acceleration profile 0 [ 11455.406] (**) /dev/wsmouse: (accel) acceleration factor: 2.000 [ 11455.406] (**) /dev/wsmouse: (accel) acceleration threshold: 4 [ 12384.156] (II) UnloadModule: "ws" > If you're using xenodm, you may find some information in > ~/.xsession-errors or /var/log/xenodm.log. There's nothing in ~/xsession-errors , and /var/log/xenodm.log is also just startup info. Also, by suggestion I've run the iperf test through ssh, and I am still able to recreate this. >> I've also made X exit on a fresh install with no custom setup. > > Strange, normally fvwm doesn't die if one of its clients crashes, and > iircs runs them in separate process groups to avoid being killed by a > signal sent to a process group. > > Using ktrace there would help, at the expense of gigantic ktrace.out > files if run on fvwm, the X server or iperf. I'll try to give it a go...
