[04:05] if you have "start on started" by itself, when does the task start? [04:05] i'm having this issue where i can't seem to stop one of my tasks [04:05] i'm wondering if that's the reason [08:16] hi there [08:17] how does upstart knows to not start tty1.conf when console.conf is runnned on tty1 ? [08:19] afournier: console.conf is for lxc containers only and there both tty1.conf and console.conf are started. [08:19] you mean virtualisation ? [08:20] kinda.. ok [08:20] thanks [09:59] hi [09:59] how can I make upstart respan my service infinitely? respawn limit 0 5 for example? [10:23] fish_: either set a very high number ('respawn 999 1') or create a helper job that specifies something like 'start on stopped JOB=myjob RESULT=failed PROCESS=respawn\n start myjob'. [10:24] jodh: hrm, I just tried 0 and it seems to work [10:26] fish_: ah - just checked the code. You can indeed specify zero or in fact you can say "respawn limit unlimited" which is clearer. [10:27] jodh: would be even clearer to add this to the docs :) [10:27] jodh: but thanks for looking that up [10:27] fish_: indeed - an oversight. I'll add it... [10:27] jodh: cool! [10:43] fish_: ok, man page updated in upstream Upstart (soon to filter down to Ubuntu) and cookbook updated: http://upstart.ubuntu.com/cookbook/#respawn-limit , http://upstart.ubuntu.com/cookbook/#making-a-job-respawn-indefinitely . [10:52] jodh: cool! but it still only mention 'unlimited' where you might want to still specify the interval. guess you can do 'unlimited 5' as well, right? but that's not mentioned [13:07] dbus-daemon man page does not say anything about --activation=upstart, is this handled by a patch ? [13:09] 0003-upstart-add-upstart-as-a-possible-activation-type.patch [13:09] looks like it [13:13] afournier: yes, unfortunately that functionality was removed and I believe was not acceptable to upstream. [13:14] strange... [13:18] no place where those patches are downloadable ? [13:20] i found the source package [13:24] afournier: sure - http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/precise/dbus/precise/files/head:/debian/patches/ (or 'bzr branch lp:ubuntu/precise/dbus') [13:24] thanks [13:27] they dont fit well with the 1.6.18 [13:30] or maybe my version of dbus is already patched :/ [14:17] hi guys [14:17] I need to set an .env file (automating $ source file.env) [14:18] which stanza should I use to do that? [14:44] marianogg9: is this for a system job or a session job? [14:45] jodh: system [14:46] marianogg9: in which case, you need to source it in *every* script section your job has that needs those variables ("script", "pre-start", etc). [14:49] jodh, is there a chance that these patches get applied upstream ? [14:51] jodh: ok, I write it down like "pre-start script / source file.env / end script [14:52] *wrote [14:55] afournier: I think it's unlikely at this point - https://bugs.freedesktop.org/show_bug.cgi?id=34526 [15:08] jodh: https://gist.github.com/marianogg9/11258170 [15:15] marianogg9: that won't work if you want file.env to apply to app_gem - you need to turn the 'exec' into 'script'+'end script', add the 'source file.env' to that block before app_gem. [15:19] jodh: https://gist.github.com/marianogg9/11258507 ? [15:20] marianogg9: right. I suggest you test it fully (start/stop/restart) before adding the respawn stanza though. [15:20] ok :) [15:28] jodh, i will redo the patches myself then (working on an upstart layer for openembedded) hope a solution will be found very soon so i won't have to redo it :) [15:29] jodh: "main process (6939) terminated with status 127" [15:29] ouch [15:30] marianogg9: http://upstart.ubuntu.com/cookbook/#determining-why-your-service-fails-to-start, http://upstart.ubuntu.com/cookbook/#obtaining-a-log-of-a-script-section, http://upstart.ubuntu.com/cookbook/#checking-how-a-service-might-react-when-run-as-a-job