[00:14] Keybuk: yes 441339 is true in sysvinit [00:14] * warren checks for K* scripts [00:18] /etc/rc.d/rc1.d has K scripts of everything [16:10] Keybuk: So, we want to add telinit u functionality to upstart [16:11] Keybuk: we can do it the quick way, just make telinit u kill -TERM 1, or we can add an upstart message [16:11] right [16:12] We're currently looking at the latter. Which would you expect is preferable? [16:12] latter is harder [16:12] true, but its not that hard [16:12] though obviously more forwards portable since in 0.5 it would be a dbus command anyway [16:14] does doing it as a message make it more forwards portable? I would assume less since all that messaging stuff is getting replaced with dbus [16:15] right [16:15] but at least it would be ideologically the same [16:15] so 0.5 won't handle TERM? [16:24] it might [16:24] but that obviously won't be the preferred way to do it [16:24] since it doesn't fail well if init is /bin/bash ;) [16:25] true [16:34] Keybuk: so within init, should I a) move term-handler so its somewhere I can call it. b) copy the restarting code from term_handler. c) selfTERM? [16:36] self term is probably easier ;) [16:36] yes it is. [17:21] gaaah! [17:22] Apr 8 12:21:24 foucault init: control.c:364: Assertion failed in control_version_query: type == UPSTART_VERSION_QUERY [17:22] I ALREADY FIXED THAT ASSERTION! [17:22] why? why won't the problem go away when I fix it? [17:24] ... or maybe that just isn't the line I thought it was... [18:05] lol [18:05] assert() is your friend [18:06] Assertion bugs are the easiest to fix, just remove the assert() call. :-P [18:06] that's like saying that gout is easy to fix, just remove the leg === cdahlin is now known as sadmac2 [18:39] \\oo/ ← Zaphod === elmarco_ is now known as elmarco