[00:34] ubiquity: cjwatson * r3223 karmic/ (4 files in 3 dirs): Strip whitespace from username (LP: #358656). [06:30] i know this probably isnt a tech support forum, i need to find someone who is pretty knowledgeable, and im guessing they dont hang out in the support channel troubleshooting 1300 users.i figured this to be a good place to try because my problem is deeply related to the install. the b43 error kicks in when ubiquity launches or so it seems in verbose mode so here goes: [06:30] ok Im gonna throw this out there. I'm doing a wubi install on an averatec 3200. I'm getting the b43 phy0 ucode5.fw error on the first reboot. i know how to fix it in 3 diferent ways. it goes to bash after sitting for a minute or 5, i fix it over and over and changes are lost after reboot. i need to find out why the changes aren't saved even when i sync and reboot, also it would help to know how to start ubuntu from the bash [06:30] the changes are in memory should it not be possible to have them written before i boot in and finish the install. if anyone knows what i need please message me or email me. by the way i am using 8.10 xubuntu wubi, figured id try that to minimixe the possibility of problems just until i get it running once. the only distro that has ever booted on this turd was goblinx although i didnt get wireless working,(didn't try either) [06:30] b43 firmware didnt send it for a loop. if i can figure this out im gonna ditch windows altogether (i'm a semi newbie but i've tried several distros and setup several installations, my little personal laptop is the only on thats ever given me any problems but its helped me learn alot) thanks [07:25] http://ubuntuforums.org/showthread.php?t=1127051 === evand1 is now known as evand [09:54] trying to install 8.10, once past language select, I cannot select any of the options... I hit enter and nothing happens... thoughts? [09:55] jfogg: Is this on the alternate CD or the desktop CD? [09:56] Desktop CD [09:56] The only option I can select is "Boot from first hard disk". [09:57] ah, I understand your problem better now. [09:58] anything else, I hit enter, the disk spins up and nothing happens [09:59] Hrm, not sure what would be useful information for an isolinux bug. cjwatson? [10:06] Does anyone have the MD5 handy for U8.10 Desktop? can't seem to find it on the site.. [10:07] jacobfoggg: alternate/live i386/amd64 ? [10:08] ubuntu-8.10-desktop-i386.iso [10:08] 24ea1163ea6c9f5dae77de8c49ee7c03 *ubuntu-8.10-desktop-i386.iso [10:09] bummer... I came up with - 9d6041d1188afb8a9c0bd802142609f4... DL'ing it again... =/ [10:09] thx... [10:12] Yes... Success - 24ea1163ea6c9f5dae77de8c49ee7c03 - Thx again! === evand1 is now known as evand [10:31] wl.ko is full of fail today. [10:47] evand: wubi is a lot better :) just trying to break in now and it's having none of it :) [10:47] davmor2: hooray! [10:49] evand: I've done kubuntu over ubuntu without removing ubuntu, across 2 partitions etc and it's just installing and removing without flaw [10:50] So I've got a branch to fix the use_biggest_free option at ~evand/ubiquity/use_biggest_free, just trying to get it sorted for KDE. I've already updated KDE for the changes to base.py and partman.py though. [10:50] davmor2: very glad to hear that [10:50] davmor2: how about uninstalling? :) [10:50] and CD eject [10:51] only issue I see that is still glaringly obvious is username from windows [10:51] everything is fine [10:52] evand: will that make the respin on friday? slangasek is doing us a we fixed this cd on friday to test stuff that wasn't going into rc [10:54] I'll try to get to it today, but I make no guarantees as there are other things on my plate at the moment. I'd suggest asking Ago about it, and I'll do the same. [10:54] What's the bug number for that/ [10:54] evand: sorry I read your kubutnu fix [10:54] crossed wires [10:55] ohh [10:55] I hope to upload that today [10:55] after I get another pair of eyes on the code [10:55] cool [10:56] evand: should I be able to upgrade wubi intrepid to jaunty? I'm guessing that is something that hasn't been done [10:57] where is it failing? [10:57] evand: it isn't I'm asking if it can, I'm just thinking of test that aren't covered by the tracker that people might try :)# [10:59] I'd definitely suggest giving that a try [10:59] it should work [10:59] will do I'll let you know how I get on :) [10:59] great, thank you! [10:59] evand: anything else you can think of? [11:01] suspend / resume, and making sure hibernate isn't a present option [11:01] but I think that's far more minor than everything elese [11:01] else* === evand1 is now known as evand [12:36] evand: lp:~evand/ubiquity/use_biggest_free looks like good progress [12:37] Thanks, I'm just about to commit KDE support [12:38] LP code review is really nice. I'm finding the inline diff quite helpful. [13:01] evand: upgrade worked :) [13:19] davmor2: great! [14:02] partman-auto: cjwatson * r289 ubuntu/debian/changelog: releasing version 84ubuntu4 [14:23] user-setup: cjwatson * r177 ubuntu/ (3 files in 2 dirs): [14:23] user-setup: Add preseedable passwd/auto-login-backup question; if set (e.g. to [14:23] user-setup: "oem") and passwd/auto-login=true, this backs up the previous contents [14:23] user-setup: of the configuration files (e.g. to /etc/kde4/kdm/kdmrc.oem). This can [14:23] user-setup: be used by oem-config to recover the old contents of the configuration [14:23] user-setup: files later (LP: #361272). [14:39] user-setup: cjwatson * r178 ubuntu/debian/changelog: releasing version 1.23ubuntu20 [14:40] oem-config: cjwatson * r659 trunk/debian/ (changelog oem-config-udeb.postinst): [14:40] oem-config: Preseed passwd/auto-login-backup=oem in the udeb, so that the original [14:40] oem-config: configuration files are available for restoration (LP: #361272). [14:41] ubiquity: cjwatson * r3224 ubiquity/ (debian/changelog ubiquity/frontend/base.py): [14:41] ubiquity: Preseed passwd/auto-login-backup=oem in OEM mode, so that the original [14:41] ubiquity: configuration files are available for restoration (LP: #361272). === evand1 is now known as evand [14:59] ubiquity: cjwatson * r3225 ubiquity/debian/ (11 files in 2 dirs): Update translations from Launchpad. [15:00] ^- included manual correction to Malay translation, notified to the translator; but we don't offer Malay just now anyway so it's not too critical [15:00] ok [15:21] cjwatson: Can you take a look at ~evand/ubiquity/use_biggest_free again when you have a chance? I've fixed the remaining issues and tested it [15:43] evand2: close the bug in the changelog please. Otherwise I think it's OK; the one thing that concerns me is that it looks like the biggest_free before/after frames will always be displayed in the KDE frontend, even if a different option is selected [15:44] doesn't it need a signal handler on that choice to deal with switching the frame visibility around? [15:44] _on_extra_toggle might do the job, and set the initial visibility to False rather than True? [15:45] * evand2 looks [15:46] if you haven't already, it's well worth setting up a test disk such that all the options are visible ... [15:46] (if necessary, munge the scripts in /lib/partman/automatically_partition/ ;-) ) [15:46] indeed, that's what I've been doing with the GTK frontend [15:51] _on_extra_toggle is already wired up for biggest_free [15:51] and I don't think the frames are always displayed, but I'm checking now [15:58] indeed, it works as expected [16:00] oh, it is? [16:01] I couldn't see how, but if it works I'll take your word for it [16:01] bah [16:03] Which reminds me, vim really needs a "draw vertical lines at each indentation level, and still work when expandtab is set" option [16:03] but yes, from my testing it works fine [16:03] evand2: I couldn't see how (no mention of _on_extra_toggle in the 'bzr merge --preview' diff), but if it works I'll take your word for it [16:04] oh cool, I didn't know about --preview, but you'll need to look at the code outside of a diff to see it [16:04] I did [16:05] that signal is connected in an 'elif choice != manual_choice:' block, the head of which now starts 'if choice == biggest_free_choice:' [16:06] No, it's (_on_choice_toggle) an indentation level back from that block [16:06] oh, but I guess it only needs to be connected for one choice maybe? I don't know, this code is awfully confusing [16:06] oh, right [16:07] ok, well if it works, go ahead and merge after adding the bug number, I'd say [16:07] will do [16:07] sorry for the confusion [16:07] thanks for that [16:08] meanwhile I've reproduced bug 361828 in debugging mode and am beating on it [16:09] neat, ubotu died. [16:09] trying to figure out whether the right solution is to mark partman-auto/automatically_partition as unseen when going back to it, or to just preseed it as unseen from the start [16:10] ah, yikes [16:13] that reminds me, bug 289663 is a tricky one. I need to find out why GTK/metacity hates us so much. [16:14] evand2: err, isn't that what your use_biggest_free branch fixes? [16:14] whoops [16:14] https://bugs.edge.launchpad.net/ubuntu/+source/ubiquity/+bug/361112 [16:15] mm, I think I ignored that because you seemed to be on top of it :) [16:16] ask seb/robert? [16:16] collins? [16:17] ancell [16:17] new gnome maintainer in addition to seb128 [16:18] ah cool [16:24] ubiquity: evand * r3226 ubiquity/ (6 files in 3 dirs): [16:24] ubiquity: Add support for displaying the "Use the largest continuous free space" [16:24] ubiquity: option in the partition bars. Previously ubiquity would create a single [16:24] ubiquity: Ubuntu partition that filled the entire drive when this option was [16:24] ubiquity: selected (LP: #289663). === evand2 is now known as evand [16:37] * jtholmes is away: for about 3 hours [16:43] http://paste.ubuntu.com/152169/ is the fix for 361828, I think; it turns out that it was overcomplicated to use self.autopartition_question there since it's guaranteed to be identical to self.current_question at that point [16:44] I believe that that was a relic of the way that method used to be organised [16:47] ah, good deal! [16:49] oh, wow, automatic mode really screws about with partman's logic [16:49] the fact that it doesn't call into the widget at all if the question is preseeded means that the state machine gets really confused and crashes [16:50] yikes [16:51] there's some really baffling behaviour here that I'm going to have to sort out [16:58] ah, race condition I think [17:01] why do we allow_change_step(True) in error_dialog and question_dialog? [17:01] that seems weird and I think is the root cause of a number of weird bugs [17:02] ubiquity: cjwatson * r3227 ubiquity/ (debian/changelog ubiquity/components/partman.py): Add a little extra state machine debugging to the partman component. [17:04] maybe it was to get the cursor back out of the watch state [17:07] * cjwatson considers http://paste.ubuntu.com/152187/ [17:09] yeah, that's better. It does look a bit weird because the Forward button gets enabled while the dialogs are up, but I think the dialogs are modal anyway - I'll leave a TODO to clean that up [17:16] ubiquity: cjwatson * r3228 ubiquity/ (3 files in 2 dirs): (log message trimmed) [17:16] ubiquity: If an error/question dialog is displayed while changing steps is [17:16] ubiquity: disabled, we currently need to re-enable changing steps in order to get [17:16] ubiquity: a normal cursor, but that also enables the Back/Forward buttons and [17:16] ubiquity: allows people to violate assertions by being too quick (e.g. selecting [17:16] ubiquity: an autopartitioning choice while the installer is still busy analysing [17:16] ubiquity: the possible choices). We should clean this up properly later, but for [17:19] ubiquity: cjwatson * r3229 ubiquity/ (debian/changelog ubiquity/components/partman.py): [17:19] ubiquity: Fix preseeding of the autopartitioning question to work in automatic [17:19] ubiquity: mode: it needs to be preseeded as unseen, otherwise going back from a [17:19] ubiquity: confirmation question doesn't work (LP: #361828). Remove an old code [17:19] ubiquity: path that would have been very complicated to fix for this and is [17:19] ubiquity: guaranteed to never be used nowadays. [17:27] I think I'll skip trying to fix partial preseeding; that sounds too risky for jaunty [18:59] ubiquity: superm1 * r3230 ubiquity/ (debian/changelog scripts/mythbuntu/mythbuntu-setup): [18:59] ubiquity: * Mythbuntu Frontend: [18:59] ubiquity: - Don't launch GUI applications from within the chroot as this causes [18:59] ubiquity: problems when libGL.so.1 and friends are diverted. Instead bind [18:59] ubiquity: mount the necessary files to launch them outside the chroot. [19:00] ubiquity: (LP: #362419) [19:54] ubiquity: superm1 * r3231 ubiquity/scripts/mythbuntu/mythbuntu-setup: fix umount command if setup is run to prevent multiple failures [20:28] ubiquity: evand * r3232 ubiquity/ (debian/changelog gui/glade/stepPartAuto.glade): [20:28] ubiquity: Pack the SegmentedBar widgets in GtkFrames rather than GtkEventBoxes [20:28] ubiquity: to work around repainting issues (LP: #362334). [20:30] ubiquity: cjwatson * r3233 ubiquity/ (3 files in 3 dirs): [20:30] ubiquity: Use a better colour to represent the results of automatic partitioning [20:30] ubiquity: (LP: #348461). [20:38] ubiquity: cjwatson * r3234 ubiquity/debian/changelog: =dev=-stripping fix in r3219 deals with LP: #362466 (I love it when I fix bugs before they're filed ...) === dpm_ is now known as dpm [21:36] Aside from doing a late_command in-target rm -rf /etc/network/interfaces; wget http://path_to_correct/intefaces_file is there any way to tell the installer to assign a static configuration to the network adapter even if it is using DHCP to get the original preseed file? [21:36] It seems to ignore preseed commands related to networking if it already has an IP from DHCP [21:46] DogWater: I have to admit I'm not sure offhand; my intention when writing that code (according to memory, and supported by comments in the code) was to permit network reconfiguration with different preseeded answers [21:46] so it sounds like a bug if that isn't working [21:47] installation-guide: cjwatson * r455 ubuntu/ (debian/changelog en/appendix/preseed.xml): Document controlling how partitions are mounted (LP: #347817). [21:55] installation-guide: cjwatson * r456 ubuntu/ (debian/changelog en/install-methods/automatic-install.xml): [21:55] installation-guide: Mention that Kickstart LVM configuration is now experimentally [21:55] installation-guide: supported, and document the pieces currently known to be missing. [21:56] installation-guide: cjwatson * r457 ubuntu/ (debian/changelog en/bookinfo.xml): Update Canonical's copyright years. [22:00] installation-guide: cjwatson * r458 ubuntu/debian/changelog: releasing version 20081208ubuntu3 === mcasadevall is now known as NCommander === NCommander is now known as mcasadevall === mcasadevall is now known as NCommander === Daviey_ is now known as Daviey