[00:09] ok, I found a similar manual (different revision it seems) on the net, and it crashes evince as well; added to LP [00:09] thanks for your help! === Pumpernickle is now known as Pumpernickel === rob1 is now known as rob [02:38] can somebody explain what's the use case for adding an entry for 127.0.1.1 in /etc/hosts ? [02:39] some apps use it? [02:39] its only purpose I could find seems to be to slow everything down considerably? ;) [02:39] what apps use it? [02:39] I'm not sure, but I've run across a couple [02:40] booting goes about 2x faster when I comment it out, and some apps start up 10x faster without it... [02:40] its a loopback address that points to your own local computer. as mentioned, some apps rely on it. i would not recommend commenting it out [02:41] eh, I think that 127.0.0.1 is perfectly fine for that? [02:41] some apps use 127.0.1.1 for loopback [02:42] I don't know if those apps could be told to use 127.0.0.1 or if there's a reason to separate them [02:43] 127.0.1.1 has "hostname.local" after it, which causes delays because of DNS-SD timouts or something [02:43] localhost.localdomain [02:43] shouldn't conflict with hostname.local [02:43] no, it has hostname.local after it... [02:44] mine hasn't [02:44] so, that would be an installer bug? [02:45] 127.0.0.1 localhost.localdomain localhost [02:45] I'm talking about 127.0.1.1 [02:45] which is added to /etc/hosts by the installer [02:45] ah. [02:46] must have changed that on mine :-P [02:46] I could understand if there was a good reason for that, but i couldn't find one [02:46] mine just has the name of my computer [02:46] I have no domains ;-) [02:46] Nafallo: I commented it out too ;) [02:47] I don't think I've touched it. [02:49] hm, could this be Avahi- or DHCP-related? [02:49] In Debian, new installs now have two 127.x.x.x IPs in /etc/hosts. [02:49] minghua: yes, that's what I see too [02:50] one is 127.0.0.1, mapping to your hostname (and FQDN if you have one). [02:50] one is 127.0.1.1, mapping to localhost. [02:50] ehrm [02:50] we have the reverse [02:50] 127.0.0.1 localhost hostname [02:50] It may be the same in Ubuntu. [02:50] #127.0.1.1 hostname.local [02:50] AFAIK [02:50] taht's what I have now [02:50] the "#" was added by me ;) [02:51] I have: [02:51] 127.0.0.1 localhost [02:51] 127.0.1.1 photon [02:51] Nafallo: Maybe it's the reverse, I am not sure on that. [02:51] minghua: I can't figure a reason we would change it, and for Debian for doing wrong :-) [02:52] JanC: Your slow boot/start time problem is apparently your own. I don't think .local is a valid domain name. [02:52] if you read the info in 'man hosts', it sounds like /etc/hosts is not widely relied on anymore. but is still needed in certain cases [02:53] minghua: I didn't put it there ;) [02:53] Nafallo: I don't think we changed it either, but I only know the Debian part of the story, and this is an Ubuntu channel, so better be safe. :-) [02:53] JanC: Your answer in the installation process put it there, I suppose. [02:53] minghua: 127.0.0.1 is localhost, has always been etc.. :-) [02:53] minghua: Ubutnu doesn't ask such things ;) [02:53] I'm not familiar with the ubiquity installer to be sure, though. [02:54] and .local is the default extension for DNS-SD (aka Avahi / bonjour /etc.) [02:54] minghua: in 50 to 70 years we might have only ::1 though ;-) [02:54] Nafallo: I'll be dead by then, so I don't care. :-) [02:54] I won't :-) [02:56] Nafallo: only 1 (consumer) ISP in Belgium offers IPv6 tunnels (and they are mostly business oriented anyway), all others think IPv6 is something for the next century or something :-/ [02:57] JanC: tunnels? what about proper IPv6 then? [02:57] one problem is that a lot of ISP-size router equipment doesn't support IPv6 correctly, if I understand correctly [02:58] also, isn't the educational sector running on IPv6 already? [02:58] e.g. Motorola (headend) & Cisco stuff [02:58] yeah, Belnet has IPv6 [02:58] but consumers can't use that :P [02:58] that depends on who they know ;-) [02:59] Belnet has an IPv6 tunnel "entrance" [02:59] that you can use for free AFAIK [02:59] SixXS? [03:00] yeah, I think they work with SixXS [03:01] my main transit provider is also a SixXS PoP ;-) [03:01] http://www.sixxs.net/pops/belnet/ [03:01] goscomb :-) [03:02] I currently use SixXS, but my ISP is going to offer free native IPv6 soon. \o/ [03:02] ion_: hehe. I will take a BGP-session with v6 ;-) [03:03] I should try to set up an IPv6 tunnel one day, but I think it's quite a lot of work for something that should just work ;) [03:04] and I would have to buy a new router too [03:04] it's not a lot of work. [03:04] it's dead easy. [03:04] last time I checked (some years ago) you had to go through RIPE etc.? [03:05] the registration process has nothing to do with setting it up, and RIPE isn't the case anymore anyway. [03:05] ah? [03:05] JanC: I use easynet pop for years and it works great :) [03:05] well, registration *has* to do with IMO :P [03:06] JanC: SixXS has there own registry. [03:06] hm, I'll look again then ツ [03:06] I rather peer and have my own /48 to announce anyway :-) [03:06] tsu [03:07] hm, their FAQ still mentions RIPE etc. ? [03:07] what does it say about it? [03:08] JanC: you can use a ripe account [03:08] too [03:08] but as said they have their own registry [03:08] bigon: right, but I don't have one ;) [03:08] and they sync the ripe into that one ;-) [03:08] https://noc.sixxs.net/faq/account/?faq=10steps [03:08] the FAQ points to a form on ripe.net [03:09] https://noc.sixxs.net/signup/create/ [03:09] so it's probably outdated :P [03:09] JanC: enter real informations because they verify them [03:09] bigon: that's no issue for me [03:10] I have real data on my domains to [03:10] (DNS.be also checks it now) === asac_ is now known as asac [04:29] geser: sylpheed given back [05:04] Hobbsee: do you have any insight into the i386 build of Hardy's pulseaudio 0.9.8-1ubuntu2? [05:24] crimsun: what about it? [05:25] crimsun: i think parts of launchpad are lying to you, it appears to be pending build [05:25] Hobbsee: I can't seem to figure out anything about the i386 build. All the other arches have built and are available in pool. [05:26] wait [05:26] launchpad is screwing up somewhere [05:27] crimsun: "launchpad is on crack' [05:27] crimsun: wait till monday, i guess [05:27] Hobbsee: ok, thanks. [05:28] crimsun: i can't cancel the build, etc [05:28] maybe i can force it though [05:28] "... bang!" Ooops [05:29] nah, i can't even force it [05:29] LP is reporting conflicting things [05:30] So will require a wizard to make sense of it? [05:30] it'll require a cprov to fix it, yes [05:30] i'd guessed queue-builder or something has fallen over, partway thru. *shrug* [06:34] Anyone having issues with sun-java6-plugin in hardy? (icedtea-java7-plugin doesn't work either) [08:02] where can I find xorg.conf ...in hardy? [08:02] lol [08:02] MuNzE_: It's gone. [08:03] i try to find but nothing [08:03] /etc/X11 ...?? [08:06] MuNzE_: if you miss it, you can just run sudo dexconf to generate it [08:06] MuNzE_: ideally you should no longer need it though [08:06] bryce, i need for compiz === RAOF_ is now known as RAOF [08:15] when did xorg.conf die? i missed this [08:16] lol [08:17] pwnguin: New in hardy. [08:17] as of? [08:18] as of the first X upload to hardy [08:19] pwnguin: 1:7.3+4ubuntu1 [08:19] I only have an InputDevice section for my keyboard (so that i can set Xkb* options) and a Device section for my GPU (so that i can use the proprietary :-( nvidia driver). Everything else is handled automatically by xorg. [08:22] well [08:22] im sure my tablet still needs a section [10:28] * Hobbsee waves [10:39] hi Hobbsee [10:39] heya mdke! how's it going? [10:39] Hobbsee: ok thanks. Busy. What's all this about you using Ubuntu? [10:39] the rumours are everywhere [10:40] mdke: i'm afraid that the rumours are true. [10:40] although compiz and my video card are not in love, for hardy. [10:40] extraordinary [10:40] permanent move? [10:40] so far [10:40] mdke: i didn't know it was that slow a news day :P [10:40] Just wait, she'll start bad-mouthing KDE, and then we'll know we have her... [10:41] heh [10:41] welcome to the light [10:41] StevenK: like i'd be stupid enough to comment on things that i don't like about KDE, or kubuntu for that matter, in a public, logged channel :) [10:41] Heh [10:42] * Hobbsee just wishes hardy would stop freezing, and behave. [10:43] at least you can run a free operating system. I've been stuck on windows for several weeks now [10:43] ew [10:43] makes it very hard to get anything done [10:43] i'll bet. [11:08] Hobbsee, slangasek - could you have a look at bug 174886, it's not something ubuntu-doc can take care of, I think [11:08] Launchpad bug 174886 in ubuntu-docs "Hardy Alpha Release Notes have no links to daily builds" [Undecided,New] https://launchpad.net/bugs/174886 [11:09] mdke: looks like that should be assigned to ubuntu-release [11:09] mdke: and slangasek remember to do that each release [11:09] ok. is there a project for that? [11:09] no [11:09] u-r is a team [11:09] fine [11:10] well, at least, i don't know of one :) [11:10] assigned :) [11:11] :) === nand_ is now known as nand [13:03] Hobbsee: Hi, can you use build.py right now? [13:05] geser: yes [13:06] Hobbsee: please give-back: libmrss libp11 libpreludedb. Thanks. [13:08] geser: It's buildd.py [13:13] geser: all given back [13:48] er, how do i debug why my keyboard and mouse have lockedup? [13:49] as in, i can move the mouse, but it won't click, etc, and nothing on my keyboard appears to work [13:49] If you force a VT switch does the keyboard work? [13:49] hwo do i force a VT switch? [13:49] Run 'sudo chvt 1' in an ssh session [13:49] ahhh [13:50] StevenK: it goes black, and then gets sent back to the X session [13:51] you know, i wonder if this is related to the consolekit bug thing [13:51] sorry, policykit [13:51] I wasn't expecting chvt 1 to send you right back to X [13:53] neither [13:53] but there is a bug about when you're logging in, where it won't let you get to a VT - it goes black, then sends you back to gdm. [13:53] Sounds sinister [13:54] ah, bingo [13:54] after killing consolekit, chvt 1 works fine, and i have keyboard. [13:54] which then locks up again when i go back to X [13:55] So X has gotten into some bad state [13:56] hrm, X will continually lock, even with consolekit killed. [13:56] Amaranth: fix it! :) [13:57] I daresay your video card does hardware pointer which is why the mouse moves -- the graphics hardware is updating the mouse pointer and keeping track of it, but X isn't listening since the keyboard and mouse clicks aren't working [13:57] Does killing compiz help? [13:57] Either X has gotten stuck or compiz has [13:57] Right [13:58] compiz usually gets stuck when a driver does something weird [13:58] Amaranth: Hopefully my stuff about the hardware pointer isn't complete bollocks [13:58] oh yes, i was going to try that [13:58] StevenK: Sounds right [13:58] Although when compiz is the problem it's not actually the case, everything is still getting events and such you just can't see it [13:58] Amaranth: then it manages to quit gdm, and drop me back at a console. [13:59] Amaranth: what's weird is that with EXA, my screen is still refreshing. [13:59] gah i can't wait for gallium [14:01] Amaranth: er, now killing compiz gets me back to a usable X window, yes. [14:01] So compiz is busticated. [14:01] What else is new? [14:01] * StevenK ducks [14:01] I suck with gdb but isn't it possible to attach to compiz and see what it is doing? [14:01] StevenK: heh [14:01] It's obviously not crashed so that would mean it's in an infinite loop somewhere [14:01] or stalled waiting for something [14:01] Amaranth: this reproducable by hitting meta+e, then alt+space [14:02] oh that one [14:02] Hah [14:02] I hate X11 [14:02] yeah, that seems to be oen of the easiest ways to do it [14:02] the problem is the menu gets a screengrab [14:02] meta+e and then alt-space opens two window menus for me [14:02] and keyboard grab [14:03] StevenK: yes, then click somewhere, and try to go back itno one of the desktops. [14:03] it's possible to escape from but not easy [14:03] LongPointyStick: Works for me [14:03] StevenK: lucky you. [14:03] I manage to get it stuck about every other time I try [14:03] I don't think I'll do that again [14:03] hehe [14:04] haha [14:05] Amaranth: how do i start compiz, etc, on a separate display? [14:05] eh? [14:05] displays are always strange creatures to me :) [14:05] DISPLAY=:0 compiz & [14:05] surely if i'm running compiz thru gdb, i need to start it from the ssh'd machine? [14:05] ahhhh [14:05] i'm not saying to start compiz in gdb [14:05] i'm saying to attach to it when it gets stuck [14:07] LongPointyStick: To do what Amaranth is asking, start compiz, get it stuck, find compiz's process id, run gdb and then run attach [14:07] or gdb -p [14:07] iirc [14:07] That will stop compiz, so you need to run continue, but then how to see what it's doing, I'm stuck [14:09] hum. attaching to it before it gets stuck just makes it freeze. [14:09] Of course, you need to run continue at the gdb prompt [14:09] gdb sends it a SIGSTOP, from memory [14:10] it's best to attach it when it sticks [14:11] so i continue, and it says Continuing, and it does nothing. [14:11] right, don't do that [14:11] attach to it after it gets stuck [14:11] then don't run continue [14:11] that was just after it got stuck [14:11] oh. [14:11] you should be able to run bt to see what it's just been doing [14:13] Maybe the last function on the call stack is annoy_hobbsee_lots_and_lots() [14:13] haha [14:13] Amaranth: no debugging symbols. [14:14] does the automatic repo thingy have them yet? [14:14] this would be easier on gutsy :P [14:14] i have no idea [14:15] there appears to be no compiz debug package in the reops [14:15] https://wiki.ubuntu.com/DebuggingProgramCrash [14:15] it's a separate repo [14:15] debug packages are created automatically [14:15] Hmm.. ddebs.ubuntu.com doesn't seem to have updated in the past 10 days [14:15] yeah, i thought so [14:16] ok then... [14:16] easier on gutsy :P [14:16] unless you feel like installing basically every X, GNOME, and KDE dev package and building compiz :) [14:17] well, i'll see if it crashes on gutsy then [14:17] it does [14:17] this particular bug has been around as long as compiz has [14:17] well, sort of [14:18] expo just made it easier to hit :) [14:18] what, the freezes? [14:19] Haha [14:20] crashes on gutsy too. [14:27] LongPointyStick: good, you can get debug symbols there :) [14:28] heh [14:43] Amaranth: it appears that they're broken anyway, due to the backports [14:44] Hobbsee: oh joy [14:44] yes... [16:19] hi [16:20] do you know why zsnes is in multiverse in ubuntu but in main in debian? [16:20] I think this is an error [16:20] cause zsnes is a free software [16:24] android: I see it in universe [16:24] zsnes | 1.510-1.1 | http://archive.ubuntu.com hardy/universe Sources [16:26] android: In feisty and Edgy was in multiverse, but Gutsy and Hardy have it in universe. [16:27] ok thank you the problem was solved === Adri2000_ is now known as Adri2000 === dfiloni_ is now known as dfiloni === mc44_ is now known as mc44 === TheMuso_ is now known as TheMuso [21:15] May I suggest adding a generic-pae kernel to the next version of Ubuntu? [21:15] Currently it requires a recompile. [21:16] How about making it the default if NX support or RAM mapped over 4G boundary is detected. [21:18] Also how about a non-PAE server kernel? [21:18] Make that the default if the processor does not support PAE. [21:19] And if installing a server version of Ubuntu. [21:19] yuhong: please ask on the kernel-team@ list (it's moderated, so please consider subscribing). === mpt_ is now known as mpt === ember_ is now known as ember