=== prpplague is now known as prpplague_afk === asac_ is now known as asac === NCommander is now known as Guest11203 === Guest11203 is now known as NCommander === daurnima1or is now known as daurnimator [10:56] ericm|ubuntu: the target to get a binary for a beagleboaard on natty/master is binary-omap, right? [11:00] ppisati, I believe so [11:00] ppisati, cooloney should know more detail [11:01] ericm|ubuntu: k [12:45] ogra_, janimo: the test rebuild for armel did finish for main, now building the universe packages. however I would like to start another test rebuild using gcc-4.6 for main, setting the priority of this test rebuild higher than the current running rebuild for universe [12:45] ok with you? [12:45] once gcj-4.6 is built [12:54] doko, fine by me. There are not many important uploads remaining [12:54] ok, thanks [13:03] * ogra_ wishes we could exclude packages on demand for such rebuilds [13:04] doko, fine with me, though if all big packages jump on the buoldds at the same time and lock all buildds i'll ask lamont to kill a build [13:04] it's done alphabetically [13:05] and if you are interested, you should be able to set the priority for single remaining builds [13:05] ah, i didnt know, yesterday we only had two left, the rest was taken by things like thunderbird, firefox, eglibc gcc etc ... [13:05] yeah, i know, doesnt help if all buildds are taken by packages that take more than a day [13:06] go ahead, just be aware that i'll ask lamot to kill stuff if its urgent and the rebuild is blocking [13:07] (i assume things will get retried anyway if they are killed) [13:59] bug 758732 === zyga is now known as zyga-food === zyga-food is now known as zyga [14:51] has anyone noticed a problem with the gcc-4.5-arm-linux-gnueabi package for natty? it has an unmet dependency: "Depends: libgcc1-armel-cross (>= 1:4.5.2-8ubuntu3cross1.47) but 1:4.5.2-8ubuntu1cross1.61 is to be installed" [14:54] hrw: ^ [14:54] i think it's a recent problem. i updated the toolchain this morning. [14:58] yes, we know that [14:59] I have a talk with steve in <1h so we will discuss this too [15:02] hrw, cool. thanks. === Sarvatt is now known as Sarvatt|2 === Sarvatt_ is now known as Sarvatt [16:22] headless images fail to start oem-config. [16:23] debconf database is locked apparently. === prpplague_afk is now known as prpplague [16:45] ogra_: I am going to have to fail the headless images due to bug 758858. [16:56] GrueMaster, hrm [16:57] netbook looks fine apart from one jasper issue with the ppa icon that i need to fix [16:59] I just booted netbook. Appears to be ok so far. [16:59] it feels quite a lot slower though [17:03] Hmmm. oem-config died near the end on my netbook image. [17:05] Generating a bug report now. Not sure if it is reproducable yet (these types of failures take ~30 minutes to reproduce). [17:06] didnt die for me here [17:07] did you get an apport report ? [17:07] yes. Looking over the logs now. I also have a stack trace. [17:07] very strange [17:11] ogra_: bug 758910. I'm going to reflash and retry. [17:24] GrueMaster, you say it restarted, did it finish on the second run ? [17:24] yes [17:25] hmm, itz gtk boog i'd say [17:26] Grumble. Need a proper German translator bot. :P [17:26] lol [17:26] well, keep an eye on it, but i doubt we'll see it again [17:26] i did three installs already and havent seen it [17:27] I'm reflashing & retrying now. Unfortunately my desktop system tends to degrade over time flashing sd cards and I end up having to reboot. [17:27] bah [17:33] Hmm. Running evolution, firefox, amarok, kde desktop, two VM's in virtualbox and flashing an sd card seems to low my system down a bit. [17:34] loagavg: 5.90 [17:35] on the panda ? [17:36] Pfft. Panda doesn't do VM. [17:37] ah, indeed :) [17:38] oh, as to the win32 image writer my son developed for jaunty, we are on our own if we want to add functionality to it. He is too busy to help before summer. [17:39] I'll see if I can get Ncommander to look at it before release. [17:44] * ogra_ takes a short break and will then jump into headless mode ... [17:47] * GrueMaster thought that was ogra_'s standard modus operendi. === prpplague^2 is now known as prpplague === prpplague^2 is now known as prpplague [18:02] Nice. My panda is now detecting my wireless AP on the other end of the house. With no antenae. [18:03] there is an antenna on the PCB [18:03] (also detecting my neighbor's wifi. Mua haha). [18:04] the socket is just for an extender [18:04] Ah. [18:04] i couldnt get bT to work though [18:04] hi all [18:04] ubuntu-arm works with Devkit8000? [18:09] ttyandroid: Possibly. Try downloading the latest natty headless image to start with. We also have a netbook image as well. May need to replace x-loader & u-boot, but the rest of the pool should work fine. [18:09] Many thks GrueMaster [18:11] ttyandroid: Be aware that since it is essentially a beagle, not an XM, memory will be a major limitation. [18:12] GrueMaster: i have buy one BeagleBoard-XM A3 and stop working [18:12] HW failure? [18:13] GrueMaster yes mmc card [18:13] GrueMaster: i try to buy other Beagleboard-XM but not find [18:13] The card or the card reader? [18:13] GrueMaster: card reader [18:14] Ouch. [18:14] GrueMaster u-boot bin from original SD don't work [18:14] GrueMaster: no load other u-boot.bin and console seems nothing [18:15] Which image are you trying? [18:15] GrueMaster original SD image [18:15] oh, intresting ! [18:16] ogra_: ??? [18:16] on headless i see asoc set up the sound stuff correctly [18:16] GrueMaster: Angstrom-demo-image [18:16] on netbook i still had the old errors [18:16] GrueMaster have another seller of BeagleBoard-xM ? [18:16] * ogra doesnt get that [18:17] ogra_: that is very odd. [18:18] ttyandroid: I'm in the US. I just used Digikey. [18:18] GrueMaster, http://paste.ubuntu.com/593227/ [18:19] WTH??? [18:19] I'll check it here, since I have two boards to play with side by side. [18:19] yeah, confusing, isnt it ? [18:20] headless was built last while netbook was built first in a chunk of ten image builds or so [18:20] I wonder if it is because of the alsa-utils stuff we added post-maverick? [18:20] * ogra goes to check the manifests [18:21] thats a good idea .... i'm just noticing that headless doesnt have any alsa userspace stuff [18:22] i bet TheMuso carried that over from maverick [18:22] Silly little aussie. :P [18:23] well, nobody told him to drop it :) [18:23] hmm, there are no processes at all on the headless image [18:23] i dont get why it would fail [18:24] oem-config? Don't know either. [18:25] bah, crap ... [18:25] * ogra forgot he has no job control in the rootshell [18:25] starting over === Lopi is now known as usb26 === usb26 is now known as Lopi [19:12] ogra_: Interesting tidbit. I removed /usr/share/alsa/init/omap4 and after a reboot, I now have no dmesg errors relating to asoc. Checking to see if sound works. [19:15] GrueMaster, that would be awesome [19:20] No sound. Mucked with alsamixer controls, nothing. [19:21] Haven't tried the patches and updates you have yet. [19:22] ah, well, you will need the new pulse from people.c.c [19:25] tried with new pulse and nothing [19:25] i get a new device in the sound properties [19:26] but it only provides an input device apparently [19:26] i sadly dont have the time to play more with wound today [19:26] *sound [19:26] will have to wait until post beta [19:28] well, in any case dropping the maverick scripts should be a step forward [19:31] GrueMaster, do you remember when we had the last working headless images ? [19:31] Beta 1. [19:31] * ogra_ tries to figure out the uboquity changes that might have caused this [19:32] i'm pretty sure there were images after b1 [19:32] but i dont know for how long [19:34] I have all dailies since Beta. Haven't been able to test them as I was hobbled last week. I can do a binary test and see where the point of failure started. [19:35] i would guess with april 3rd [19:36] i see two debconf related changes in the ubiquity upload from the 2nd [19:37] Ok, starting there. [19:57] ogra_: 20110403 fails [19:58] if 02 works i think i know what it is [19:59] i'll try to remove that code bit here, lets see what happens then [20:00] Checking 02 now. [20:00] (as well as juggling a mirade of other test images). [20:07] 02 passes. [20:08] So the relevant version of oem-config is 2.5.32->2.5.33 [20:11] yep [20:12] + * Do a quick check at startup to see if the debconf database is locked by [20:12] + something else, and exit slightly more gracefully than by raising an [20:12] + exception (LP: #746020). [20:12] thats what i suspect to be at fault [20:13] * ogra_ tests [20:14] Is there a tool to check the debconf db? [20:14] no idea [20:14] Could be a simple boolean error. [20:15] i dont think there is a tool [20:15] works ! [20:15] * ogra_ is at the language selection screen [20:22] aaaand ... install successfully finished \o/ [20:24] yup, all fine, even after a reboot [20:30] You mentioned that jasper is doing something with debconf? [20:31] yes, but it does that since before beta1 [20:31] and it does it before the reboot, i already asked colin if it could cause any issue, but he denied [20:32] it only sets debian-installer/framebuffer to false [20:33] i simply think the check is worng, but i wouldnt know how to write it different or better [20:33] i fear we need cjwatson [20:37] hey all, i have an arm machine that unless the hdmi cable is plugged in, it freezes during boot. i've set the console to equal our serial console, and i added debug=vc to the kernel command line... and it appears it gets to this point: + /scripts/init-top/plymouth [20:37] and then it stops [20:38] What is the platform? Image? [20:38] uImage, Maverick, ARM mx515 [20:39] I don't think we had an mx515 kernel for maverick. I know we didn't have images. [20:39] Might check with #linaro folks. [20:39] The Linaro kernel supports some mx515 boards, but I guess it's a custom kernel [20:39] you know it is lool :P [20:40] steev: This is in the initrd still; plymouth touches framebuffer (it outputs the boot progress splash and messages etc.) [20:40] steev: To poke at what's happening, you can add break=something to your kernel cmdline, this will give you a shell in the initrd [20:40] steev: rgrep maybe_break /usr/share/initramfs-tools to see possible breakpoints you can use [20:41] lool: my understanding was that, plymouth was supposed to see if it could use a framebuffer and if not (ttymxc0) was supposed to go down to a text console, it doesn't appear to be going to a text console [20:41] steev: For instance, break=top is a good choice to get started [20:41] okay, ill give that a shot :) [20:41] steev: I can't speak for what plymouth attempts to do, you might have to research this yourself; but I can tell you how to run things by hand ;-) [20:42] well if i can step through it, that will point where i need to go :) [20:42] steev: To read the sources, take the initrd.gz from the system and unpack it in a directory of your choice: mkdir i; cd i; gunzip -c < ../initrd.gz | cpio -i [20:42] steev: That will give you the same contents as what's in the initrd from the shell you will have [20:42] steev: You can run exactly the same shell commands to see at which points it goes wrong [20:56] is there a way to make plymouth itself be a tad more verbose, like saying what it's attempting to do? [21:02] GrueMaster, see -release ... fix underway ;) [21:03] I've been monitoring it. Cool. [21:03] :D [21:03] so apart from sound we should be good now [21:03] Yep. [21:03] and the odd color change in headless :) [21:04] I still don't see that. [21:04] i still wonder why only i seem to see it [21:04] snap :) [21:04] but i see it on all machines, its not client bound ... else i would have blamed the ac100 [21:04] Maybe it looks different when not viewed through a beer glass. [21:05] hmm, i should try that [21:05] heh [21:05] * ogra_ goes to look for a whisky botte to replace the glass [21:05] *bottle [21:06] GrueMaster, but do you use screen ? [21:06] screen+byobu [21:06] (i think i remember you said byobu) [21:07] that might apply colormap changes [21:08] Well, the system that the serial console is hooked to is headless. So ssh->byobu->many screen sessions. [21:08] ah [21:08] i wouldnt mind having it b/w [21:08] i wonder if there is a way to enforce that on the installer side [21:17] where does /conf/param.conf come from (we don't have one, which is why i ask) [21:19] ogra_: I'm going to go ahead and greenlight armel rebuilds after oem-config is fixed. I haven't found anythhing that is critical and no one on kubuntu is interested in testing prior to rebuild. [21:20] GrueMaster, yeah, just go ahead, i'll test tomorrow as soon as i get up [21:20] That will really help, as I have an am post-op appointment tomorrow. [21:20] no prob [21:42] even when setting up the postfix (calling debconf) the colors were weird [21:43] pink and gray IIRC [21:43] mine are completely inverted [21:43] rendering the whole ui unreadable [21:43] yeah, seems to be inverted [21:44] https://bugs.launchpad.net/bugs/747229 has screenshots [21:45] Wow. I am definately not seeing that. [21:46] (of course, I am currently not seeing much of any oem-config atm) :P [21:46] heh [22:05] hi is anyone available to help me with a Ubuntu 10.10 ARM/OMAP question? [22:05] Run into a bit of a problem working with the USB connection [22:10] Schamus: Give more details. What are you seeing? [22:10] Quite right, I'm running a Beagleboard xM A3,with ubuntu 10.10. I'm coding a python script to manipulate an LCD screen via USB [22:11] The python script uses import serial, sets up the connection, and shoots hex bytes controlling and manipulatingthe screen to show items, however, on the beagleboard its apparently not sending the information over correctly [22:11] Ok. Not sure how much help we can be. have you tried this on x86 first? [22:11] On any other laptop this script works perfectly, but the beagleboard does not [22:12] So i was curious if USB drivers on the Ubuntu 10.10 ARM edition are a bit different than a standard Ubuntu 10.10 install [22:12] They should be the same. I believe the kernel was the same for both x86 & armel. Could be a config issue though. [22:13] How so? [22:14] Check /bot/config-`uname -r` [22:14] boot [22:15] The code is the same tree as x86. If it is a usb issue, it is either in the core arch code or a missing config option. [22:15] The rest is arch generic. [22:16] Also, try the latest maverick kernel or the natty kernel. [22:16] yeah i did updates yesterday [22:19] so, excuse my amateur, you mean something like "gedit /boot/config-'uname -r' correct? [22:19] yes [22:20] (you can actually just use tab completion, but it was easier to type uname -r to indicate the config for the running kernel) [22:21] yeah i've been picking up linux over the past month or two so i'm a little ignorant in some basics still [22:23] im pulling up a blank with that command [22:24] type 'ls /boot-config*'. That should give you a list of config files. [22:24] Look at the latest one. [22:26] Ah, got it now. [22:26] is there a specific section im looking at for USB? [22:27] I have no idea personally. I don't have one of these lcd screens to try to reproduce this issue with. [22:28] What image are you running on your laptop? If it is running the same release, check to see if it has the same kernel rev. [22:28] That is the easiest starting point. [22:30] 10.04 on my laptop, 10.10 on the beagleboard [22:30] I would suggest upgrading the laptop. It will be easier to diagnose that way. [22:31] You can also just download the iso and use usb-creator to make a bootable usb drive to test with. [22:32] Yeah, I mean I tried many other USB devices, (i.e. bluetooth dongle, mouse, keyboard) and they all work, just this LCD screen doesn't seem to want to accept streams from the beagleboard [22:34] could it be a problem with permissions? === Michael_ is now known as NCommander