=== TheMuso_ is now known as TheMuso [01:51] did something change such that I need something else to set language? locale=en_US console-setup/layoutcode=us [02:13] hi [02:18] cjwatson, thanks for your message yesterday, about the Arabic problem in preseed [02:33] is there someone ? [02:35] once i've installed hardy using pxe and preseed, my client boots and the keyboard is in something like Arabic [02:36] as it has been reported here https://bugs.launchpad.net/ubuntu/+source/console-setup/+bug/328078 [02:36] Ubuntu bug 328078 in console-setup "surprising keyboard choice if not preseeded" [Medium,Triaged] [02:46] maxagaz: i seem to be having the same problem [02:46] my pxe default file has: [02:47] append initrd=ubuntu/jaunty/initrd.gz root=/dev/rd/0 rw locale=en_US console-setup/layoutcode=us netcfg/wireless_wep= netcfg/get_hostname= DEBCONF_DEBUG=5 url=http://shaz/ubuntu/jaunty/preseed.cfg tasksel:tasksel/first="ubuntu-desktop" [02:47] that worked a few weeks ago, but not today [02:49] my pxe default is : [02:49] append vga=normal initrd=ubuntu-installer/amd64/initrd.gz locale=en_US console-setup/layoutcode=en_US netcfg/choose_interface=eth0 netcfg/get_hostname= preseed/url=http://ubuntu.i-xanadu.com/xanadu-preseed.cfg -- [02:50] hmm, let me try en_US ... [02:55] maxagaz: well, that seemed to be my problem. so some good came out of this :) [02:56] CarlFK, you mean you don't have the problem anymore ? [03:00] well, I was having a different problem [03:00] I am just now starting the install [03:24] maxagaz: I see something odd [03:25] durring the install, I hit Alt-F2, see "hit enter for console..." [03:25] I hit enter, I get a prompt. I type "abc" I see "***" [03:25] where * = some diamond char [03:36] CarlFK, during the install, i don't have any question [03:39] evand: I tried a netboot (took a while to figure out in virtualbox..but ended up super easy), and used the preseed.cfg file the guy had attached...I didn't experience the same sort of problem (used alpha6) [03:39] I dunno if I replicated everything the same...shrug...ideas? [03:40] maxagaz: did you type something in the installers VT ? [03:41] evand: actually...I lied...I added "d-i console-setup/layoutcode string us" to the preseed file as per the suggestion in the bug [03:41] CarlFK, what's "the installers VT" [03:41] imma run once without that and see what it does [03:41] i'll check again my install using console-setup/layoutcode=us instead of console-setup/layoutcode=en_US [03:42] maxagaz: were you the one having the problem? [03:43] shtylman, my problem is about the Arabic keyboard [03:43] maxagaz: right...using the preseed file, after install arabic was the keyboard selected for you..right? [03:44] maxagaz: "the installers VT" is during the install, I hit Alt-F2, see "hit enter for console..." [03:44] shtylman, yes, that's me [03:44] shtylman, i posted a message yesterday as Gilles on https://bugs.launchpad.net/ubuntu/+source/console-setup/+bug/328078 [03:44] Ubuntu bug 328078 in console-setup "surprising keyboard choice if not preseeded" [Medium,Triaged] [03:45] maxagaz: ok, just so I know :) ... well I ran one install and it worked fine...with that line...im gonna run another without the preseed to confirm the bug [03:45] and see what I can do about it...hopefully :) [03:45] shtylman: you get what I am saying about the alt installer's VT? [03:46] CarlFK: lemme read up some...I just opened the irc window.. [03:47] same pxe line, before it even loads the preseed (I disconnected the lan). Alt-F2, enter for console, type "abc"at prompt, see "***" [03:47] CarlFK: before starting the install [03:48] I switch to VT and type... [03:48] english as normal [03:48] lemme start the install and try again... [03:48] do you autodetect keyboard layout? [03:48] or set manually? [03:48] shtylman: using these kernel parms? locale=en_US console-setup/layoutcode=en_US [03:49] CarlFK: no...lemme put those in and try [03:49] those take care of keyboard layout :) [03:49] CarlFK, a late answer... i didn't type anything in the installers VT (by the way, what does VT stands for ?) [03:50] virtual terminal [03:50] VT - Virtual Term... yeah, that [03:50] ok [03:53] CarlFK: maxagaz: behavior confirmed for "locale=en_US console-setup/layoutcode=en_US" as a kernel parameter ... I see ### in the virtual terminal instead of english... but by using "locale=en_US console-setup/layoutcode=us" it works [03:53] evand: ^ [03:54] are yall positive en_US is a valid code still? [03:54] shtylman: well, I got that from this IRC session :) [03:55] shtylman: I was using locale=en_US console-setup/layoutcode=us [03:55] CarlFK: heh...missed that... :) also...from what I gather en_US is for local, and 'us' is for keyboards [03:55] so the problem is the ungraceful default when using a non-valid layoutcode [03:56] cjwatson: evand: what is the default when using a layoutcode that is not valid? [03:56] "locale=en_US console-setup/layoutcode=us" stopped taking care of the keyboard question - console-setup/layoutcode=en_US did [03:57] CarlFK: huh? [03:57] um [03:58] (10:48:42 PM) shtylman: do you autodetect keyboard layout? [03:58] that implies the installer asks about the keyboard, right? [03:58] CarlFK: yea...after adding the kernel param it doesn't [03:58] kinda. [03:58] CarlFK: I hadn't added the param yet when I asked that [03:59] 2 weeks ago, using "locale=en_US console-setup/layoutcode=us" - no keyboard question [03:59] today, those prams, got asked about keyboard [03:59] using which netbook tar? [03:59] alpha6? [03:59] or current? [03:59] saw what maxagaz used, tried that, no keyboard q [04:00] http://archive.ubuntu.com/ubuntu/dists/jaunty/main/installer-amd64/current/images/netboot/ubuntu-installer/amd64/linux [04:04] CarlFK: just pulled latest linux and initrd and ran it with "locale=en_US console-setup/layoutcode=us ... result: no keybaord question [04:05] md5sum linux 93bdff35fdba8a7f547ebb563c3cb3c2 [04:05] is mine the same? [04:05] 4140534753c74a27c887ec49c58b20a3 initrd.gz [04:05] I have different for linux [04:06] and for initrd [04:06] where did you pull from? [04:07] that link [04:07] whoops - I didn't md5 the x64 one.. [04:07] 3be4c9beea45a3f5ae7bd2a72761234f initrd.gz [04:07] 42d4d7a9bb1333aff2a1580eddc5aa02 linux [04:08] k... yea [04:08] ran in virtual machine...no keyboard question... [04:08] and VT's were normal [04:08] can you try the x86? [04:09] sure [04:09] thanks [04:09] sorry bout that - forgot that my script pulled both [04:13] CarlFK: no prob.... result: i386 behavior = amd64 behavior -> no keyboard question [04:13] huh [04:13] what are you testing with? virtual machines or physical? [04:14] physical [04:15] http://shtylman.com/stuff/shtylman.netboot.tar.gz [04:16] thats a tar of my tftp directory [04:16] obviously you will need to change the preseed/url in the text.cfg file...but try that out if you want...see what it does [04:26] shtylman: does your's ask about keyboard ? [04:26] nope [04:26] goes straight to hostname [04:26] after network conf [04:28] which of these files has the kernel/append lines? [04:29] append vga=normal initrd=ubuntu-installer/i386/initrd.gz locale=en_US console-setup/layoutcode=us preseed/url=http://192.168.1.105/preseed/preseed.cfg -- quiet [04:30] text.cfg [04:30] in the boot-screens [04:30] yeah - grepped everything [04:30] k [04:42] well, im out...if there are more things needing testing, just leave a msg and I will try it tomorrow [04:42] thanks - im having lan issues... [09:36] good grief, that was a vast amount of confusion over a layout code :) [09:36] maxagaz: en_US has never been and will never be a valid keyboard layout code; it's a locale [09:37] maxagaz: so console-setup/layoutcode=us is indeed the proper fix for your problem [09:37] shtylman: well, that's exactly the bug that maxagaz quoted; it just picks the first in the list or some such, while it would probably be better if it stopped and asked [09:38] hello crazybyte [09:38] hello evand [09:38] see you later then :) [09:38] sure thing [09:39] maxagaz: BTW, I'd appreciate it if you were a bit more accurate in bug reports; in your statement in the bug report you said that you were using console-setup/layoutcode=us, not console-setup/layoutcode=en_US as you later clarified. If you'd been accurate from the start then I could have helped you straight away [10:03] cjwatson: if some crazy german people are contacting you directly about unattended ubuntu installation you can forward them to me;) i have link your launchpad profil and blog in my german howto with a thanking for your support [10:04] ok [10:16] cjwatson, that's right... it only when CarlFK gave us his pxe config that i thought about doing the same [10:16] it [10:16] it's [10:47] crazybyte: perhaps one approach to "grub-restore" would be to arrange for the "rescue" package to be usable on the live CD and have a UI and stuff [10:47] cjwatson, ok [10:47] i will look into it [10:47] since it has a grub-reinstall plugin [12:26] evand, hey [12:27] mat_t: hi [12:27] evand, alpha 5 installer didn't fit on my screen (mini 9)... Is that fixed for the beta? [12:28] mat_t: not quite yet: https://bugs.edge.launchpad.net/ubuntu/jaunty/+source/ubiquity/+bug/325958 [12:28] Ubuntu bug 325958 in ubiquity "Jaunty Alpha 4: Ubiquity windows does not fit on 1024x600 screen" [High,Triaged] [12:29] evand, thx [12:29] sure thing [14:06] evand: bug 334284 is still milestoned for beta; since I'm going to be asked about it in the release meeting, do you think it's going to make it? [14:06] Launchpad bug 334284 in ubiquity "Timezone map cities are incorrectly placed" [Medium,In progress] https://launchpad.net/bugs/334284 [14:10] I suspect not for beta, but hopefully shortly thereafter [14:10] I'll move the target [14:11] ok, thanks [14:28] evand: regarding the city placement....I worked at it quite a bit and I think I got the cities a bit closer...it was similar to your code with just a few changes for the offsets... its in the kde_frontend folder [14:29] shtylman: ok thanks, I'll take a look in a little bit. Trying to fix resizing issues. [14:29] k [14:37] cjwatson, Noting your comments in the release meeting: mcasadevall and I have been walking through armel installs today, tracking down failure points. [14:43] persia: at the moment it's a lot faster to deal with uninstallable packages by way of the testing-ports output and chdist [14:43] that's what I've been doing [14:43] Oh, then we're talking about different things. We've been walking through d-i :) [14:44] ah [14:44] I feared duplication, but it appears there isn't any :) [14:44] please do continue walking through d-i ;-) [15:10] cjwatson: are you gonna handle the keyboard preseed bug? and does its status need to change in launchpad? [15:13] shtylman: yeah, it's on my list. What's wrong with its status? [15:15] cjwatson: on second though...ignore the status comment :) [15:16] *thought [15:18] cjwatson, looking at partman/lib/disk-label.sh, I'm curious if you happen to know the historical reasons why each armel subarchitecture is separately set to "msdos" and *) is set to "UNKNOWN". [15:21] persia: I don't, but it's moot because that's been changed upstream [15:21] apparently I forgot to merge that. I will. [15:21] beta-critical? [15:22] Actually merging the upstream fix isn't, but we'd like the imx51 subarch to not have a warning for beta, if we can. [15:23] I can look at that merge, or just add the subarch, if it saves you time. [15:24] I have the merge in progress [15:24] OK. In that case, it would be nice if it happened by beta, but I'd say it's beta-medium or beta-high, rather than beta-critical. [15:25] ok, I'll see what I can do today [15:26] Thanks. I'll drop it from the list of imx51 stuff. [15:29] partman-partitioning: cjwatson * r703 ubuntu/ (14 files in 3 dirs): merge from Debian 65 [15:32] evand: I've been think (and yes I know it's dangerous) about the map. Is it relatively to scale? If so could you not just x,y plot it using the cities gps co-ordinates? [15:32] partman-partitioning: cjwatson * r704 ubuntu/debian/changelog: releasing version 65ubuntu1 [15:32] thinking even which is far more dangerous :) [15:32] davmor2: it's not that simple. [15:32] evand: Why? [15:32] persia: ^- [15:33] the points are in latitude and longitude, the map is a projection of the surface of a sphere onto a 2d surface. [15:34] cjwatson, Thanks a lot. [15:35] evand: Okay just thought I'd ask :) [15:36] sure, and thanks for the suggestion [15:38] hw-detect: cjwatson * r111 ubuntu/hw-detect.sh: typo [16:25] I need to install lenny on 8 boxes. Is it worth me trying to port my ubuntu pxe/preseed configs, or just burn 8 cd's and start answering questions? [16:26] I think I did some debian installs a few years ago. [16:27] I would start with the Ubuntu configuration but also compare the Debian and Ubuntu example-preseed.txt files and make modifications based on that [17:55] apt-setup: cjwatson * r164 ubuntu/ (debian/changelog generators/90services-select): [17:55] apt-setup: Make sure to exit services-select generator cleanly if [17:55] apt-setup: apt-setup/services-select is empty. [18:25] grub-installer: cjwatson * r774 ubuntu/ (debian/changelog grub-installer): [18:25] grub-installer: Only check for the boot partition being on dmraid if the user allowed us [18:25] grub-installer: to activate dmraid, to cope with cases where dmraid is misdetected [18:25] grub-installer: (LP: #346001). [18:28] grub-installer: cjwatson * r775 ubuntu/debian/changelog: releasing version 1.36ubuntu4 [18:48] * mcasadevall kicks his AJOIN list [19:09] cjwatson, https://bugs.edge.launchpad.net/ubuntu/+bug/di-imx51-support [19:10] cjwatson, I've added support to di for the imx51, and I've done a successful installation with it [19:10] mcasadevall: ok, thanks; being called away to dinner now but I'll queue that up [19:10] cjwatson, all the work is in bazaar branches :-) [19:10] so it should be easy to merge === mcasadevall is now known as NCommander