=== JohnFlux [n=JohnFlux@konversation/developer/JohnFlux] has joined #upstart [03:55] Hey all [03:56] to any developers.. could you include a package with the header files etc? [03:56] it's currently impossible to write a program that queries upstart etc === Starting logfile irclogs/upstart.log === ubuntulog [i=ubuntulo@ubuntu/bot/ubuntulog] has joined #upstart === Topic for #upstart: Upstart 0.3.9 | http://upstart.ubuntu.com/ | http://people.ubuntu.com/~fabbione/irclogs/ | https://lists.ubuntu.com/archives/upstart-devel/2007-October/000468.html === Topic (#upstart): set by Keybuk at Thu Oct 11 23:46:59 2007 === juergbi [n=juerg@80-219-21-79.dclient.hispeed.ch] has joined #upstart === GodEater_ [n=bryan@rockbox/staff/GodEater] has joined #upstart [08:52] JohnFlux: Huh? How do you figure that? [09:03] soren: ah hey! [09:04] soren: I have ubuntu [09:04] soren: If I look at what packages there are, there's no libupstart of libupstart-dev etc [09:05] JohnFlux: Why do you need them? [09:05] soren: if I look at the files that come with it just "upstart" there's no libraries nor header files [09:05] soren: so that I can make a gui for upstart [09:05] JohnFlux: Er.. [09:06] JohnFlux: You don't need header files or libraries to do anything like that. [09:06] soren: in particular, I want to be able to get a list of running services, for example [09:06] status [09:07] I'd much prefer a library to trying to parse the output of programs :/ [09:07] Sorry, I meant "initctl list". === JohnFlux nods [09:09] soren: so how about? a libupstart ? :) [09:10] There's no such thing. You can try convincing Keybuk that it'd be useful and you might get lucky :) [09:10] he's comes in here right? [09:10] I'll give that a shot thanks [09:10] Yeah, he'll be around in an hour or so, I'd guess. [09:10] btw I'm the maintainer of the kde "task manager" type thinig [09:11] thing [09:11] just looking to use upstart info to improve the info of what's shown [10:06] JohnFlux: i think libupstart is about to be killed off [10:06] see the mailing list post that the topic refers to [10:06] "libupstart goes away, which nobody else is using anyway." [10:09] instead there will be dbus aplenty === Md [i=md@freenode/staff/md] has joined #upstart === Keybuk [n=scott@wing-commander.netsplit.com] has joined #upstart === Keybuk amuses at the upcoming feature list for Bazaar [01:42] "Pack-based repositories" [01:42] THE RAPTORS ARE COMING! === soren wonders if Ubuntu 13.04 will be the rowdy raptor release. [01:47] ravenous raptor [01:48] Good one. [01:50] I need a brain-to-tomboy sync plugin [01:51] so when I think of neat code changes in bed, I don't need to have to try and remember them in the morning [01:53] Any reasonable person would suggest some sort of intermediate storage like a piece of paper, but I haven't managed to implement that either. [01:53] Althought it is a bit alarming to imagine how many good ideas have just vanished because I haven't bothered to write them down before falling asleep. [01:54] aye === mbiebl [n=michael@p4FD2E0A7.dip.t-dialin.net] has joined #upstart [02:44] Jc2k: nobody is using it because it's impossible to use [02:44] Jc2k: it's not actually shipped [02:44] Jc2k: I've been wanting to use it for ages [02:44] Jc2k: I asked on the mailing list about it like 6 months ago [02:45] Jc2k: it will be a shame if it's just killed off [02:46] why? isnt libupstart mostly about ipc? which will be handled via dbus, and thus *easily* accessible? [02:49] it's entirely about IPC === Amaranth_ [n=travis@ubuntu/member/Amaranth] has joined #upstart [02:50] so therefore JohnFlux has nothing to fear by libupstart death, in fact has much to gain from it. [02:51] he won't have to rewrite his front-end with every minor release, for a start ;) [02:52] actually that does sound good [02:52] JohnFlux: libupstart has never been a stable API [02:52] initctl uses it, but that's it [02:53] how far along is it btw? [02:53] the dcop thing [02:53] ? [02:54] how far along is what? [02:54] s/dcop/dbus/ [02:54] right hehe [02:54] Does it offer a dbus service at the moment? [02:54] Keybuk: how long till 0.5 basically [02:54] no [02:54] far along = not even started [02:54] heh [02:54] fair enough [02:55] do any other distros use upstart btw? [02:55] or just ubuntu? [02:56] just ubuntu at the moment [04:53] I've made a few commits to Upstart in the last few days [04:53] and lots to libnih :p [04:53] so I'm at least starting on the roadmap to 0.5 [04:54] \o/ [04:54] oh [04:54] questionage [04:55] can upstart be forced to run as a slave of "old" init? === Jc2k is thinking of a less 'controversial' migration path and can't spell [04:56] Having two init implementations running side by side is hardly less controversial than what Ubuntu currently does? [04:59] if something depended on upstart and upstart could be installed without ripping the heart out of your distrothen it would be a lot more viable [05:00] *shrugs* [05:01] the problem with being a slave is that it isn't init [05:01] so doesn't get afforded init's special privileges [05:01] bleh === Jc2k kicks unix [05:01] Jc2k: You should know that UNIX kicks harder. :) [05:02] :) [05:02] you could write an upstart parser for inittab :) [05:02] then it could just be a drop-in replacement for sysvinit [05:02] on trunk, that should be quite easy [05:03] interesting [05:05] inittab would be a ConfSource, register jobs with useful names like "inittab/$id", etc. [05:05] certainly upstart supports a superset of inittab's features === toidinamai [n=frank@i59F7021D.versanet.de] has joined #upstart === Amaranth [n=travis@ubuntu/member/Amaranth] has joined #upstart === Md [i=md@freenode/staff/md] has joined #upstart === Starting logfile irclogs/upstart.log === ubuntulog [i=ubuntulo@ubuntu/bot/ubuntulog] has joined #upstart === Topic for #upstart: Upstart 0.3.9 | http://upstart.ubuntu.com/ | http://people.ubuntu.com/~fabbione/irclogs/ | https://lists.ubuntu.com/archives/upstart-devel/2007-October/000468.html === Topic (#upstart): set by Keybuk at Thu Oct 11 23:46:59 2007 [09:32] :-) [09:47] now, of course, I have to work around the ways I was attempting to work around the lack of something I've now implemented [10:15] yay, ok init builds again [10:15] now for the tests [11:09] ==13312== Address 0x1C is not stack'd, malloc'd or (recently) free'd [11:09] that would be bad, then === Keybuk is having strange valgrind bugs [12:23] ==32728== Invalid write of size 4 [12:23] ==32728== at 0x8093F89: nih_list_cut (list.c:202) [12:23] ==32728== by 0x809405D: nih_list_destroy (list.c:246) [12:23] ==32728== by 0x8092C09: nih_free (alloc.c:330) [12:23] ==32728== by 0x8092C34: nih_free (alloc.c:336) [12:23] ==32728== by 0x808D243: parse_job (parse_job.c:278) [12:23] ==32728== by 0x805DF5B: test_stanza_start (test_parse_job.c:2185) [12:23] ==32728== by 0x8087030: main (test_parse_job.c:7303) [12:23] ==32728== Address 0xBEEFA174 is just below the stack ptr. To suppress, use: --workaround-gcc296-bugs=yes [12:41] oh I see