=== bladernr_ is now known as bladernr_afk === smb` is now known as smb === tjaalton_ is now known as tjaalton [08:16] * smb wonders what a "strut" is... [08:19] hi apw [08:20] apw: would you update aufs when you get simply instructions to do so? already verified with an iso image using aufs3? [08:21] apw: also ndiswrapper 1.57 compiles with 3.2 kernel, why dont intergrate that? [08:23] bleagh === _ruben_ is now known as _ruben [11:39] * ppisati -> reboot === chuck_ is now known as zul [12:21] who looks after the 3.3 shit? :p [12:21] kernel.ubuntu.com [12:22] ooo theres an rc2 now ;) [12:25] rc1 wasnt booting on an amd zacate system of mine [12:25] but it does on bulldozer 8150 [13:27] 1276852736 [13:49] Hi, all. I'm using ubuntu-oneiric kernel on the ARM device. There is a problem: when I halt the system it have a segfault after unmounting file system... please, advice, how can I fix it? [13:49] ppisati, ^^ [13:50] reisei: kernel version? [13:50] ppisati: 3.0.13 [13:51] reisei: i'm checking, hold on [13:52] reisei: uname -a [13:53] ppisati: [13:54] Linux omap4430-panda 3.0.13 #0 SMP PREEMPT Wed Jan 25 13:16:31 MSK 2012 armv7l GNU/Linux [13:58] reisei: did you build it by yourself? [13:59] ppisati: yep === dduffey_afk is now known as dduffey [14:00] reisei: then it's unsupported [14:00] reisei: btw, do you have musb enabled? [14:01] ppisati: :-( Even, you can't suggest what's wrong? [14:01] ppisati: yes. [14:01] reisei: ok, disable it, it causes panic at shutdown [14:01] reisei: i swa some fixes for it upstream, but never ehcked if they hit -stable [14:02] ppisati: thanks! I'll try it. [14:11] ppisati: yes, it works. [14:12] reisei: ok then, either you update to a more recent -stable version of you r kernel [14:12] reisei: or you could look for "musb shutdown fixes" on the arm lkml ml [14:13] tgardner, did ogasawara confirm it was the procps patches that were her boot problem do you know ? [14:13] apw: I did [14:13] apw: I yanked em [14:13] apw, no, I do not know for sure [14:13] ah, nm [14:13] ogasawara, oh heh hi [14:13] it early there isn't it ? [14:14] ppisati: ok. Thanks a lot... Can't solve that problem for a week. [14:14] apw: it is, but it's my normal start time anyways [14:14] apw: I try to squeeze in as much time while the nugget is asleep [14:15] apw: I tested on i386. it would appear to boot properly, but then hang and vomit an oops for a null pointer dereference [14:16] ogasawara, not good. will have a play with it and try and figure that one out [14:16] apw: I can try booting it again here and get more detailed info [14:17] apw: hrm, except I didn't save the bad build nor the patches. you have them somewhere so I can re-create? [14:17] ogasawara, i have em, but don't worry i'll play with em [14:18] commit 633b45454503489209b0d9a45f9e3cd1b852c614 [14:18] Author: Eric Paris [14:18] Date: Tue Jan 3 14:23:08 2012 -0500 [14:18] audit: only allow tasks to set their loginuid if it is -1 [14:18] ogasawara, i'll let you investigate that one, which is a new tunable for systemd systems whcih we may need to be aware of [14:18] apw: ack [14:29] ok, configuring /etc/asound.conf to use pulse made audio in flash video work again [14:32] my flash sound is coming out the internal speaker instead of the usb headset, even though any other sound is coming out the headset [14:34] brendand: out of curiosity, what's thje content of /etc/asound.conf for you? [14:34] ppisati, eh - it doesn't exist? any other place it could be? [14:35] cat ~/.asoundrc [14:36] ppisati, nor that [14:39] brendand: than i don't know, i had to do all this crap to make it work [14:40] ppisati, not sure why i don't have those files [14:40] brendand: what i read around is that if you don't get any noise out of flash is beacuse the audio is nit routed through pulse-audio [14:41] brendand: in fact, without it, i don't see any flash-stuff in the consumer list of pulseaudio (Sound Settings -> Applications) [14:42] ppisati, i think i have a different problem. i'll need to find exactly what's going on when i get a chance. probably a reboot will fix it though [14:55] * ppisati -> coffee, brb [14:56] * cking -> pick up kid from school, back in 25 mins or so === shadeslayer_ is now known as shadeslayer [15:35] ogasawara, am i correct in thinking you are using master-next in ubuntu-q ? [15:35] apw, I created master-next in that repo [15:35] apw: I am [15:35] apw: I need to sync it up again with the latest precise master-next though [15:36] ogasawara, yeah am starting to wonder if i should be building against it [15:36] ogasawara, actually that came up recently, should we be uploading that regularly, like we do the pre-proposed heads ? [15:37] apw: for q? I don't think it's necessary at this point in time [15:37] apw: for precise master-next maybe [15:39] ogasawara, yeah we cirtainly can add precise to the mix there, if it is helpful. there is no obvious reason not to [15:39] now that PPAs work [15:39] apw: yah, that seems reasonable. I think maybe once we hit kernel freeze for precise, then q would be reasonable to start building etc. [15:41] * ogasawara back in 20 [15:43] tgardner, is that AMD laptop en-route to me now? [15:44] ogasawara, ok, done, precisise should now be pushed from tommorrow [15:46] cking, yep, went out yesterday afternoon [15:46] tgardner, thanks! [16:01] apw: ack, thanks! [16:23] apw: so for this systemd loginuid patch, did you just see this in passing or is there another source which triggered this investigation, ie a LP bug? [16:24] ogasawara, i was looking through the patches against fs/proc to see why the poo i pushed yesterday was exploding on you, and that one jumped out at me [16:25] ogasawara, so no, nothing other than randomly bumping into it, no idea what we even have that set to [16:25] ogasawara, i assume its an 'affecting Q' think now I think about it [16:25] apw: I assume it is something we'd want to toggle on, I'd have to double check Q to make sure we did. [16:26] apw: I'm actually running the Q kernel on my of my machines here [16:26] config.common.ubuntu:460:CONFIG_AUDIT_LOGINUID_IMMUTABLE=y [16:26] ogasawara, actually i have no idea what way it needs to be, upstart (i assume) may be compatible with it being systemds way or not [16:27] ogasawara, so i don't even know it is exposed hrrmmm ... perhaps we should be referring it to jhunt [16:28] apw: from reading the commit message and Kconfig details, it does say that it should be toggled on for systems which use systemd or similar central process services which I would interpret to be upstart [16:29] ogasawara, so probabally we just should tell jhunt about it, and possibly security as it only affects what i think of as a security id for yourself [16:29] apw: yep [16:30] jjohansen, ^ one for your endless todo list ... this option may affect us in Q [16:31] ok, i definitely have a new bug in precise. no matter what i do audio from the browser always comes out of the external speaker. it's verrrry annoying [16:34] pacmd list-sink-inputs doesn't list anything when browser audio is playing [16:34] hmmm. wonder if it's just chrome [16:35] nope [16:35] ppisati, does this sound more like what you were seeing? [16:37] ogasawara, https://wiki.ubuntu.com/Kernel/Reference/fwts/s3 and https://wiki.ubuntu.com/Kernel/Reference/fwts/s4 [16:38] cking: awesome, thank you! [16:38] and fwts is in universe, so it's all ready to go. [16:38] cking: sweet [16:39] brendand: nope, for me it was just flash-audio not working [16:40] ppisati, sure just flash? i thought it was as well, but noticed while in a g+ hangout that the voice audio was coming through the headset fine (because it's coming from the gtalk plugin) whereas the audio effects in the hangout were coming out of internal [16:41] brendand: yep, it was just flash [16:44] tgardner: you din't get this one - http://static.1wt.eu/img/articles/guruplug-slow-heater/guruplug.jpg - right? [16:44] jsalisbury_, Maybe that might be interesting de47a4176c532ef5961b8a46a2d541a3517412d3 [16:45] ppisati, nope. gimme a sec [16:45] ppisati, http://www.globalscaletechnologies.com/t-dreamplugdetails.aspx [16:45] tgardner: http://www.globalscaletechnologies.com/t-dreamplugdetails.aspx? [16:45] ok [16:45] ppisati, which is better :) [16:46] well, the one i posted is a fire hazard [16:46] http://1wt.eu/articles/guruplug-slow-heater/ [16:46] that's why i asked for the fan [16:49] ppisati, one interesting feature of the dreamplug is that it also has wifi built into it and comes up as an AP by default [16:49] tgardner: ah, nice, but for my LAN i opted for this one: http://pcengines.ch/alix2d13.htm [16:50] raised: https://bugs.launchpad.net/ubuntu/+source/pulseaudio/+bug/928980 [16:50] Launchpad bug 928980 in pulseaudio "Browser generated audio comes out the wrong sink" [Undecided,New] [16:50] ppisati, yeah, I've used a lot of PC Engines boards in another life. [16:50] diwic - any idea why this could be happening? [16:51] diwic - seems like pulseaudio won't manage the browser audio [16:51] brendand, yes. [16:52] brendand, if TheMuso does not fix it tonight I'll fix it tomorrow, since it seems to hit a lot of people [16:53] brendand, I think the same as pgraner, that it was related to ALSA 1.0.25 [16:53] diwic: does it apply tp flash stuff too? [16:53] ah k, you just answered [16:53] ppisati, it applies to anything using the alsa-plugin bridge to pulseaudio [16:53] ppisati, which Flash does. As for other apps, it differs. [16:54] diwic - i'm guessing it's a duplicated then. do you have a master bug number? [16:54] brendand, no, just saw it on the mailinglist. [16:55] diwic, ok. thanks! [16:55] (and can replicate here, I believe) [17:17] cking: i happened to be running inotifywait -m -r / for something last night and noticed ibus opens and reads an svg every time the focus changes, lots of icons are read all the time and .cache/indicator-applet*.log is written to pretty often, if you're still looking for things that wake up the disk there might be something to see [17:28] apw: hrm thanks [17:31] ohsix, can you file a bug and add "Also affects project" to ubuntu-power-consumption - that way it will get some love and attention [17:32] hm maybe, i'm on natty though; don't know how easily i can test the latest versions of things, just thought it was worth mentioning as a source of more data to look at [17:33] plus the .svg reads even if they're redundant should hit the disk cache, unless you're already under memory pressure then it just makes the situation a tinytinytiny bit worse :p [17:38] ohsix, thanks, lemme put that down on a list of things to eyeball next [17:39] before latencytop got a fancy ui it'd pretty readily tell you about file access [17:39] apw: what is the status of bug 882147? [17:39] Launchpad bug 882147 in linux "overlayfs does not implement inotify interfaces correctly" [High,Triaged] https://launchpad.net/bugs/882147 [17:39] yep, some tools just overstep themselves [17:40] bdmurray, in progress, its not entirly obvious we can fix it at the moment [17:40] bdmurray, i've tried telling userspace that inotify doesn't work on that filesystem and that just makes it puke errors violently [17:40] apw: so inotify might not be available then? [17:40] it still says "press F to" check out the file in latencytop (paraphrasing) but it doesn't do anything; they just didn't change the string :P [17:41] bdmurray, i just don't know at the moment, its much harder than i had hoped thats for sure [17:41] i am about to start an upstream converation on what the heck we can do [17:41] apw: okay, I'm really interested in seeing that work [17:42] bdmurray, yep, and its not clear that the current structure lets that work === jsalisbury_ is now known as jsalisbury [18:03] o [18:03] ogasawara, ok i've managed to reproduce some kind of panic with the procps changes, though not until i tried to change the options ... it booted just fine for me ... [18:04] though of course it may depend what you have installed on the system in question [18:04] apw: I'd tested with the latest precise and what was on master-next [18:05] ogasawara, yeah, there is definatly somethign screwey, just it didn't reproduce instantly for me, but i am looking at this problem and that may be what you hit [18:05] apw: ack. let me know if you want me to help test anything. [18:05] ogasawara, so once i've fixed it i'll get you to retest :) [18:05] * ogasawara nods [18:06] * ogasawara back in 10 [18:07] ogasawara, ahh this is completley broken, and only luck lets me boot if anyone looks in /proc/mounts [18:32] ogasawara, ok sorted i think, i've now pushed out some replacement patches fixing the issue i think you were seeing [18:32] apw: cool, I'll pull and test [18:32] ogasawara, basically any boot was a random tossup, so your positive testing was just as much luck as mine [18:33] ogasawara, a change of parameters to the show_options superblock options [18:34] * tgardner -> lunch [19:12] * cking -> EOD [19:16] * tgardner relocates === tgardner is now known as tgardner-afk [19:17] bdmurray: you see any recent breakage wi LP api? on Precise, task.date_created seems to be returning 'unicode' object instead of 'datetime' object [19:17] bdmurray: doesn't have to be LP api. there are several places it could be broken [19:18] bjf: haven't noticed [19:18] bdmurray: i don't run LP scripts on precise very often so I'd not seen it [19:19] bjf: just tested and yeah it looks like unicode [19:19] * bdmurray cries a little [19:20] apw: test machine boots fine with the latest set of patches (have done 3 reboots so far, all successful) [19:21] ogasawara, sounds good, it was an obvious flaw when i found it [19:22] apw: how urgently do we need these to go up, do we need an upload today or can I wait till tomorrow/fri? [19:23] ogasawara, no urgency really, we need them so security can test whether they are enough of a fix, but they arn't going to get to that in short order, and if they do, well they can make their own [19:24] ogasawara, just wanted them on their way [19:24] apw: ack [19:25] tgardner-afk, just found a new shortcut in unity launcher (not listed on the help) super-tab, does alt-style round your launchapplication [19:29] bjf, jsalisbury, i note that the meeting summary that got sent out this time is missing the links i put in ... am i doing them in the wrong format ? [19:30] apw, hmm. Let me take a look. [19:32] apw, your format during the meeting is ok. I must have improperly formatted the moin file. [19:43] jsalisbury, well i'd not sweat about sending out another version or anyting, just see if we can figure out what occurs for next time [19:44] apw, sure thing. I'll be sure it's fixed for next meeting. The script to generate the moin file stripped the links, but I'll fix it. [19:59] * apw calls it a day ... have fun [20:22] http://paste.ubuntu.com/834410/ [20:22] ^^^^^ [20:23] make-kpkg broken? [20:23] apw: super + ` was a cool find for me :) [20:23] or does it not work with mainline sources? [20:23] apw: same w/ apt + ` [20:23] er alt [20:25] this is holding up my work, can someone look at how make-kpkg is not working? [20:32] hmm, this time it seemed to work, i guess nvm [23:31] vanhoof, what does super+` do for you, doesn't seem to be a key binding here ?