[00:01] \sh_away: inkscape 0.46 is already in hardy [00:19] oi we have bug with ubuntu gutsy and local time in australia in in nsw [00:19] we need the time to move forwards one hour -- ntp is not providing the proper time [00:20] this is a major problem for security etc. and needs to be worked on as soon as possible [00:20] StevenK: ^^^ How's your time now? [00:20] 11:20 instead of 10:20 [00:21] i will to go tzdata now and see if i can fix it [00:21] http://www.timeanddate.com/worldclock/city.html?n=240 [00:21] dbmood1: Timezones won't change for another week. [00:21] It changed this year. It's the first week in April. [00:21] http://www.abc.net.au/news/stories/2008/03/30/2202804.htm [00:22] it needs to be moved -- forwards and then back [00:22] Fujitsu: yes you are normal correct it was changed [00:22] We should still be +11. [00:22] Is it not 0021 UTC at the moment? [00:23] well, UTC doesn't change, does it? [00:23] let me try apply utc [00:23] It doesn't, but my system says it's 1123 local time. [00:23] its 00:23:31 according to my altered clock [00:23] as in put back for daylight savings [00:24] All of my systems are correctly still +11. [00:24] THey weren't meant to switch back, and didn't. [00:24] I checked this some weeks ago. [00:24] what ? [00:24] dbmood1: I saw a new tzdata coming in recently (maybe today, not too sure). do you have that installed already? [00:24] (for gutsy) [00:24] i'm using ntpd to update too... [00:25] is that a problem ? [00:25] We're not meant to switch from +11 to +10 for another week. My Dapper, Gutsy, Hardy systems haven't switched yet. [00:25] well i have just switched [00:25] dbmood1: Which version of Ubuntu are you running? [00:25] Fujitsu: lucky you... I'm loosing an hour tonight already :( [00:25] 7.10 here i can check hardy [00:25] dpkg -l tzdata [00:26] Desired=Unknown/Install/Remove/Purge/Hold [00:26] | Status=Not/Installed/Config-f/Unpacked/Failed-cfg/Half-inst/t-aWait/T-pend [00:26] |/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad) [00:26] ||/ Name Version Description [00:26] +++-==============-==============-============================================ [00:26] ii tzdata 2007f-3ubuntu1 time zone and daylight-saving time data [00:26] i might have my updates screwy ? [00:27] Ooh, damn. [00:27] You're right. [00:28] is it just me ? [00:28] or ... [00:28] 2007f-3ubuntu1 is too old, I think. [00:28] * Fujitsu looks harder. [00:28] rofl oh cool [00:28] I'm thoroughly confused. [00:28] thank you [00:28] sorry [00:28] seems all is fine [00:28] but iinet my isp repos are slightly outdated [00:28] or something [00:29] oh wait .... was the tz.. thing a recommended update ? because i don't have those ticked [00:29] Let's just get rid of the stupid DST altogether. :-) [00:29] Oh what. [00:29] WHy the heck did this only go into -updates? [00:29] let me tick recommended and see what happens [00:29] dbmood1: That's the problem. [00:30] its a security bug it should be in security too tho [00:30] The update has been in *-updates since 2008/03/07. [00:30] i don't imagine many people to have recommended ticked [00:30] But not in -security. [00:30] Most people should. [00:30] This sort of update should really go into -security, I think... [00:30] is it default ? [00:30] As it's utterly critical. [00:30] I believe it is on by default. [00:30] hopefully [00:31] -- what about for lts and the server releases -- yeah that fixed it [00:35] how is a tzdata bug a security bug? [00:35] well if you have encryption that is based on time etc. / you get time based attacks [00:36] because you're an hour off from everyone else? [00:36] ........a whole hour [00:36] yes [00:40] got an end user to confirm (hopefully didn't touch it) that she doesnt' have the correct time .... [00:40] no wait [00:40] she got the update [00:40] well, there's a hardy tz bug [00:42] well i will know the full answer when she returns... [00:43] doko: got any pointers for pycentral usage with a package that has SWIG python bindings? I seem to be missing some critical step -- everything gets built and installed but python can't seem to find it... [00:44] but my debian box has the correct time it seems and this is an etch box... zdata 2007j-1etch1 Time Zone and Daylight Saving Time Data Australia/Sydney Sat Apr 5 15:59:59 2008 UTC = Sun Apr 6 02:59:59 2008 EST isdst=1 gmtoff=39600 [00:44] Australia/Sydney Sat Apr 5 16:00:00 2008 UTC = Sun Apr 6 02:00:00 2008 EST isdst=0 gmtoff=36000 [00:44] sorry to be so spamish i will be quiet except to tell you the out come of the lady's update [00:46] The change was made in 2007g. [00:46] pwnguin: hrm? what tz bugs are there in hardy? [00:46] pwnguin: more precisely, why is there no bug report about this that's importance: high and milestoned for the release? :) [00:46] So anybody who has updated from -updates since 2007/10/20 will have the changes. [00:47] If tzdata updates are only pushed to -updates, it shouldn't be `Recommended updates', it should be `Critical (unless you don't care about even slight accuracy of time) updates' [00:48] slangasek: they're actually bugs in gnome-clock-applet [00:48] in this case it is an hour off and they - end users who might not have it ticked will have it off - some admins might not want to allow the recommended -- or know about the change yet -- they probably have them off [00:49] slangasek: apparently it doesn [00:49] pwnguin: oh, I guess you mean the fact that gnome-panel guesses your timezone wrong === dbmood1 is now known as dbmoodb [00:50] yes. just mentioning it in case the above "end user" was merely judging from hardy, etc [00:53] apparently if you do not turn ntp on the time even with the correct tzdata will not fix the problem immediately (end user result) -- hope you guys can start looking into this --- seems easy to fix and blah (package seems to be fine -- waiting for sometime i guess) [00:55] i installed the beta,, everthing went well but grub doesn't like the partition [00:55] mttr: #ubuntu+1. [00:55] np === tkamppeter_ is now known as tkamppeter [02:27] <_MMA_> gah.. After new updates GDM sound nor logout sounds work. :( [02:28] _MMA_: and you're sad? [02:30] <_MMA_> Well a little more than sad because this has been something that has broke for one reason or another for 2 releases now. But I don't wanna be a di*k about it yet. [02:32] dirk? [02:32] <_MMA_> dick [02:32] Ah, right. I thought you mean dink. [02:32] <_MMA_> If ya wanna be a smart-ass. ;) [02:35] <_MMA_> Oh well. I'll work on it Monday. === tkamppeter_ is now known as tkamppeter [03:04] there will be a short outage to the wiki and bazaar.launchpad.net, to hoepfully address the bazaar.launchpad.net performance problems === kitterma is now known as ScottK2 === Igorot_ is now known as Igorot [06:05] hey === macd_ is now known as macd === temuge1 is now known as temugen [08:15] good morning === Seveaz is now known as Seveas === toresbe_ is now known as toresbe === asac_ is now known as asac === giskard_ is now known as giskard === doko_ is now known as doko === jsgotangco is now known as greeneggsnospam [11:17] ah yes, australian time confusion [11:18] we got 11 hours gap or 9 now? [11:18] you fell back and we went fortward [11:18] @now sydney [11:18] Current time in Australia/Sydney: March 30 2008, 20:18:58 - Next meeting: Server Team in 3 days [11:18] so 9 [11:19] no [11:19] 8 for me [11:19] 9 for you [11:19] Seveas: that's wrong. [11:19] Seveas: it's currently 9.18pm [11:19] ah [11:19] Hobbsee, you don't change for two weeks iirc [11:19] Hobbsee, hmm, odd [11:19] ompaul: yeah. something like that. i only noticed when the pin clock for work was showing the wrong time. [11:19] Seveas, they are auzzies ;-) [11:20] * Hobbsee beats ompaul before he can run away [11:20] * ompaul buys Hobbsee a cup of tea and tells her to work on getting her country to change time in sync with the rest of the planet (or close enough that it don't matter) [11:20] there's no python-tz update for dapper [11:21] fun :/ === thekorn_ is now known as thekorn === toresbe_ is now known as toresbe [11:36] ompaul: so every time some government nutter up here decides to extend DST by another few weeks, Australia should get a few weeks less DST? :) [11:38] sounds great [11:39] * persia discourages synchronisation. Conference calls are annoying when there is more than one time transition happening the same week [11:40] persia: this is why we all deal in UTC. [11:41] this must mean that the UK/AU time crossover gets significantly worse again, though [11:41] Hobbsee: True, although I've recently discovered that some people confuse UTC and British Time [11:41] yeah, true [11:43] I find it most unfortunate that those without -updates enabled are doomed to have incorrect times. [11:43] I think anyone living in a country that changes the clocks is doomed to have incorrect times once in a while. [11:44] If you have -updates, you're safe. [11:47] (excepting at least Argentina) [11:50] Well, if we're *really* quick... [11:51] heh [12:38] slangasek, na there is normally a two week lag between .eu and .au [12:38] slangasek, after that you factor in the nutters ;-) [12:38] ompaul, that's not just with DST [12:40] Actually, NSW comes first. Except they've changed it so that all of .au changes DST (the states that use it) at the same time [12:41] StevenK: Not only the same day, but the same time? [12:43] persia: Well, I mean the same day. It used to be NSW dropped DST today, and the rest followed a week later. Now we all have to wait until next week [12:43] (And well done Telstra for having your time sources on the phone network jump back an hour today) [12:44] lol [12:44] Ah. That makes more sense. I thought it was something like 4am in the east, 3am in the middle, and 1am in the west [12:45] Actually, 4am in the east, 3:30am in the middle and 1am in the west [12:45] Err. Never mind the bit about the west [12:45] Huh? I thought there was a timezone between Melbourne and Adelaide. Time to study the time map again [12:46] There is. [12:46] Melbourne is +1100, and Adelaide is +1030 [12:46] Perth is +0800 [12:47] (After next week, it's +1000, +0930, +0800) [12:48] Ah. And Victoria and NSW are the same. Right. And summer time expires in West Australia in 2009. [12:50] persia: Yes, Victoria and NSW are the same in terms of DST. Queensland is not. [12:51] Ah. That was the source of confusion. Thank you. [13:38] hello [13:55] lamont: look at what? [13:56] lamont: sorry, tab error; unping === fta_ is now known as fta === evalles is now known as effie_jayx [14:58] hi dendrobates :) [16:51] calc: KDE 4 uses shared-mime-info, KDE 3 not [18:24] why does glib2.0 have such an odd version number? it's less than the unstable version it's based off of [18:24] it really should be named 2ubuntu1 not ~hardy1 [18:27] I was wondering about that too [18:38] Riddell: is there a way for me to query what kde 3 uses for particular files to see what i need [18:46] jdong: 209308 [18:46] bug 209308 [18:46] Launchpad bug 209308 in glib2.0 "glib2.0 version is wrong" [Undecided,Confirmed] https://launchpad.net/bugs/209308 [18:47] cool [18:54] doko: how is the progress on the i18n stuff? [18:54] looking ... [18:55] doko: i'm testing my first full build right now, so it might be ready for upload tomorrow [18:55] yes, second build finished, I'll let you know tonight [18:55] doko: great :) [19:11] Does anyone here work on wubi? [19:13] the cairo patch fixes OOo font display issues :) [19:14] calc: which patch? [19:14] calc: so I have to scroll back 24hrs in my IRC logs to understand the context of that statement? :D [19:15] jdong: no, if you notice OOo fonts have looked wrong for like forever, they will be fixed probably tomorrow (when i manage to upload) [19:15] it was using cairo/libfreetype incorrectly :) [19:15] calc: awesome! and yes, I have noticed that [19:16] i don't remember which bug report had the patch there were many related bug reports about OOo fonts looking bad [19:18] it appears it doesn't do subpixel smoothing but works for the other settings [19:21] jdong: yea i see you were one of the people commenting on the oldest of the fonts bugs [19:22] I vaguely recall filing a sarcastic bug or two on that when I felt in a particularly bad mood :D [19:22] jdong: heh :)( [19:22] er :) [19:22] * calc has fat fingers [19:24] calc: will impress support 3D effects too? [19:25] tjaalton: not sure i haven't looked at that yet [19:25] tjaalton: probably if it is listed as a 2.4 feature upstream [19:26] calc: thank you!! (font bugs with OO) [19:26] it's probably a config option or something, the current version in Hardy doesn't support it [19:26] * milli hates ugly fonts [19:26] so far the font bugs look to be fixed EXCEPT for the subpixel stuff, but that is a lot better than it was [19:26] I mean build config option.. [19:26] tjaalton: hmm ok [19:27] i'll need to dig to see if i can find a patch to fix the subpixel part also [19:32] erm [19:32] it looks like it is working now, wtf [19:33] yea its definitely working, i thought it wasn't earlier, so now i am confuse [19:33] confused [19:34] ok yea its consistently working now, so i am not sure what happened earlier [19:34] one minor issue is you have to restart OOo to get the font setting to change, you can't change it while running [19:51] calc: --enable-ogltrans [19:51] tjaalton: ah [19:51] calc: don't know what it depends on [19:51] let me see if it isn't getting enabled for some reason [19:52] calc: it was added after rc2 :) [19:53] it appears to be attempted to be used via [19:53] --enable-opengl [19:53] so that might be a bug then [19:53] ok [19:54] well its still not there in the debian version as of 2.4 final [19:54] i'll look into it and see if the arg needs to be changed or added in addition to the other one [19:56] cool, thanks [20:02] calc: kde 3 uses /usr/share/mime [20:03] calc: /usr/share/mimelnk/ rather [20:18] Riddell: does it just do matches based off of file extension? or is there a way to determine what mimetype it would consider a particular given file to be? [20:19] Riddell: i'm wanting to make sure that i list all the needed mimetypes for KDE in OOo to open the files correctly [20:20] Riddell: i already have it for all the shared-mime-info using gvfs-info to verify [20:20] Riddell: so i think that will probably already work correctly for KDE4 (haven't tested) [20:20] calc: kde 3 also does some magic detection, using kdelibs-3.5.9/kio/magic [20:21] Riddell: is there any kde util that i can use to have it spit out the mimetype? [20:22] calc :) [20:25] calc: don't think so, probably not hard to write on in python [20:38] calc: http://kubuntu.org/~jriddell/tmp/mimetype2.py === jtt is now known as jtholmes === jtholmes is now known as jt2 === jt2 is now known as jtx === jtx is now known as jtisme === hunger_t is now known as hunger === jtt is now known as jtisme2 [21:19] Riddell: thanks! === jtt is now known as jtisme [21:58] Riddell: still here? [21:58] Riddell: i tried running that script and its bombing out [21:58] Riddell: tells me list index out of range [22:00] Riddell: i can paste it to pastebin if you are around, not sure why it is saying that [22:01] http://pastebin.ubuntu.com/6234/ [22:01] i ran it as ./mimetype2.py filename [22:02] ah it worked when i ran it as python mimetype2.py filename [22:02] i'm not quite sure why but it worked [22:15] i am confused, this script is telling me kde thinks a excel file is type application/msword [22:15] but in dolphin it says its an excel file [22:16] is dolphin using shared-mime-info or kde's mime stuff [22:18] if this is working right then kde's mime support is crummy [22:18] it claims xml file is text/plain [22:19] is there a bug open on the AEST/AEDT tzdata error ? [22:19] * calc thinks he'll let a kde person determine if the files are opening correctly, since it seems kde mime support is very weird [22:20] borderline broken it would seem [22:25] calc: http://kubuntu.org/~jriddell/tmp/mimetype2.py updated [22:27] Riddell: ok testing that [22:29] Riddell: works much better! :) [22:29] now it thinks excel files are... excel files ;-) [22:30] yipee it looks like it is working much better all around :) [22:30] * calc can now make sure all these mimetypes are in the files as well [22:31] lifeless: Which error? The updates have been in -updates since late October. [22:31] But I'm not sure what you're expected to do if you only use -security. [22:32] Fujitsu: I'm running hardy and its wrong [22:32] Fujitsu: by which I mean, the NSW senate changed the daylight saving rules [22:32] lifeless: Ermm. [22:32] I'm in Vic, I know of the change. [22:32] dpkg -l tzdata [22:32] well its still a bit buggy but i wouldn't be surprised if it was actually real bugs at this point since most of the types are showing up right [22:33] Fujitsu: 2008b-1ubuntu1 :P [22:33] Ah, I'm only on 2008a... [22:33] cat /etc/timezone [22:33] I bet it's User defined. [22:33] no such file [22:34] Um. [22:34] Anyway, I have to head off to uni now... but tzdata in all releases has been updated with the right dates. [22:34] Another ubuntu user I know of has the same symptoms of an early change back [22:35] bye [22:35] lifeless: My (Hardy) clock is correctly showing DST. [22:36] RAOF: thanks [22:36] my gutsy clock handles DST correctly too, although I use -proposed, -updates. [22:37] lifeless: you're using the gnome clock applet? [22:37] slangasek: well two things got it wrong: my server, and my gnome clock applet [22:37] the server I can tell via irssi :> [22:37] lifeless: click, Locations, Edit, verify that the timezone automatically populated for your city isn't wrong [22:38] server is 7.04; probably just haven't updated in an age [22:38] I think this will be a great excuse for not turning in my paper on time tonight! [22:38] slangasek: tiemzone "australia/sydney" with caps; looks right to me [22:39] slangasek: and 'date' in a console has the same time [22:39] hmm, ok [22:39] so what time is it supposed to be there right now? [22:39] did feisty get the tz updates? [22:39] its supposed to be 0838 [22:40] lifeless: that's what I see on the console with current hardy when I type TZ=Australia/Sydney date [22:40] Wow. A 1 minute offset? [22:40] :p [22:40] soren: race condition on the minute boundary :-} [22:40] lifeless: so I don't think the problem lies with the tzdata package [22:41] slangasek: well, thats quite bizaare then. [22:41] Since it was :39:58, the boundary was quite wide. :-) [22:41] whats current utc ? [22:41] lifeless: hwclock in UTC? does it display the right time? :) [22:41] 21:40 [22:41] Sun Mar 30 21:41:59 UTC 2008 [22:42] TZ=utc hwclock --show --utc [22:42] Sun 30 Mar 2008 21:43:07 UTC -0.117192 seconds [22:42] so, other than fucking around with kerberos, the hwclock seems ok. [22:42] * slangasek grumbles and adjusts his clock by hand, since repeatedly setting it via the clock panel causes it to drift backwards :P [22:42] yet another bug I should file on that damn thing... [22:43] TZ=Australia/Sydney date [22:43] Mon Mar 31 08:43:39 EST 2008 [22:43] so my current time zone is apparently not Australia/Sydney [22:44] score [22:44] blame bug #185190 [22:44] Launchpad bug 185190 in gnome-panel "Clock applet chooses wrong timezone for many cities (eg Pittsburgh, Beijing)" [Low,Triaged] https://launchpad.net/bugs/185190 [22:44] even if it's not the bug at issue ;) [22:45] ~$ echo $TZ [22:45] [22:45] ls -l /etc/localtime? [22:45] (well, "readlink /etc/localtime") [22:45] not a symlink ! [22:45] and I'll bet that my friends bug is the same [22:45] and you said you don't have an /etc/timezone, so you get to try to guess which file under /usr/share/zoneinfo it corresponds to, enjoy [22:46] welcome the world of precision timekeeping [22:46] s/the/to &/ [22:46] running tzselect doesn't make it a symlink eithe [22:46] well, I bet I know what file it is [22:46] its the *old* Australia/Sydney file. [22:46] oh, heh [22:47] in fact, wtf is tzselect meant to do these days [22:47] well, the reason it's not a symlink is that it screws things up if /usr is on a separate partition at boot [22:47] cause it sure as hell ain't writing to /etc/localtime [22:47] but /etc/timezone is supposed to be used to track this [22:47] nor to /etc/localtime [22:48] sorry, not to /etc/timezone [22:48] and obviously we need to be updating /etc/localtime when the package updates [22:48] lifeless: dpkg-reconfigure tzdata? :/ [22:48] yes just found via the tzselect man page [22:48] :) [22:49] and it too writes a static file [22:49] yup [22:49] you want to file the bug? [22:52] no, I want you to file the bug and me to fix the bug ;) [22:52] blah [22:53] lp account name ? [22:53] vorlon [22:54] mailed [22:54] we're going to have to get a gnome-panel task for this well, since the gnome panel is not interacting appropriately with the tzdata-controlled files [22:54] thanks [22:55] breakfast, then to see about unfucking the wiki & b.l.n [22:57] lifeless: looking at the postinst, the failure here is because /etc/timezone didn't exist; I'm guessing you don't have any overt knowledge of removing this file yourself... :) [22:57] (in which case I need to track down why it wasn't created on upgrade) [22:58] I'm glad I poked around a bit first :) [22:58] and no [23:06] slangasek, which upgrade should have created it? I'm missing the file as well but my hardy is slightly outdated (a week or so). Might help tracking it down [23:08] Seveas: an upgrade long ago, I think. There's mention of a preinst having been added in Aug 2007, but that preinst doesn't exist now; I'm going to chase this with the Debian glibc maintainers [23:17] lifeless: what was this machine originally installed as? [23:18] -EDUNNO [23:19] feisty or gutsy 64 bit; then I copied various things from my 32-bit laptop (like exim config files) [23:19] probably the fister [23:21] great i found all the mimetypes i needed for kde :) [23:22] lifeless: do you happen to remember if it was liveCD or alternate? It's conceivable that this was a ubiquity bug [23:26] slangasek, mine (2 of them) were desktop installs [23:27] Seveas: installed from which release? [23:28] * slangasek needs to know which CD to start downloading :) [23:28] that's what I'm trying to find out, edgy or later for sure [23:28] one of them ubiquity 1.2.5 [23:29] that's edgy [23:29] another 1.4.11 [23:29] feisty [23:29] fun fun [23:30] uhm cacti FTBFS, and patch is ready. If someone can please see bug #194687 [23:30] Launchpad bug 194687 in cacti "cacti web frontend fails with 'Invalid PHP_SELF Path' after upgrade" [Medium,In progress] https://launchpad.net/bugs/194687 [23:40] ScottK: ping [23:41] emgent: Pong [23:42] about bug #194190 [23:42] Launchpad bug 194190 in cacti "Please sync cacti 0.8.7b-1 (universe) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/194190 [23:43] i saw now that debian version bump fix all. [23:43] i go to remove ubuntu patch and i will attach debdiff. [23:43] slangasek: no idea [23:43] okie-day