=== sadmac_ is now known as sadmac2 [16:31] anyone around who can help me get a script running [16:31] or point me at how to debug what is going wrong [16:32] are there any log files etc, or anything like thta [16:32] currently when i start my process i seem to get 4 processes for some reason [16:33] currently looking like i am going to have to stick to using rc.d to launch apps as nothing i try in upstart seems to work [16:33] and it never tells me why, [18:12] oly: /var/log/messages [18:57] thanks for the tip sadmac2 [18:57] when i set up a script and run sudo start script it shows 4 processes is that normal [18:58] or is it actualy launching 4 instances of the process for some reason ? [19:06] http://ubuntu.pastebin.com/m5edd07f2 [19:07] if anyone can tell me if that is correct it shows no processes the script then the processes after its started [19:07] as you can see there are 4 processes at the end instead of 1 [19:26] oly: the logs indicate upstart only spawned the first one [19:26] oly: it must be forking [19:27] you mean the fact that the start test3 command only printed one pid [19:28] just confusing because if i start it using /etc/init.d inadyn start [19:28] it would only show a single process [19:28] i was wondering if it was my lack of knowledge or something [19:29] been trying allsorts for ages [19:32] nothing appears in /var/log/messages by the way from starting that [19:35] hmm. not sure then. doesn't look like upstart is running more than oe [19:35] *one [19:35] maybe yoru init.d script passes more arguments [19:38] well, if i start inadyn from the terminal it starts fine and shows only one process [19:38] perhaps i will try and write some scripts for my own applications and see if i get the same behavior [20:12] wow [20:12] just confirmed an old mysterious 0.3 bug on 0.6 [20:12] Keybuk: details [20:14] assert() bug [20:14] you need two jobs to replicate it, so it shows up for people who've tried to do native boots [20:14] e.g. [20:14] foo: [20:14] exec foo [20:14] bar: [20:14] start on stopping foo [20:14] exec killall foo [20:14] then "stop foo" [20:15] Upstart will assert() [20:15] ie. if you stop a job, and another job that reacts to its "stopping" event *causes* the original job's process to exit [20:17] hmm [20:17] NetworkManager-vpnc's error messages are for shit. [20:17] "Failed." [20:18] "Reason: 6" [20:18] thanks a lot. [20:18] off-topic, but still infuriating [20:20] :-) [20:20] anyway, I've done a patch and will probably do a 0.6.3 early next week [20:21] sweet [20:25] *hints that you might want to upgrade from 0.3.x now :p* [20:27] Keybuk: we can't take the regression on that state-transfer thing, even as ugly as it is. Not having that pretty much kills the distro (Thanks Jakub!). [20:42] it was your patch ;) [20:42] this is why you should work to get all your patches upstream [20:43] and why hoarding patches by yourself and not working to get them upstream is bad [20:43] m'kay? :p [20:43] Keybuk: I thought upstream was dead as far as 0.3.* [20:43] it is [20:43] Keybuk: 0.3.10 was a shock to me [20:43] it's doubly dead now [20:43] as in I will now actively *not* fix critical bugs in it [20:44] what about LTS? [20:44] what about it? [20:44] you shipped 0.3 with it didn't you? [20:44] yes? [20:44] this doesn't affect the Support team's ability to provide telephone Support for it? [20:44] and the Security team will still provide Security fixes [20:45] we don't do what you guys do [20:45] once our releases are out, we largely *don't* backport non-security bug fixes to them [20:45] instead we suggest people update [20:46] oh god. the thought! [20:46] our customers expect us to time travel and make the bug not appear in the release they depolyed rather than upgrade a single package [20:48] :-) [20:48] yes, well [20:48] ;-) [20:49] life is long and hard [20:49] then you die [20:50] aka. life's too short to provide support for older releases [20:50] Keybuk: you haven't seen the ongoing support plan for RHEL 2.1 [20:51] Keybuk: the cost grows exponentially. If 3 customers keep that thing for 2 years RH will purchase a country