[02:00] anyone remember when gvfs fuse support was added, was it in Intrepid (gnome 2.24) or earlier? [02:02] hmm looks like it was in hardy as well [02:02] gvfs-fuse has been available since gvfs has been [02:03] ok [02:03] looks like gnome 2.22 was when that happened [03:10] calc: still chasing that one down? [03:10] calc: ever find a debug build of OOo? [03:14] cj: got it working :) [03:14] cj: i just forced gvfs fuse paths in the file dialog so it uses fuse now throughout [03:14] cj: i'll be uploading a new version to jaunty in a few minutes [03:15] woo hoo :) [03:16] you know if there are any job openings on your team? :) Seems I don't have enough hard-core OpenGL experience to get the gtk+/OpenGL position [03:16] * cj is annoyed by the commute [03:16] cj: all i know about is what is on the ubuntu.com careers page [03:17] I guess what I'm asking is whether any of those are on this team :) [03:17] i guy i have known for about 13 years might be coming to work for the oem team soon if his interviews go well :) [03:17] * calc thinks he only has one left now [03:17] cj: oh i am not sure, i'll see if i can figure it out from the descriptions [03:18] ah, don't put too much effort into it. I'll take any position I can hack :) [03:18] http://webapps.ubuntu.com/employment/canonical_GD/ <- is for DX team which isn't the one i am on but works with us [03:18] DX != DirectX, I assume :) [03:18] Desktop Experience? [03:18] cj: if you like to do new design type stuff that would be a good position [03:18] cj: yea [03:19] cj: the team did the new notification work, etc [03:19] my wife is an excellent designer. http://www.cadmiumyellow.com/ [03:19] I like the notification :) Looks good with the transparency and whatnot [03:19] * cj hugs screen copy/paste [03:20] cj: she works for microsoft? [03:20] did for years. I'm on contract there now. [03:20] she's pretty busy with the kids, but she's interviewing for a position in the surface team, I think. [03:21] so, my X is not *excellent* but I've poked at it here and there while doing work with gtk+. How deep do they want the X fu to be? [03:22] cool, surface sounds neat [03:22] not sure probably doesn't have to be that great, gtk definitely would be important though [03:22] Hmmm, haven't done anything professionally in C/C++ since '03 or so... mostly perl and some c# since then... [03:23] I've read it pretty often, but not written anything more complex than a patch here and there [03:23] looks like knowing how to autotools code is wanted too but that is fairly simple [03:23] * cj wrote a series of autotools articles... I have a feeling I'm in the first page of google hits for mono+autotools [03:23] * cj checks [03:23] cj: doesn't hurt to apply for a position perhaps they will like you :) [03:24] awesome. first hit :) [03:24] writing articles on autotools doesn't hurt either ;-) [03:24] ooh, and second and third :) [04:30] new OOo 1:3.0.1-5ubuntu1 uploading now using gvfs fuse instead of gnome-vfs/gio :) [04:40] calc: only using one of two competing systems is always a good thing [04:40] calc: even if you're completely destroying performance :P [04:45] vuntz: how maintained is gnome-menus? libxfce4menu recently lost all dependencies other than glib, gobject, and gtk... :) [04:50] Amaranth: gnome-vfs and gio are both broken in OOo so fuse was the only option, it didn't work out of the box either but was relatively easy to patch in OOo to work [07:19] hello gentleman [07:19] men [08:00] pitti, around? [08:32] Good morning [08:32] crevette: hey [08:32] hello [08:32] crevette: I noticed the bug mail yesterday, but the followup said it wouldn't work? [08:35] sorry for being late; hacked until 2 am... [08:35] pitti: oh, you too ;-) [08:35] seb128: yeah, couldn't stop until my apport launchpadlib test suite would succeed :) [08:35] ;-) [08:35] I just closed IRC in the evening [08:35] dear rosetta, please stop spamming me now! [08:36] gosh, launchpadlib still has quite some quirks to hack around.. [08:38] mvo: hello [08:38] mvo: do you plan to update libgksu in jaunty? [08:39] mvo: the new version fixes "gnome-system-monitor crashes when asking for credential" bug which is open for a while and got lot of duplicates [08:39] and any idea about bug #342043 [08:39] Launchpad bug 342043 in gimp "gimp- install prob" [Undecided,New] https://launchpad.net/bugs/342043 [08:39] "list file for package `gimp-data' contains empty filename" error [08:39] seb128: I asked for a FFe [08:39] let me check the status [08:39] mvo: where? there was no open bug yesterday for that [08:40] mvo: https://edge.launchpad.net/ubuntu/+source/libgksu/+bugs [08:40] https://bugs.edge.launchpad.net/ubuntu/+source/gksu/+bug/339949 [08:40] Ubuntu bug 339949 in gksu "feature freeze exception for 2.0.2" [Undecided,Confirmed] [08:40] looks like its fine [08:40] I will test/prepare/upload [08:41] it also fixes a long and anoying problem that the return code of the sudo ed app is not returend [08:41] mvo: thanks! [08:41] afterwards we can work on the panel indicator thing (maybe when kenvindane comes too) [08:42] mvo: ok, for the fusa thing we use the update-notifier framework no? [08:42] yes [08:42] we could do the same again, but ted was suggesting to just run it automatically [08:43] and not bother asking [08:43] because it does only add something and does nothing else (the fusa one removed the logout applet) [08:43] right, good point [08:43] btw, the change to remove system/logout is confusing :) [08:44] I think so too [08:44] I'm wondering if we should have a gconf key to disable that behaviour [08:44] and set it only on new installs [08:48] that is a good idea [08:48] hm [08:49] how expensive is a gconf call? I was thinking of a simple file assuming its the "cheapest" solution. because the script would run on each login ./ [08:49] if we put it into autostart [08:49] what autostart? [08:49] gconf call are less expensive that a test call in an autostart [08:50] could we write the autostart to the user directory by some way? [08:50] so the tool could delete the autostart after running [08:50] so we would have no cost [08:50] and it would run once [08:55] I don't know how to do that savely [08:55] (think of nfsrootsquash for example) [08:56] we could patch gnome-panel itself to create the applet [08:56] mvo: that would mean extra code diff to carry until next lts [08:56] I don't like that much [08:56] can we just drop the autostart in /etc/xdg/autostart then for upgrades and not new installs? [08:57] yes [08:57] autostart can be conditional on gconf value nowadays [08:57] so we just need to make the script do the change and set the key* [08:58] so we don't even need code checking for this gconf key [08:58] aha, cool [08:58] lut huats [08:59] hello seb128 [08:59] what gconf tree is the autostart config? [08:59] ? [09:00] gnome-session runs the autostart [09:00] so it looks to the user's gconf configuration [09:00] if that's the question [09:00] I'm not sure I parsed that correctly though ;-) [09:00] or you asked how to writte the .desktop line? [09:01] "autostart can be conditional on gconf value nowadays" - what gconf keys :) ? [09:01] whatever you want [09:01] seahorse-daemon.desktop:AutostartCondition=GNOME /apps/seahorse/sharing/sharing_enabled [09:02] that's one example [09:02] which means [09:02] start if /apps/seahorse/sharing/sharing_enabled is true [09:02] cool [09:02] I prepare a update based on this, good stuff [09:02] thanks [09:03] you rock! [09:03] * seb128 hugs mvo [09:07] hmm. could it be that metacity recently got more composition features? [09:08] there is a branch that gives it all sorts of new features based on clutter [09:08] but the stock one has none of that AFAIK [09:08] (well, drop shadows and a new alt-tab switcher) [09:09] mvo: the alt-tab switcher ... when did that land? [09:09] i have composition enabled for the bubbles and it worked well ... until yesterday [09:09] now the alt-tab switcher is really slow [09:09] and firefox gets the same BadWindow problems i have with compiz [09:09] no X update in between ;) ? [09:10] but without compositing in metacity those problems are gone? [09:10] thats something i will now figure [09:11] what gconf key was that? [09:11] found it [09:11] nice! devel/ in alt-f2 opens nautilus in that folder .) [09:11] mvo: yes. firefox start issues are gone [09:11] i couldnt start firefox from the gnome panel at all [09:12] now i can start it again [09:12] bryce_: backout whatever you updated in the last two three days ;) [09:15] asac: there is a new libdrm version uploaded recently you can try downgrading this one [09:15] mvo: that reminds me that alt-f2 is broken on my laptop for a few weeks [09:15] e.g. does nothing [09:15] still works here though [09:15] seb128: thanks will check that out [09:15] asac: known problem with compiz [09:15] asac: keep it like this, so you are my tester [09:17] mvo: ok ;) [09:20] hmm ... seems i get more driver bits from xorg today ... lets see if that fixes it [09:21] asac: what videocard? [09:22] ati [09:23] 5:00.0 VGA compatible controller: ATI Technologies Inc R580 [Radeon X1900] [09:23] with -ati driver [09:23] let me relog in to see if the latest xorg updates helped [09:23] otherwise will try drm downgrade [09:23] morning o/ [09:25] lut didrocks [09:25] i think i think i think ... that the driver updates from today fixed it [09:25] but could be that X just needs to run longer to get into that bogus state again [09:26] alt-tab however still has kind of high latency until it appears [09:26] probably takes a bit to get screens of windows [09:44] anybody on jaunty amd64? === mpt_ is now known as mpt [09:50] yes, why? [09:51] geser: can you run [09:51] pitti, sorry for the delay, I don't know if I can trust only one user. would you have time to test? [09:51] geser: chrpath -l /usr/lib/python-support/python-gconf/python2.5/gtk-2.0/gconf.so [09:51] geser: or 2.6 [09:52] /usr/lib/python-support/python-gconf/python2.5/gtk-2.0/gconf.so: RPATH=/usr/lib [09:52] /usr/lib/python-support/python-gconf/python2.6/gtk-2.0/gconf.so: RPATH=/usr/lib [09:52] geser: thanks [10:10] didrocks: want to do some updates today? [10:12] seb128: I'm afraid, I will not have the time. I leave earlier to take the train [10:12] seb128: if it can wait for Monday, no pb :) [10:12] didrocks: no, therei no hurry, enjoy your weekend! [10:12] seb128: thanks a lot, hope to have a nice weather for skiing :) [10:13] oh, you go skiing [10:13] cool [10:13] seb128: I will be there from Monday evening and ready for new GNOME release ;) [10:13] (in vacations \o/) [10:13] ;-) [10:15] crevette: yes, I'll test it, but can't do right now [10:23] so where can i mount nfs through gnome? [10:23] starred like 5 minutes at preferences/admin menu now [10:23] stared [10:25] "Connect to Server ..." no nfs :( [10:26] last try google: "mount nfs in gnome" [10:27] asac: you can'ty [10:27] asac: nfs is not a standard user thing [10:27] sigh [10:27] so all is lost now [10:27] ? [10:27] no nfs ... no remote mount ... no mail [10:27] no remount mount? no mail? [10:27] seb128: one of the recent updates broke nfs ... i cannot mount my nfs anymore :( [10:28] seb128: well. for ages i mount my maildir folder and use mutt [10:28] i hoped that doing it through gvfs might workaround some issue i do not yet see [10:29] i guess this means i should introduce offlineimap on this system too [10:29] i will ask the server folks now [10:29] asac: gnome bug #536292 for the record and a discussion about nfs in GNOME [10:29] Gnome bug 536292 in hal volume monitor "user-mountable nfs filesystems are not showing up in Computer view of nautilus" [Normal,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=536292 [10:30] seb128: oh. yeah. i even hoped that i could just mount nfs like i can mount ftp, dav, samba et al [10:30] asac: who browse nfs as you browse ftp etc nowadays? [10:30] so now i can share through samba maybe ... which sounds a bit up-side-down ;) [10:30] asac: usually nfs mounts are sysadmin thing [10:31] asac: use ssh? [10:31] seb128: but why is samba? [10:31] a user thing [10:31] seb128: ssh? [10:31] asac: because that what zindozs users use all the time to share anything? [10:31] seb128: you mean open mutt on that system? [10:31] its far too slow ;) [10:31] asac: no, gvfs-mount ssh://server, cd .gvfs/ssh...; mutt [10:32] seb128: sounds like an option. but also sounds slower [10:32] i will come back if nfs really is dead officially [10:32] asac: why would ssh be slower than nfs? [10:32] seb128: because nfs is optimized for remote file access [10:32] and not ssh? [10:32] seb128: i dont think ssh is optimized for that. its just a remote session protocol [10:33] nfs is even on the kernel side [10:33] but i can try and report back [10:33] do you really see a difference to read emails? [10:33] ideally ssh server and run mutt there should be the fastest [10:33] seb128: i will report back. havent tried. but i would think. if i open mutt i have to open all maildir files [10:33] local disk access [10:33] only remote mutt display [10:33] seb128: unfortunately its not faster to run mutt on that system [10:34] than running it on this system using nfs [10:34] weird [10:34] the server is really slow [10:34] remote mail client sucks IMHO [10:34] yeah. the nfs client side cache is responsible for the win [10:34] if you ever have an attachment, you need to jump through hoops [10:34] pitti: yeah. thats another reason [10:34] i will probably migrate to offlineimpa for this system too [10:35] * pitti pats imap [10:35] but i first want to understand why nfs broke after ages [10:35] asac: offlineimap is a 100% pure ♥ [10:35] :) [10:35] pitti: i use offlineimap on my laptop against the same server [10:35] pitti: so its all setup [10:35] but nevertheless, nfs should work, of couse [10:35] its just that i used nfs here for years [10:35] it doesnt work anymore and that scares me [10:35] nfs didn't change for ages [10:35] indeed [10:35] now going to server folks ;) [10:35] weird that it broke [10:35] asac: thanks for catching [10:35] seb128: right. i think its bootup optimization ;) [10:36] offlineimap++ [10:36] blame Scott :) [10:36] pitti: yeah well. its hard to proof if you have no clue ;) [10:36] asac: well, they didn't touch nfs at all, not even changed rc init scripts [10:36] seb128: yes. i browsed my upgrade log ... only found low level stuff liks initscripts or fuse stuff ... but fuse isnt used for nfs i hope [10:36] heh, no [10:37] not yet ;) [10:37] nfsv5 might [10:37] (just kidding) [10:37] hmm [10:37] heh [10:39] i saw some scary modprobe issues on boot this time. output looked like what you get if modprobe is run with wrong parameters (e.g the help output) [10:39] besides from that the boot was really fast [10:40] problem is that it happens before it writes any logs ... so probably the initramfs thing [10:40] is there any log somewhere for that stage? [10:41] * asac reboots into a bloody old kernel [11:11] wow. now it works ;) [11:11] rebooting again [11:18] asac: itz kernel bug? [11:19] i am not yet sure [11:19] i had to change my "lo" iface address [11:19] not sure how that might have changed [11:19] i dont use ifupdown for ages now [11:20] and NM just does ifconfig lo up [11:21] i will go back to this after i caught up with important mail ;) [11:27] pitti: SystemError: W:Failed to fetch http://ddebs.ubuntu.com/dists/jaunty/main/binary-i386/Packages.gz Hash Sum mismatch [11:27] pitti: /me rm lock [11:28] pitti: (that's the recent retracer fail, just to let you know I've looked at it) [11:28] seb128: ugh, if I'd were to buy you a beer for every time you do that, you couldn't work for half a year.. [11:28] seb128: merci [11:28] lol [11:28] * seb128 hugs pitti [11:28] * pitti currently figures out why removing tags suddenly doesn't work any more [11:28] the scary thing about test suites is that they uncover so many bugs.. [11:29] they could break bug opening for a change [11:30] that would give us some time to catch up on the open ones ;-) [11:30] haha === asac_ is now known as asac === sabdfl1 is now known as sabdfl [12:18] hey gnomefreak [12:18] hi kenvandine_wk [12:18] gnomefreak: i hear you live in the apex area.... not far from me :) [12:19] i live about 2 miles from southpoint [12:20] yep im about 2 miles from 751 and 64 intersection ;) [12:20] ah... not far at all :) [13:00] tedg: Ping. Just testing xchat, please ignore. [13:01] tedg: One more. [13:05] tedg*: the panel applet adding stuff is making some progress [13:05] tedg1: indicator goodness for xchat? [13:06] tedg: when you have a few minutes... i have some questions about debugging a problem with the applet [13:10] morning rickspencer3 [13:12] kenvandine_wk: good mornging [13:12] kenvandine_wk: any chance you could me your work irc nick *short* than your non-work one? [13:13] it would save me about 2 key clicks :) [13:13] hehe [13:13] suggestions? [13:13] do you need to be logged twice? [13:14] that breaks tab completion and is pretty annoying I've to say [13:14] i guess i could just join this channel from the one [13:14] i am sitting in front of both of them all day [13:15] but at night just the other one [13:15] you have one work machine and one out_of_work not running ubuntu? ;-) [13:16] both running jaunty [13:16] using them both all day... then carry the laptop down with me in the evening :) [13:58] re [13:58] mvo: you're gnome-panel debdiff looks fine to me, feel free to upload [13:58] seb128: I'm doing some final testing now [13:59] mvo: you can also upload the libgksu update I think since it got an ack === onestone_ is now known as onestone [14:13] seb128: hm, it seems to not run on gnome-session login, how can I set gnome session into debug mode to see what is going on? [14:13] mvo: gnome-session --debug [14:14] mvo: where did you install it and did you desktop-file-valide the .desktop too? [14:19] pitti: I'm not sure to understand what you did on notify-osd [14:19] seb128: desktop-file-validate give me the same as for seahorse [14:19] the desktop file is in /etc/xdg/indicator-applet.desktop [14:20] pitti: you switched from 0.9-0ubuntu1 to 0.9ubuntu2 is that normal? [14:20] mvo: should be /etc/xdg/autostart [14:20] seb128: eh, my bad. there it is [14:20] seb128: might have been a merge glitch; the ppa branch uses native versions AFAIK [14:20] (copy-n-paste error) [14:20] mvo: should be /etc/xdg/autostart/indicate-applet.desktop [14:20] mvo: no problem [14:20] pitti: ok thanks [14:20] seb128: but since we have orig.tar.gz's, we should use normal revisions [14:21] pitti: right, that's what I though I was just checking in case you did the change on purpose to build from bzr rather than using tarballs or something [14:21] seb128: nah, I just suck [14:21] pitti: no you don't! [14:21] * seb128 hugs pitti [14:22] * pitti hugs seb128 [14:22] mvo: it's /etc/xdg/autostart/indicate-applet.desktop? [14:22] mvo: run a non GNOME session and gnome-session --debug [14:22] mvo: maybe the .py is not +x or something [14:24] I wonder if gconf is not available yet when it runs or something [14:25] hm, it runs [14:25] then it hangs [14:26] and then the session says it fails to register [14:26] does it kill it then? [14:26] or will session just ignore it at this point? [14:27] mvo: it means it has run and gnome-session goes to timeout [14:27] mvo: drop the "X-GNOME-Autostart-Phase=Initialization" that's wrong there [14:27] mvo: that means it should be ran early and register [14:28] sorry for overlooking that in the debdiff [14:28] normal autostart should use no such key [14:28] aha, thanks [14:31] mvo: that's using that which will lead to several seconds blanks in the bootchart and Keybuk will hunt you down ;-) [14:32] mvo: btw any reason to just not dput the libgksu from your ppa now? [14:32] seb128: yes, because I'm debugging a gnome-panel indicator applet problem right now ;) [14:32] lol [14:32] mvo: sorry ;-) [14:32] :P [14:32] no problem [14:33] I forgot that most men don't do multi-tasking ;-) [14:33] * seb128 runs [14:33] human multitasking is a myth [14:33] I can only do debugging and tea drinking at the same time, not more ;) [14:33] yeah, eating food and thinking works :) [14:33] but thinking about two different things is a myth IME [14:35] I find it difficult enough to think about one thing already ;) [14:36] * pitti hugs mvo [14:36] * seb128 hugs mvo and pitti [14:37] pitti: notify-osd 0.9.3 uploaded and I verified it fixes the epiphany download crasher testcase [14:37] yay you [14:38] seb128: hm, seems they didn't update to 0.9.3 in the ppa :( [14:38] pitti: I didn't look to their ppa just got it from bzr [14:38] right, I mean the ppa branch [14:38] ah [14:39] lp:~notify-osd-developers/notify-osd/packaging/ [14:39] are they still using that since they got it uploaded to jaunty? [14:39] seb128: nevermind [14:39] well, I guess they do, for their own testing [14:39] I used lp:notify-osd to do the merge [14:39] anyway, so they can merge from us for a change :) [14:39] and updated the packaging over that [14:39] yeah ;-) [14:46] davidbarth: ^ FYI [14:46] davidbarth: i. e. you might want to consider merging lp:~ubuntu-desktop/notify-osd/ubuntu/ into lp:~notify-osd-developers/notify-osd/packaging/ if you still care about the ppa branch [14:47] seb128: thanks for sponsoring alacarte. However, i have received 4 e-mails from "rosetta" stating translations could not be imported. Should/Can I do anything about this or can I safely ignore them? See http://paste.ubuntu.com/130645/ for the mail. [14:47] pitti: ok, will take a look at the changelog [14:47] davidbarth: just a merge to trunk, to get 0.9.3 [14:51] Ampelbein: you can ignore those [14:51] ok, thanks. [14:52] pitti: i will in fact let mirco do that, as he's now officially the maintainer of notify-osd (he loves it!) [14:52] davidbarth: ah, so he can finally do some packaging again :-) [14:52] DEBUILD, I'LL BE BACK! [14:53] heh [14:59] kenvandine_wk: do you think you could test https://edge.launchpad.net/~mvo/+archive/ppa?field.name_filter=panel&field.status_filter=published&field.series_filter=any ? the auto-adding of the indicator applet? [15:01] seb128: is there a way with gnome-session to ensure that the add-indicator thing is run before the panel? it seems the parallel startup confuses the panel sometimes (I just got two workspace switcher on the panel, but just one in gconf) [15:01] mvo: certainly! [15:02] mvo: not easy, you would have to add back the line your dropped and make sure the script register to the session to not create a timeout [15:02] hmmm [15:02] the way it works now is that session components are started first and they register correctly [15:02] so it knows when to start the next one [15:02] and then everything else is started [15:03] that's an efficient way to do thing [15:03] right [15:03] but that doesn't include the usecase where random script need to be ran before a session component [15:03] maybe ensure that the panel is fully loaded and then run it? [15:03] maybe a simple sleep is enough in the script [15:03] add a sleep? [15:03] right ... [15:03] heh :) [15:03]