[10:00] fta_: does fta@ubuntu.com now work? === fta_ is now known as fta [10:00] bug #227711 [10:00] Launchpad bug 227711 in xuxen-eu-spell "hunspell-en-us conflicts with thunderbird (unversioned)" [Undecided,New] https://launchpad.net/bugs/227711 [10:00] no idea, try me :) [10:00] * asac sends a mail ;) [10:01] is it supposed to be a redirection/alias ? [10:04] fta: not sure :( [10:04] fta: at least its still not set up [10:04] but i am sure its supposed to work [10:04] maybe one has to do something? [10:05] fta: https://wiki.ubuntu.com/Membership [10:05] reading.. [10:08] looks like it should happen automatically [10:09] (reason: 550 5.1.1 : Recipient address rejected: User unknown in virtual alias table) [10:09] yeah [10:10] either its a batch [10:10] or a manual action [10:10] i'd wait maybe 12 more hours [10:10] and then bug them in #launchpad [10:10] while talking to mx.canonical.com, so it's not on my side ;) [10:10] yeah [10:10] i saw that [10:14] inode->properties->next->next->next->next->next->next [10:14] oops [10:14] i will ping daniel [10:14] yeah ask him ;) [10:15] fta: that pointer deref looks like art ;) [10:15] in C you certainly will risk a null-deref ;) [10:15] debugging a crash [10:16] it's gdb, i don't code like that [10:16] thats in the gdb code? or do you type that in (gdb) ? [10:16] i typed that in gdb [10:17] well, i clicked that in ddd [10:17] ;) [10:17] ok [10:17] thats ok i guess [10:19] http://www.sofaraway.org/ubuntu/tmp/ddd.png [10:20] ddd is still an old ugly motif app, no gtk interface, at least not in ubuntu [10:21] yeah. but not that bac [10:37] "It should be working before the end of the week" [10:41] morning [10:42] asac: debugging on xulrunner... will the configure option "--enable-debug" enable the module logging, or do I have to do something more? [10:53] IntuitiveNipple: hi [10:53] IntuitiveNipple: wait a second [10:53] thanks :) [10:54] IntuitiveNipple: http://paste.ubuntu.com/44847/ [10:54] IntuitiveNipple: i'd suggest not to use a package [10:54] IntuitiveNipple: for that you just get the source tree [10:54] (either from package or from CVS) [10:54] and then you add that paste above as .mozconfig in the top level dir of the tree [10:54] e.g. inside mozilla/ [10:54] then you build with: [10:54] make -f client.mk build [10:54] and [10:54] sudo make -f client.mk install [10:55] IntuitiveNipple: oh adapt or drop the MOZ_OBJDIRT [10:55] from .mozconfig [10:55] so --enable-debug is sufficient then [10:55] i just use that dir because i have a fast disc there [10:55] IntuitiveNipple: --enable-debug + --disable-optimize [10:55] IntuitiveNipple: but take care [10:55] the package usually strips [10:55] which might not be what you want [10:55] thus i suggest just to use make -f client.mk like above [10:56] The package always gets built since the build-process is shipped off to a buildd - I don't let my development machine get filled up with dev header packages [10:56] IntuitiveNipple: once you have finished that just take care that the /etc/gre.d entry that points to your system install has a commented out PATH so firefox cannot find it [10:56] IntuitiveNipple: well. then try. [10:57] IntuitiveNipple: also pass --disable-strip [10:57] Interestingly, the packages are built with that [10:57] yeah [10:58] IntuitiveNipple: because dh_strip strips them [10:58] thats why i say its easier to build the tree ;) [10:58] maybe it works well with -dbgsym packages [10:58] never mind... this'll work for my purposes [10:58] install pkg-create-dbgsym [10:58] otherwise you wont get the -dbgsym packages [10:59] as long as the PR_DEBUG is enabled from configure that'll enable logging [11:01] IntuitiveNipple: NSPR_LOG_MODULES=HelperAppService:5 [11:01] is the switch you want to use then [11:01] asac: you dont have a nvidia card with 3d drivers i Intrrepid? [11:01] s/Intrrepid/Intrepid [11:01] unfortunately i dumped my nvidia system is rotting in the closet [11:02] IntuitiveNipple: and add the patch to the package that i gave you ... most likely it fixes this bug. and if not its surely where we want to build on top of [11:03] thanks maybe ill find timo today and find out why 3d isnt enabled and cant be for some reason [11:03] I'll do that later; for now I just want to track what is happening and why [11:03] IntuitiveNipple: well. if you want to scrutinize you ;) [11:03] IntuitiveNipple: my guess is that its really the issue we see fixed in the patch [11:04] e.g. it finds an app in gnome registry for the extension .txt [11:04] and then it uses that to get the mime-type and accidentially finds a match in mailcap which makes it use that app [11:04] The 'why' is more important right now since it might point to wider issues, plus, it might indicate a user work-around that doesn't require a new package [11:04] but tries to run it with gnome_vfs_mime function [11:04] which causes the app issue [11:05] "app not found" [11:05] IntuitiveNipple: we will have an upload soon anyway. so the sooner you find the patch the better. users properly prefer a fix instead of a workaround hidden in some launchpad id [11:06] the crux seems to be why "gnome_vfs_mime_get_default_application()" fails, or the code converting the result to mozApp fails. [11:07] IntuitiveNipple: imo it doesnt fail [11:07] what fails is the running of the process [11:08] and that happens because it tries to run the process with the mailcap way even though it has a struct for gnome [11:08] or vice-versa [11:08] but well. if you verfiy the cause i am happy too [11:08] i just dont want to waste your time - which would be more or less the case if this patch works - unless you like debugging of course ;) [11:09] let me know what you find ;) [11:13] "what fails is the running of the process" - you mean the external handler? No, I don't think that is correct. In my debug logs it shows that the alternative to VFS path is taken, but mDefaultApplication=null ( that is before it tries to execute it) [11:13] asac, my email should be fine now. [11:15] fta: testing [11:16] fta: did it require manual interaction? [11:16] fta: send mail btw [11:17] IntuitiveNipple: i tried the get_default_application for text/plain yesterday [11:17] it worked [11:17] where is brusles(sp) its a country i think [11:18] IntuitiveNipple: http://people.ubuntu.com/~asac/test/ [11:18] thats the code for it [11:18] asac, yes, it's a 2d cron task [11:18] gnomefreak: most likely brussels [11:18] ? [11:18] asac, https://answers.edge.launchpad.net/launchpad/+faq/71 [11:19] ah ... then its their own problem if people bug them ;) [11:19] thanks for fixing spelling [11:19] they should run it twice a day :) [11:19] gnomefreak: brussel is in belgium ;) [11:19] asac: ah thanks [11:20] http://maps.google.com/maps?f=q&hl=en&geocode=&q=brussels&ie=UTF8&ll=50.867511,4.367752&spn=0.186342,0.397568&t=h&z=11&iwloc=addr [11:20] gnomefreak: ^^ [11:20] thanks looking [11:21] gnomefreak: its more or less the European capital ;) ... i think the parliament is there ;) [11:21] * gnomefreak notes google maps [11:25] * gnomefreak was planning on playing with compiz today while i finish chatzilla and push both to PPA [11:26] xorg.conf looks right [11:29] asac, got it [11:58] asac: mozilla bug 440075 <- apply! [11:58] Mozilla bug 440075 in Location Bar and Autocomplete "Location bar doesn't encode the address when using UNIX copy/paste (selecting)" [Normal,Assigned] http://bugzilla.mozilla.org/show_bug.cgi?id=440075 === vk5foss_ is now known as kgoetz [12:11] armin76: please get that landed ;) [12:12] me? it needs to be reviewed first :P [12:13] but it isn't going to be on 3.0.2 anyway, so better get it applied [12:53] my dquilt script is very handy now that quilt is no longer doing those patches symlinks [13:10] anyone know the nvidia command to generate xorg.conf? [13:11] did you try without xorg.conf ? [13:11] or dpkg-reconfigure -phigh xserver-xorg [13:11] fta: yes [13:11] it doesnt help [13:13] # You should use dexconf or another such tool for creating a "real" xorg.conf [13:13] # For example: [13:13] # sudo dpkg-reconfigure -phigh xserver-xorg [13:14] yeah it faile dto help i normally use it but im not sure what went wrong so i used sudo nvidia-xconfig (normally this helps) brb reboot [13:43] asac: Would you like to *shoot* the ubuntuforums web admins?!? [13:49] IntuitiveNipple: i have no connection to them ;) [13:49] IntuitiveNipple: why? [13:50] good... we can throw cream cakes at them! :D [13:52] I'm just writing up a summary in the bug report... basically it is a combination of things. #1 is misconfigured mime-types (plain/text instead of text/plain !!) amongst others, BUT there's also in issue with xulrunner where it fails the "plain/text" lookup, then must look at the file extension, and gets the *correct* mime-type from that, but *then* doesn't store the resulting app that gnome-vfs returns in the mDefaultApplication a [13:52] nd therefore fails to load the helper [13:52] It'll be clearer in the bug write-up [14:01] IntuitiveNipple: afaict xulrunner always tries to use the file extension [14:02] it uses the mime-type to find the file extension and uses that to lookup [14:02] IntuitiveNipple: anyway. thanks [14:02] IntuitiveNipple: have you tried the patch? [14:06] No, I've just looked at the logs of my tracing reports. Gonna figure out why xulrunner 'forgets' the launcher that gnome-vfs gives it when it uses the file-type not the Content-Type. [14:09] Any idea who the web-server admins are? [14:12] IntuitiveNipple: admins of what server? [14:12] I'm checking in #ubuntu-website [14:12] Volans: ubuntuforums.com [14:13] the admins of the server are the canonical-sysadmins (see the relative page on launchpad) [14:13] just got there, thanks :) [15:42] asac: update: no change in alpha 5 re NFS/FF3 issue [15:51] pmatulis: ok. [15:51] pmatulis: how is network managerd? [15:52] networkmanager? [15:52] or ifupdown? [15:57] asac: one moment, overflowing here [16:13] http://www.journaldugeek.com/2008/09/09/google-chrome-utilise-par-1-des-internautes-francais/ [16:14] 1% of french users switched to chrome, mostly from firefox [16:20] asac: network-manager is looking good (wired and wireless); ifdown eth0 says 'device not configured' [16:20] fta: how do you know that they switched from firefox [16:20] % [16:20] pmatulis: maybe this issue just happens with NM? and if you use ifupdown for networking it works? [16:21] pmatulis: point is that NM still has some issues with tearing down interfaces before we umount nfs shared [16:27] asac: if you want me to test something, best to let me know the exact steps you want me to go through (i.e. how to disable NM, not sure these days, edit interfaces file?) === asac_ is now known as asac [16:45] asac, they say firefox dropped 0.7% and ie 0.21%, accounting for the 1% of google [16:49] pmatulis: 1. ensure that NetworkManager isnt installed [16:49] 2. setup your eth0 in /etc/network/interfaces [16:50] 3. check if its still reproducible [16:54] asac: that's weird [16:56] pmatulis: ? [16:56] fta: how did they measure that? [16:58] fta: how recent of a ffox 3.1 build are you running? [16:59] ~3 hours old [16:59] fta: does it crash for you regularly? does jemalloc build work? [16:59] crash at shutdown [17:00] asac: evolution is a reverse dependency of network-manager [17:03] no problem with jemalloc [17:03] asac, but a lot with asound vs p-a [17:05] pmatulis: hmm of network-manager itself? [17:05] pmatulis: not libnm-glib? [17:05] asac, 3.1~b1~hg20080909r19012 [17:06] ok thanks [17:06] https://bugzilla.mozilla.org/show_bug.cgi?id=453933 [17:06] Mozilla bug 453933 in ImageLib "Crash in [@ start_pass_fdctmgr]" [Critical,Unconfirmed] [17:07] nope, never seen that [17:10] pmatulis: for the sake of trying you could also move the NetworkManager files in /etc/dbus-1/events.d/ in some other directory [17:10] and restart [17:10] that should make NM not start (assuming you are on hardy) [17:11] on /win30 [17:11] sorry that went to wrong channel [17:57] asac: i actually performed a package removal of NM. 'ifdown eth0' works, 'ifup eth0' complains of missing file: /etc/network/run/ifstate as well as ifup script problem: '[: 164: 0: unexpected operator' [18:00] asac: however, the interface comes up fine [18:01] asac, stats are from xiti: http://www.xiti.com/ [18:04] pmatulis: what do you have in /etc/network/interfaces? [18:04] pmatulis: did you reboot? [18:05] asac: auto eth0 [18:05] asac: iface eth0 inet static [18:05] asac: address ... [18:05] asac: netmask ... [18:05] asac: gateway ... [18:05] i dont even have /etc/network/run/ifstate [18:06] me neither [18:06] asac: no, did not reboot, will try [18:06] pmatulis: do that [18:06] looks too busted to be true ;) [18:09] asac: same problem :( [18:09] asac: ifdown good, ifup throws errors but interface comes up fine [18:09] pmatulis: ok. but NM isnt running? [18:10] asac: NM is not even installed [18:10] if nfs is properly mount after boot it should be ok to try that way [18:12] asac: not sure what you mean by 'ok to try that way' [18:16] pmatulis: when you boot. is all fine? e..g is nfs mounted and system works? [18:17] pmatulis: can you then still reproduce? [18:18] pmatulis: or did i wrongl assume that a nfs unmount/reboot is involved when it happens? [18:19] asac: i believe you mis-assumed, yes. don't believe this is related to NFS [18:19] asac: you want me to double-check? [18:20] pmatulis: so how do you reproduce this bug again? [18:20] just stopping and starting firefox? [18:20] or is there something else involved? [18:20] asac: ok, ok, me thinks we are mixing up bugs [18:20] yeah most likely. [18:21] right now i'm talking about ifup throwing errors, not related to NFS [18:21] pmatulis: right. i am talking about the final goal ;) [18:21] how do you reproduce the firefox issue again? [18:21] other issue still survives: when user home mounted via NFS and cold boot occurs then FF3 is unuseable [18:21] ok ... cold boot [18:22] workaround: mv .mozilla to /tmp (say) and mv back to ~ [18:23] pmatulis: is that 100% reproducible? [18:23] asac: yes [18:24] pmatulis: ok. and you tracked it down to the places.sqlite file right? [18:24] asac: unfortunately not [18:24] pmatulis: oh. i think thats likely [18:25] pmatulis: at least there were other bugs that said that the places.sqlite was the broken file [18:25] asac: it said 'locked' when i issue .tables command but then i tried on working system and gave same msg [18:25] maybe try to just move the places* files away and back into the location [18:25] to narrow down the cause [18:25] will do [18:25] pmatulis: is there anyway to preserve that file in such a "locked" state? [18:26] i have the feeling that tarring up and untarring would also resolve it :( [18:26] asac: heh, yeah, don't know [18:26] pmatulis: i am sorryy, but my day ended already ... and now i have to go [18:26] pmatulis: which timezoe are you in [18:26] ? [18:26] UTC-5? [18:27] asac: EDT (13:23 right now) [18:28] asac: hmph, didn't happen just now, make that 99% reproducible [18:57] asac: appppppppply the patch! works fine here [18:58] got a question about outgoing smtp on thunderbird running under ubuntu. ,cananyone offer assistance? [19:01] question is as follows....... [19:02] i have thunderbird running on a dual boot system windows xp home adn ubuntu 8.07 LTS [19:02] on windows i can send and receive emails perfectly. [19:03] however as of the beginning of september, i can only receive under Linux despite the fad that settings are identical in bothoperating systems. [21:15] asac, [reed]: flash10 is barely usable for me... it sky rockets my cpu and look at this: http://www.sofaraway.org/ubuntu/tmp/flash10.gif [22:11] fta, the same is happening here... [22:11] (regarding flash 10) [22:12] great [22:12] well, sort of [22:12] do you remember when it started ? [22:12] yeah, at least you know you're not alone ;) [22:12] umm... well, it is like that for a while [22:12] me too :( [22:12] I can't remember exactly how long, but I'd say a week or two for sure... [22:13] apprently, the liferea bug was about that too [22:13] Bug 203157 [22:13] Launchpad bug 203157 in liferea "Liferea uses a lot of CPU at times" [Medium,Confirmed] https://launchpad.net/bugs/203157 [22:13] and i have the same with prism [22:13] the commom thing is xul + flash [22:13] you're not using nspluginwrapper? [22:13] no [22:14] yeah... then it's xul or flash... (I'll be evil and blame flash :P) [22:16] restarting X, brb [22:19] youhou, new irc cloak [22:19] yay :) [22:20] i'm no longer unaffiliated ;) [22:20] i hope I'll have a cloak soon ;) [22:26] http://ubuntuforums.org/showthread.php?t=914314 [22:29] hmm http://ubuntuforums.org/showpost.php?p=5754265&postcount=4 [22:30] bug 192888 [22:30] Launchpad bug 192888 in libflashsupport "firefox crashes on flash contents when using libflashsupport" [High,Confirmed] https://launchpad.net/bugs/192888 [22:33] "This bug has 62 duplicates" [22:34] and 261 subscribers, oh my. [22:43] crimsun, you don't use your ppa for your alsa-lib ? [22:45] hm, strange, i already built your branch but i'm not running it [22:46] oh, it's older [23:15] موزيلا فايرفوكس [23:15] it's supposed to mean Mozilla Firefox [23:27] [reed], today's trunk: Program received signal SIGSEGV, Segmentation fault. [23:27] 0xb7754003 in JS_GetClass () from /usr/lib/xulrunner-1.9.1b1pre/libmozjs.so [23:32] fta: looks like quite a generi backtrace ... you have more context?