**** BEGIN LOGGING AT Thu Mar 17 02:59:57 2011 Mar 17 13:58:32 hi Mar 17 13:58:37 the installation broke down Mar 17 13:58:43 how can I get into the TTY Mar 17 13:58:50 I have no default username/password Mar 17 14:44:19 vin: you can mount the filesystem on another linux machine, chroot into it and do $passwd Mar 17 14:44:44 you can do even more stuff using qemu: http://wiki.debian.org/QemuUserEmulation Mar 17 15:13:03 hey guys can somebody tell me how to install cross-compiled modules? i cant use make install... can i? :-/ Mar 17 15:13:14 (im talking about non-kernel modules) Mar 17 15:13:35 ((meaning its a driver that isnt included in the kernel)) :D Mar 17 16:04:58 GrueMaster, how does your panel crash ? Mar 17 16:05:03 works fine for me Mar 17 16:05:34 Not sure of the nature of it yet, but it dies & respawns continuously atm. Mar 17 16:06:06 could be the resolution. Mar 17 16:06:30 ah, yeah, i remember something that requires 800px with Mar 17 16:06:35 *width Mar 17 16:06:59 you shouldnt see gnome-panel at all though Mar 17 16:07:13 did you explicitly pick gnome-session at gdm ? Mar 17 16:07:37 Well, whatever is generating the top panel is respawning. Mar 17 16:07:56 thats unity-2d-panel Mar 17 16:08:09 (Qt app) Mar 17 16:08:10 Tomato tomato. Mar 17 16:16:06 GrueMaster: issue we had was with gnome-panel Mar 17 16:16:14 and not unity-2d-panel, so it's probably a different bug this time Mar 17 16:16:30 Yea I know. Still, same visual effect. Mar 17 16:17:24 does it go away if you use a higher res ? Mar 17 16:17:43 Don't know. Just rebooting with the fix now. Mar 17 16:17:55 nope. Mar 17 16:19:31 Hmmm. Why is there a /home/buildd dir? Mar 17 16:20:49 that shouldnt be there Mar 17 16:20:53 * rsalveti lunch Mar 17 16:21:39 It's on the image I downloaded. Very odd. Mar 17 16:21:53 i see it here too Mar 17 16:21:55 Thought it might have been a glitch in the SD. Mar 17 16:21:59 ok Mar 17 16:22:25 * ogra_ just copies a loop mounted natty netbook into the ac100 Mar 17 16:23:01 there is only the homedir though, no user Mar 17 16:27:29 root@isis:/# dpkg -S /home/buildd Mar 17 16:27:29 dpkg-query: no path found matching pattern /home/buildd Mar 17 16:27:43 doesnt come from a package (would have surprised me) Mar 17 16:29:51 not in any postinst either Mar 17 16:34:48 ls Mar 17 16:34:55 gah Mar 17 16:35:58 I think I know why the bug reports are broken. There is a chromium-browser bug that the script is failing on. Mar 17 16:36:32 oh? Mar 17 16:36:37 how can that be Mar 17 16:41:36 Hmm. Not related. Working with Brian now. Mar 17 17:47:20 The unity-2d-panel issue appears to be sound related (figures). Mar 17 17:49:04 dmesg has a long list of "asoc: no valid backend routes for PCM: SDP4430 Media" and .xsession-errors shows libsoundmenu.so is respawning. Mar 17 19:57:09 Interesting. This issue is omap4 only (beaglexm works fine). Mar 17 19:59:37 dual core? Mar 17 20:00:56 which issue? Mar 17 20:03:43 janimo: The libsoundmenu respawn issue. And I am not sure yet if it is dual core related. That is my next step, although I doubt it. Mar 17 20:04:08 ah, I am not uptodate with U-2d related bugs Mar 17 20:04:30 I think I should test that too again Mar 17 20:04:38 This is new as of today's image. Mar 17 20:04:45 I found both compiz an u-2d frustrating to use on x86 Mar 17 20:04:52 so did not even tried to use it on arm Mar 17 20:04:56 And I didn't see it yesterday when updating an older image. Mar 17 20:05:33 GrueMaster, there were again libindicator related updates since yesterday, is the soundmenu among them? Mar 17 20:06:43 Might be. Let me work to dig deeper for a bit. Mar 17 20:06:49 Need food atm. Mar 17 22:05:40 After much digging (mainly in the dark), it appears that the schema from indicator-sound is not being properly registered. Mar 17 22:08:15 GrueMaster, i claimed the flash-kernel doesnt respect double dash notiation bug back Mar 17 22:08:39 will work on that during the flash-kernel reworking Mar 17 22:08:45 ok. Are you actually going to fix it? It has been there since karmic. Mar 17 22:08:51 ok Mar 17 22:10:40 its a pretty big task and bound to architectural design changes for flash-kernel-installer Mar 17 22:11:05 i always planned to work on it ... currently we dont use live or d-i images so its not pressing Mar 17 22:12:17 I'm just trying to whittle down open bugs. Mar 17 22:13:14 btw: I am making some progress on the unity-2d-panel respawn. It so far appears to be indicator-sound update. Mar 17 22:13:28 I'm getting ready to backrev and see if it still fails. Mar 17 22:13:54 It is not smp related, but it is omap4 specific (runs fine on beaglexm). Mar 17 22:16:45 try linking the binary of indicator-sound to /bin/true ;) Mar 17 22:17:06 runs fine here too Mar 17 22:17:15 even with broken sound in the kernel Mar 17 22:18:24 Looks like there is a new version as of 3 hours ago. Will try it first. Mar 17 22:19:11 I have a feeling something broke in the newer releases for the postinst, as upgrading from 20110311 works fine. Mar 17 22:19:45 fun Mar 17 22:20:09 its probably the buildd user that broke it :P Mar 17 22:20:20 heh Mar 17 22:20:24 * ogra_ really has no clue where that could come from Mar 17 22:21:16 Simple. A botched update to the posinst script or a bad tweak to the schema.xml file. The old settings would be still valid. Mar 17 22:21:36 i was referring to /home/buildd Mar 17 22:22:27 Oh, that. Mar 17 22:23:37 Hmm. Looks like all of the com.canonical.indicator.* schemas may be hosed. Updating indicator-sound got passed it's issue, now datetime failes. Mar 17 22:23:52 (as does my typing) Mar 17 22:24:09 weird Mar 17 22:24:27 i'm using an omap3 image here, freshly installed on the ac100 Mar 17 22:24:46 (with rolled back libc but no other modifications) Mar 17 22:25:02 apart from the g-s-d breakage all seems to be fine Mar 17 22:25:55 I can't really consider running on a non-supported system with a non-supported image completely valid. Besides, it also works fine on my beaglexm Mar 17 22:26:18 pfft Mar 17 22:26:25 userspace is always the same Mar 17 22:26:51 There can always be subtle hardware differences. Mar 17 22:26:52 as long as you have all kernel options it should be fine Mar 17 22:27:02 which we need to cover :) Mar 17 22:27:22 Not up to us to validate hardware. Mar 17 22:27:36 yet Mar 17 22:27:39 up to us to make sure userspace runs on all v7 HW Mar 17 22:27:57 which is one of the reasons to not hard build with NEON enabled Mar 17 22:28:13 Then give me more v7 systems to test on. Otherwise I can't comment on failures for which I have no hardware. Mar 17 22:28:27 imho we all should have ac100s and efikas Mar 17 22:28:48 Well, that I'll definitely agree to. :D Mar 17 22:28:54 just to be able to validate Mar 17 22:29:33 i'm considering buying a second one ... i just found out the 3G version is available for 199€ now Mar 17 22:30:11 http://launchpadlibrarian.net/66592855/20110317-ubiquity-sarcasm.png Mar 17 22:30:11 Not bad. Wish I had the $$$ to spend. Mar 17 22:30:16 * ogra_ grins Mar 18 00:32:14 GrueMaster, any idea why florence isnt in the archive ? the source was accepted and the binaries are lintian clean, so i dont get why they didnt make it into the archive Mar 18 00:32:24 (your mail indicates you know more) Mar 18 02:18:50 Hi, I had a couple of questions. **** ENDING LOGGING AT Fri Mar 18 02:59:57 2011