[12:02] ok how do i run strace for a stalled configure. xulrunner is stallling to configure runnnin gdpkg [12:10] for example running dpkg --configure or upgrade xulrunner stalls when trying to be configured. i need to run an strace do i run it on the 5 or so processes or dpkg or on xul...? [12:28] i feel like this is useless http://paste.ubuntu.com/516206/ [12:36] would someone please fix xulrunner* [12:37] its stalling on 1.9.2 and 2.02.0~b8 [13:15] dpm, yt? [13:18] dpm, i wanted to replace stuff like by $HOST in my strings to make things simpler to translate, but i see $ is used in 1 lang: '$$Овај додатак ће имати приступ приватним подацима на .' (lang=sr) [13:47] hey fta, I'm back from lunch. As an alternative, you could use %(HOST)s as in python formatted arguments. It is something translators are used to, although I'm not sure it's the best solution [13:49] hi chrisccoulson, do think we should have a Firefox roundtable at UDS? Or shall we just have a talk in the corridors? [13:49] chrisccoulson, I mean a FF translations roundtable [13:49] dpm - we have a session for firefox-4 - you should probably come along to that :) [13:50] chrisccoulson, sure. Will there be time to discuss translations there? Is there a tight agenda? [13:50] we should be able to discuss translations there [13:50] i'm not sure what the agenda is for discussion yet ;) [13:51] but translations would be a good one to add [13:52] chrisccoulson, cool. Is there a blueprint for that already? I'll subscribe to it and add a translations agenda item to the whiteboard [13:52] dpm - i think it's other-desktop-n-firefox-4 (off the top of my head) [13:54] chrisccoulson, https://blueprints.edge.launchpad.net/ubuntu/+spec/packageselection-desktop-n-firefox-4 - is that the one? [13:54] dpm - yeah, that's it. it got renamed ;) [13:55] chrisccoulson, no worries :). Ok, subscribing and adding translations to the discussion agenda [14:01] k, done [14:10] ok i cant get anything when trying to run strace on xulrunnner or dpkg [14:10] * gnomefreak suggestss that someone looks at it but hey thqat is me [14:11] http://paste.ubuntu.com/516206/ is what i get [14:12] * gnomefreak going back to bed, i still feel alot like crap [15:25] dpm, %() should be ok [15:26] dpm, hm.. "(%)..." would become "(%(PROGRESS_PERCENT)%)..." is that readable enough? [15:27] maybe %{} [15:29] fta, yeah, perhaps, I'm just concerned about introducing a new custom variable markup, but yes, %{} would be more readable than the other options [15:42] grr, more \\n vs \n issues. i really need to write a proper test suite for that, to make sure that (grit2gettext ∘ gettext2git)(s) == s [15:49] micahg: i tried getting strace of dpkg and of xulrunner and both end up with the output that i posted at: http://paste.ubuntu.com/516206/ [15:50] gnomefreak: :( [15:50] not sure why i cant get one sorry [15:51] i was using strace -Ff -tt -p 2>&1 | tee strace-.log [15:52] that strace output looks pretty normal if you attached it to something that is hung.... [15:52] if i knew the args. to use with strace -Ff -tt 2>&1 | tee strace-.log maybe i would get a different output since it wont be running? [15:52] chrisccoulson: that is corrent [15:52] correct [15:52] xulrunner 1.9.2 and 2.0 are both hanging [15:53] during configure on upgrade [15:55] gnomefreak, couldn't you just try running dpkg --configure xulrunner-1.9.2 in strace? [15:55] chrisccoulson: i ran strace when xulrunner hung [15:56] but i didnt use xulrunner i use PID [15:56] configure -a [15:59] chrisccoulson: i can only stace dpkg since it is hung in dpkg [15:59] and output is same [16:00] chrisccoulson: strace -Ff -tt 2>&1 | tee strace-.log what arg would i use. [16:01] maybe that might work better instead of already having it running [16:03] micahg: chrisccoulson http://paste.ubuntu.com/516291/ [16:03] that gives a better strace :) [16:04] gnomefreak, are you using maverick or natty? [16:04] natty [16:04] daily builds? [16:04] yes [16:04] ok, that's expected [16:05] well i cant fonfigure the other packages like this [16:05] configure even [16:05] gnomefreak, bug 663294 (which i'm currently working very hard on to try and resolve) [16:05] Launchpad bug 663294 in gcc-4.5 (Ubuntu) "Firefox built with gcc-4.5 is a non-starter on i386 (affects: 1) (heat: 8)" [Undecided,New] https://launchpad.net/bugs/663294 [16:05] thats the same? [16:05] if you're on x86, yes [16:05] yes [16:06] ff works here i think [16:06] ok, so at least we've got a confirmation that this doesn't just affect ff-4.0 [16:07] marked confirmed and left a short message. thanks [16:15] chrisccoulson: eta on the fix? [16:15] i cant even install a package now [16:15] gnomefreak, not until i've figured out whats happening [16:15] 0ok thanks [16:16] welcome to natty :) [16:16] ;) [16:16] i'm discussing it with glandium atm too [16:16] :) [16:17] k [16:17] i going to lay back down for awhile maybe ill be back later today [18:09] jdstrand, linux/stable (6.0.472.63 -> 7.0.517.41) [18:12] fta: thanks. do you have packages available for lucid and maverick? [18:13] jdstrand, not yet, it's a fresh 0sec release :) [18:13] oh, heh [18:13] fta: just ping me whenever you have them and are satisfied with the packages [18:13] jdstrand, sure [18:13] fta: thanks for the notification [18:15] hello jdstrand ... i think i asked you something a few days ago, but now i cant remember :/ ... [18:19] asac: 08:41 < asac> jdstrand_: is [ 0.325499] audit: initializing netlink socket (disabled) from apparmor? [18:19] right [18:19] yeah. that one jdstrand [18:20] asac: I don't think it has anything to do with apparmor as a google search shows it popping up in various places. I don't know what it is tbh [18:21] kk [18:21] jdstrand: so i have a local build of firefox somewhere in /home/.... would our apparomor policies pick that up at all? [18:22] asac: if I were to guess, I would say that it is part of the audit framework, and that particular part is not enabled [18:22] * asac assumes not [18:22] asac: no, it wouldn't [18:22] kk [18:22] asac: not unless you defined a policy for it in /etc/apparmor.d/ [18:23] chrisccoulson: do you have a ffox 4 local build? [18:23] chrisccoulson: http://pastebin.com/sM6Ayt8J [18:23] i have that patch and wonder if that works better there than on my old 3.6 checkout [18:24] asac - i can't try it just yet, still investigating gcc-4.5 issues [18:25] chrisccoulson: np. === davida is now known as davidascher [18:54] chrisccoulson, chromium doesn't even build with gcc 4.5 [18:55] fta - oh, that's not good [18:55] firefox builds, but it then hangs at start [18:56] http://code.google.com/p/chromium/issues/detail?id=49533 [19:12] i things also fail on armel you can subscribe linaro-toolchain-wg team [19:13] to launchpad bugs for everything gcc related [19:13] chrisccoulson: whats the build failure? [19:13] (for ffox) [19:14] asac - it builds fine, but it just doesn't run [19:14] asac - bug 663294 [19:14] Launchpad bug 663294 in gcc-4.5 (Ubuntu) "Firefox built with gcc-4.5 is a non-starter on i386 (affects: 1) (heat: 8)" [Undecided,Confirmed] https://launchpad.net/bugs/663294 [19:19] chrisccoulson: nothing in bugzilla about 4.5 bustage? [19:19] asac - no, i've been talking to ted and glandium about it this afternoon [19:19] glandium can't reproduce it, although i still managed to reproduce it with the debian build of gcc [19:20] chrisccoulson: debian gcc build + ubuntu rest of toolchain? [19:20] or all debian? [19:20] i'm currently doing a build of mozilla-central with minimum build options [19:20] asac - just debian gcc [19:20] glandium also tried with our binutils, and it still worked [19:23] also libgcc1 ? ? [19:23] i would try swapping that out [19:23] asac - yeah, i'm running the debian libgcc1 too [19:25] chrisccoulson: goes away if you disable JEMALLOC ? [19:25] asac - i haven't tried that, but i would imagine so [19:26] the parts of the test suite that aren't using jemalloc are working ok anyway [19:26] chrisccoulson: also swapped out plain big libc? [19:27] ;) [19:28] anyway ... talk is cheap. me stops now [19:28] * asac goes away for a bit === kancerman_ is now known as kancerman [20:50] asac - ok, building from mozilla-central with minimal build options works fine with gcc-4.5 [20:51] so, it's either one of our build options or patches [20:58] cool [20:59] guess build options [20:59] though its odd. most build options are really coming from default gcc settings and not from our packaging [21:40] jdstrand, do i still need to create a bug for releasing chromium? [21:40] fta: yes please [21:51] jdstrand, bug 663523 [21:51] Launchpad bug 663523 in chromium-browser (Ubuntu Natty) (and 3 other projects) "6.0.472.63~r59945 -> 7.0.517.41~r62167 upgrade (affects: 1) (heat: 6)" [High,In progress] https://launchpad.net/bugs/663523 [21:51] i guess i should build with gcc 4.4 in natty for now :( [23:39] micahg, https://edge.launchpad.net/~ferramroberto/+archive/thunderbird [23:51] fta: thanks for the tip, I need to be more vigilant about the stable and daily PPAs