=== fta_ is now known as fta === max101 is now known as _max_ [01:29] what is the procedure if there is a spec that you want to help, however it seems stalled and you can't get the assignee's contact information? === beuno_ is now known as beuno [02:55] wow. intel driver is a bit trippy now. [03:00] Hobbsee: intrepid ? === hankjin_ is now known as hankjin [03:12] emgent: yes [03:15] Hobbsee: Should I not upgrade today? [03:16] wgrant: it just draws occasoinal artefacts when you open new windows. it's fine to upgrade, apart from that. [03:16] Ah. [03:16] it's not a big deal - just...different [03:16] as in, it's not going to really get in the way [03:18] Ah. [03:20] open virtualbox in intrepid is broken, now i use non-free. [03:22] <[Relic]> Anyone know when the ubuntu will read the temps for the 45nm intel chips? (aka coretemp. from next kernel(.25) to this kernel(.24) and new lm-sensors version(3.0.2)) [03:23] <[Relic]> I am going to have to switch to something that does do that if it doesn't happen soon. === defcon is now known as Guest52148 === Guest52148 is now known as ionstorm [04:58] What is the location of amixer on Ubuntu? [04:58] /usr/bin/amixer? [04:58] yes [05:54] moin === Zic_ is now known as Zic === elky is now known as elkbuntu === torkel_ is now known as torkel === tkamppeter_ is now known as tkamppeter [12:15] hello all, i want to make a package with a configuration file in etc. is there a way to tell the package that that file is taken and if needed, insert a default conf file? [12:18] GuySoft: Better to ask packaging questions on #ubuntu-motu [12:18] persia, thanks, i got an answer at #debian-devel [12:18] GuySoft: Excellent. I'm glad to hear it. [12:19] persia, BTW, its done automatic.. all good for debian [12:22] GuySoft: Yep. :) [13:25] here's a silly question, but i'll ask it anyways..Does anybody know which packages from ubuntu packages I would have to download to build a base, none gnome, none kde, Just no frills base? [13:27] ubuntu-minimal ? [13:27] ubuntu-minimal might be a good place to stary [13:27] s/stary/start/ [13:35] thanks guys you hit the nail on the head :) === Guest71871 is now known as hichamh [14:57] Can anyone help me to find missing seahorse-agent on intrepid? [15:29] slytherin: it got split out from the seahorse into a separate source tarball but isn't packaged yet. IIRC seb128 was working on packaging it. [15:30] geser: So someone is working on it. That is all I wanted to know. [15:53] gasp, last upgrade was bad (intrepid).. after a reboot, X crashed (once), gvfsd-trash crashed, and far worse, my keyboard is f*cked up (no more arrow keys, home/end keys, AltGr acts as Enter, ... in fact, they are all mapped to wrong codes) [16:02] fta: bug 255008 [16:02] Launchpad bug 255008 in xorg-server "Up arrow key mapped to Print [screen]" [High,Confirmed] https://launchpad.net/bugs/255008 [16:09] fta: read it through [16:11] tjaalton, yep thanks. setxkbmap -model evdev -layout fr -variant latin9 -option lv3:ralt_switch fixed it for now [16:13] fta: set the model in the keyboard capplet so it's always evdev [16:15] tjaalton, done [16:22] s/stary/start/ === asac_ is now known as asac [16:34] Hello [16:34] How are Ubuntu applications generally planned out and developed? [16:37] hello? [16:38] How do the Ubuntu developers work? [16:38] That is, what do they focus on? [16:38] https://wiki.ubuntu.com/UbuntuDevelopment [16:38] How do they organize their code, etc? [16:38] jcastro: thanks === cr3_ is now known as cr3 [17:55] heya [18:15] cr3: Hey. Now that was a fast answer :). Do you prefer me to reply on IRC or by mail? [18:15] RainCT: either way is fine :) [18:15] RainCT: both are logged anyways, which equally compensate for my lack of memory [18:15] hey jcastro! how's Argentina treating you? [18:17] RainCT: I'm particularly curious to hear how you might suggest to address the mouse and keyboard tests, I haven't been satisfied with either test for a while now [18:18] cr3: Do you mean the text or the complete test? [18:18] RainCT: the complete test could be better in both cases, for the mouse and the keyboard [18:19] (brb) === LucidFox_ is now known as LucidFox [18:45] I'm working on the moodle package, and the binary-indep rule in the rules file removes some licenses that come in with the source tree. [18:46] I've changed the name of the binary package, and now the build complains about the files not existing. [18:48] here's the error and the rules file bit: http://rafb.net/p/9afeVa67.html === smurfix is now known as smurf [18:49] moquist: you'll have to change the "moodle" in debian/moodle to the new name of the binary package [18:49] RainCT: what if I'm building two binary packages from the same source tree? [18:49] * moquist pastes the control file [18:50] control file: http://rafb.net/p/YgJCoB72.html [18:50] there was formerly one binary package named 'moodle' [18:51] moquist: you'll have to duplicate the rm lines or use a loop then [18:51] huh. OK. [18:51] moquist: I'm not sure if I've understood what you want to achieve, but I don't think that you'r taking the best aproach [18:51] RainCT: I'm wide open to criticisms of this approach. I haven't actually run this by my sponsor yet... [18:52] moquist: to be sure that I understood it, what is the difference between the two packages? [18:52] RainCT: the problem is that the single binary package dependencies are simply broken if both postgresql and mysql deps are in the same line. [18:53] even if the deps are pgsql|mysql, if you happen to have one of the mysql packages installed then you'll get a mix of pgsql & mysql, and it won't work. [18:53] (moddle is in main, btw, so the MAintainer should be Core Dev, not MOTU) [18:53] email addr? [18:54] moquist: Ubuntu Core Developers [18:54] RainCT: thank you [18:55] moquist: (update-maintainer from ubuntu-dev-tools does the Maintainer change automatically for you) [18:55] actually, this is targetted for debian, so I'm not sure what will end up there. but for now it says Ubuntu Core Dev. === RainCT is now known as RainCT_ [18:57] RainCT_: is there a good reason I shouldn't use brace expansion? e.g., 'rm debian/{pkga,pkgb}/rest/of/path.txt' ? [18:57] hm. with -f, I guess. [18:58] moquist: yea that should work [18:58] RainCT_: OK, I'll go that route then. Thanks for all your help. [18:59] moquist: so what's the problem you have with the dependencies exactly? [18:59] moquist: that if you have neither of the databases installed you'll get some packages from both or what? [19:01] cr3: Sorry; I have time now :). For the text, I think just removing the introductory sentence should be enough; about changing the test I couldn't think of anything which I like. [19:02] RainCT: changing the introductory text for which test, both mouse and keyboard, or one of them in particular? [19:02] cr3: both, imho what they say is obvious enough [19:03] RainCT: sure, I'll have a look at it then [19:04] cr3: For the $(...) things, it's not that they had text next to them or something so I don't see why you would want to put them between _( and ) (or whatever you use for the translations). [19:06] moquist: anyway, I'll show you how I'd do it [19:08] moquist: http://rafb.net/p/VillPv83.html [19:08] moquist: (moodle-postgres and moodle-mysql would just be empty packages) [19:09] RainCT: beautiful. that's obviously better; thanks [19:09] moquist: if libdbd-mysql-perl is only necessary for MySQL you can move it to moodle-mysql package, too [19:09] moquist: no problem :) [19:09] * moquist nods [19:11] moquist: it may also be good to add (= ${source:Version}) to the moodle-postres and moodle-mysql dependencies [19:12] moquist: dash doesn't support brace extension [19:12] geser: noooooo! :\ [19:13] moquist: well, now it doesn't matter with the approach I've told you, or? :) [19:15] Yeah. I just like to whine about dash. :) [19:15] hehe [19:17] RainCT: I don't understand the ${source:Version} addition. I can guess what it's doing, but I don't really know. [19:17] RainCT: and is this what you mean? http://rafb.net/p/JFsIFo54.html === Kmos_ is now known as Kmos [19:20] moquist: ${source:Version} will automatically be replaced by the current revision number of the package [19:20] moquist: so adding it you ensure that moodle and moddle-{mysql,postgre} are the same version (that's important in the case that with future version the database dependencies change) [19:21] you want to read the deb-substvars(5) manpage [19:21] RainCT: OK, that's more-or-less what I figured was going on, but ... I think I'll take a look at the man-page. [19:22] moquist: and no, it's like this how you would write them: http://rafb.net/p/uOmKSc90.html [19:23] RainCT: ah! that already makes this much clearer. [19:27] cr3: About the untranslatable strings, "Internet connection fully established" is the only one, I think. The words after "Running test:" can't be translated neither but that may be intended [19:32] cr3: ah, and on the "the following information will be send" page, 4 out of the 5 strings are cut when translated into Catalan [19:32] (as in, the first characters are replaced by "...") [19:33] (uhm.. I thought GTK would be more clever; "Detalls" gets replaced by "...talls", making it even longer then it would be LOL) [19:34] ...heh [19:34] ...h [19:39] RainCT: I just have one debian/rules file right now. Am I going to need a debian/moodle.rules, a debian/moodle-mysql.rules, etc? I'm building now and it hasn't failed yet, but I guess I don't understand why the moodle-* packages won't still have the same problem. [19:41] moquist: No. But if there is a debian/install, debian/links file or something like that you'll have to rename it [19:42] RainCT: to debian/moodle.install and debian/moodle.links? [19:42] moquist: yes [19:42] k; thx [19:44] moquist: about the failure which you got: files for binary package moodle are placed into debian/moodle at build time, and the rules file accesses those files inside there when it does the rm. when you renamed the package, they went to either debian/moodle-postgre or debian/moodle-mysql (you'd need to do more changes than the rm to debian/rules for your approach to work), so the "rm -f debian/moodle/.." didn't work anymore, of course [19:44] * moquist nods [19:44] moquist: (FYI, for the case you've ever uncompressed a .deb, you'll know that there's a control.tar.gz and a data.tar.gz inside it; the data.tar.gz is, afaik, exactly the same as the debian/ directory) [19:45] Anyone know why on latest intrepid, some of my command line programs suddenly can't make connections to the internet (lan connections work)? [19:45] now it's upset b/c it can't find debian/install that I just renamed [19:45] RainCT: Yes. [19:45] e.g. firefox can connect to www.google.com, but "telnet www.google.com 80" never connects [19:46] benc: Anything interesting in strace’s output? [19:46] BenC: do you have a proxy configured in FF? have you tried running tcpdump on your telnet? [19:46] Yeah, tcpdump, too. [19:46] even stranger, apt-get works, but telnet archive.ubuntu.com 80 doesn't :/ [19:46] no proxy [19:46] and no other machines on my network experience this [19:47] hardware failure? [19:47] I seriously doubt hardware could limit _some_ programs [19:48] how can I limit tcpdump to a certain remote host? [19:48] MTU? [19:48] Treenaks: MTU wouldn't be different betweek apt-get and telnet [19:49] *between [19:49] BenC: no, but telnet sends smaller packets [19:49] BenC: and receives [19:49] Treenaks: which should mean it would be more likely to work, not break :) [19:49] git is also broken, and is ssh [19:49] hmm.. _reverse_ path mtu failure.. [19:50] cr3: something more: it may be better to let GTK wrap the text automaticaly instead of forcing that with \n [19:52] BenC: tcpdump -i eth0 host [19:52] 14:52:09.851645 IP cunning.phunnypharm.org.59548 > yo-in-f103.google.com.www: S 3309534473:3309534473(0) win 5840 [19:52] 14:52:12.848519 IP cunning.phunnypharm.org.59548 > yo-in-f103.google.com.www: S 3309534473:3309534473(0) win 5840 [19:52] that's all I'm showing [19:53] repeats like 5 times [19:57] the syn from elinks (which works) looks to be the same [19:57] 14:56:29.489243 IP cunning.phunnypharm.org.34738 > yo-in-f99.google.com.www: S 3098605745:3098605745(0) win 5840 [19:57] I get an immediate ACK from that [19:57] BenC: this really sounds like a proxy [19:57] BenC: everything that works has proxy support [19:58] elmo: if it's a proxy, it's something intrepid installed on my system that I didn't ask for [19:58] BenC: hmm, what's $http_proxy? [19:58] elmo: I checked my system, and proxy settings are disabled... [19:58] elmo: not set [19:59] $ env | grep -i proxy [19:59] bcollins@cunning:~$ [20:00] running as root doesn't help either [20:01] ssh -vvvv shows this as the last part [20:01] debug2: fd 3 setting TCP_NODELAY [20:01] debug2: callback done [20:01] debug2: channel 0: open confirm rwindow 0 rmax 32768 [20:02] it gets through authentication, and stops just before I get a shell === cody-somerville_ is now known as cody-somerville [20:03] hmm...two things happened before this started...I dist-upgraded....and I switched to a bcm4321 card and the closed wl driver [20:03] could the wl driver be this braindead? [20:13] Only one way to find out I guess [20:19] * BenC high fives broadcom on the dumbest thing he's ever seen a driver do [20:20] BenC: wow, seriously? [20:20] elmo: yeah...I switched back to my intel 4965 and everything is working fine === vorian is now known as vorian_ [20:21] BenC: so what's the bug? [20:21] Treenaks: the bug is broadcom needs to fix their driver [20:21] BenC: obviously [20:21] I only wonder if it's a regression with 2.6.26, or if it was like this with 2.6.24 too, but I don't have time to track it down [20:22] Treenaks: If there was code for me to debug, I would track it down further...but me and binary blobs don't play well together [20:38] BenC: ssh works for me on 2.6.24-20-generic with the wl driver [20:39] johanbr: want to see if it breaks with 2.6.26 kernel? [20:40] BenC: Sure, just give me a minute. [20:54] BenC: ssh with 2.6.26-5-generic and wl works too. My card is 0c:00.0 Network controller [0280]: Broadcom Corporation BCM4328 802.11a/b/g/n [14e4:4328] (rev 03) [21:00] johanbr: odd [21:01] johanbr: Are you behind a NAT firewall? [21:01] BenC: MTU issues? [21:01] I wonder if that's part of what is breaking things [21:03] BenC: Yes, there's NAT. All wireless clients here have to go through a cisco vpn. [21:13] Then I'm not sure exactly what in my setup is triggering the problem, but it surely _shouldn't_ happen :) [21:22] BenC: Does it matter if you ssh inside or outside your lan? [21:23] johanbr: everything inside the lan worked...it was external connections that failed with _some_ programs [21:23] weird [21:23] elinks/firefox => worked....telnet to port 80 (to same hosts) didn't [21:23] ssh didn't work [21:24] pidgin works, apt-get works [21:24] and ssh connected, but halted just before dropping to a shell [21:24] telnet wouldn't even connect...even tho the syn packet looked just like the one from firefox/elinks [21:26] Very strange. I'll try again when I get home - there I have an ordinary NAT setup for the wireless. [21:28] I'm just working behind a wrt54g with dd-wrt firmware [21:28] never had a problem like this, so it's really strange === RainCT_ is now known as RainCT [22:35] i was quite curious to hear about alfresco being shipped in the Canonical repo, does anyone happen to know of an ETA on that? I'd be very interested to try that application [22:35] I'll try and install it "the linux way" but the procedure looks pretty fierce [23:37] BenC: I'm home now. ssh through NAT on 2.6.26-5-generic hangs for me too. [23:42] sweet, now I don't feel so weird [23:42] johanbr_: thanks for testing [23:42] No problem. Let me just reboot and see if it happens on -24. [23:50] BenC: Same thing on -24. And if I route the ssh traffic through a VPN, it works fine. [23:57] johanbr: ah, so not a regression [23:57] so can't blame it on the kernel [23:57] johanbr: what wireless ap are you using?