| bryce | sforshee, heya bug #1166442 has been fixed by a patch in linus' tree posted last week. I stuck the patch on the bug report. I've verified it on my samsung laptop backported to the 3.5 kernel; issue affects pretty much all recent samsung laptops I gather. | 06:52 |
|---|---|---|
| ubot2` | Launchpad bug 1166442 in linux (Ubuntu Saucy) "Elantech clickpad/touchpad lacks multitouch features." [Low,Confirmed] https://launchpad.net/bugs/1166442 | 06:52 |
| === smb` is now known as smb | ||
| ppisati | indicators today are in the correct position | 07:19 |
| ppisati | but when i click on the volume one, nothing pops up | 07:19 |
| ppisati | so i can't change the volume level | 07:19 |
| smb | Youre lucky sw update worked | 07:20 |
| ppisati | or any other audio-related thing | 07:20 |
| ppisati | :( | 07:20 |
| ppisati | actually NO indicators work today :( | 07:23 |
| * ppisati -> reboot | 07:23 | |
| * apw yawns | 07:33 | |
| * smb waits on RAOF mentioning some sort of bees | 07:38 | |
| * apw realises he isn't on irc :) [sic] him away | 07:39 | |
| apw | wtf | 07:39 |
| apw | that is like old ... | 07:40 |
| apw | the bees have takne him away | 07:40 |
| smb | giant bees | 07:40 |
| cking | smb, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1065434 | 07:41 |
| ubot2` | Ubuntu bug 1065434 in linux (Ubuntu Raring) ""unregister_netdevice: waiting for lo to become free. Usage count = 1" after LXC container shutdown" [High,Fix released] | 07:41 |
| smb | apw, and yous is not on mumble (again). Quick finish breakfast and get dressed before the doorbell rings | 07:41 |
| smb | cking, Ah, thanks... now lets see | 07:42 |
| cking | smb, I recall now it was a pain to test because it needed many multiple iterations before I found a way to reproduce it reliably | 07:43 |
| smb | cking, Yeah, the new one mentions "after 6000 or so launches and teardowns..." | 07:43 |
| smb | bug 1196295 | 07:44 |
| ubot2` | Launchpad bug 1196295 in lxc (Ubuntu) "lxc-start enters uninterruptible sleep" [High,Confirmed] https://launchpad.net/bugs/1196295 | 07:44 |
| cking | urgh | 07:44 |
| cking | hrm, very similar looking, but I do wonder if it's a different issue, I guess it may be, hard to tell | 07:46 |
| smb | Feels a bit like something that happens after many iterations which then prevents proper cleanup... but really hard to tell | 07:48 |
| cking | 6000+ iterations is a killer to bisect with | 07:48 |
| smb | And even if you could in theory bisect, my experience with network changes is that they make ones life of bisecting even harder | 07:50 |
| cking | double urgh | 07:50 |
| apw | smb, or lots of them | 07:53 |
| apw | smb, nope not expecting any builders today, thankfully | 07:53 |
| smb | apw, The fun with the doorbell is that it does even ring when you are not *expecting* it to | 07:54 |
| smb | :) | 07:54 |
| smb | And yes, lots of changes and lots of dependencies between | 07:54 |
| apw | as long as they are bringing fun things for me, that is fine | 07:55 |
| smb | Yeah, the bills they just put into the letter box (strangely) | 07:55 |
| === fmasi_afk is now known as fmasi | ||
| * smb is glad to see that the software-center maintains a consistent level of quality | 08:04 | |
| RAOF | Bumblebees! | 08:04 |
| ppisati | brb | 08:44 |
| apw | RAOF, swarms of uber-bumble-bees | 08:44 |
| * ppisati -> out for lunch | 10:45 | |
| === fmasi is now known as fmasi_afk | ||
| === fmasi_afk is now known as fmasi | ||
| psivaa | jdstrand: thanks for the fix for bug 1202203, appears to solve the issue. 20+ runs did not see the bug with today's images. | 11:48 |
| ubot2` | Launchpad bug 1202203 in isc-dhcp (Ubuntu) "Connection requests to saucy server VMs from a hosts fail after fresh preseeded VM installs" [High,Fix released] https://launchpad.net/bugs/1202203 | 11:48 |
| === ara is now known as Guest69887 | ||
| === fmasi is now known as fmasi_lunch | ||
| nessita | hello. Is there anyone available to help me with a potential kernel sound bug? LP: #1201528 | 12:38 |
| ubot2` | Launchpad bug 1201528 in alsa-driver (Ubuntu) "[Realtek ALC889] - Audio Playback Unavailable" [Undecided,New] https://launchpad.net/bugs/1201528 | 12:38 |
| rtg | nessita, you might try restarting pulseaudio when it gets in that state, e.g., http://www.linuxplanet.com/linuxplanet/tutorials/7130/2 | 12:50 |
| nessita | rtg, yeah, but it does not solve it | 12:50 |
| nessita | rtg, the sound gets "dead", and nothing fixes it but a restart | 12:51 |
| rtg | nessita, the fact that audio output lasts for many minutes before stopping kind of indicates that its a pulse issue. diwic might be able to offer some thoughts. | 12:51 |
| nessita | rtg, happy to provide any log that could be useful -- currently I can not do any meeting in this computer, all mumble/skype/hangouts will "break" | 12:52 |
| nessita | rtg, I was suggested by the support team that this could be a buffering issue | 12:52 |
| rtg | nessita, possibly, but I see no indication in your logs that its a kernel problem. without that there isn't much to go on. | 12:53 |
| diwic | nessita, this is worth trying https://wiki.ubuntu.com/Audio/PositionReporting | 12:53 |
| nessita | diwic, looking | 12:54 |
| nessita | trying with "options snd-hda-intel position_fix=1" and rebooting, brb | 12:57 |
| === ssweeny` is now known as ssweeny | ||
| nessita | diwic, hello, I'm back, So, audio worked for a while and stopped working again. I think I noticed the following pattern: | 13:08 |
| nessita | if I adjust volume levels using pavucontrol, it usually ork | 13:09 |
| nessita | works* | 13:09 |
| nessita | but if I use the control dialog below the sound icon in the indicator menu, it breaks immediatly | 13:09 |
| diwic | nessita, hmm | 13:10 |
| nessita | diwic, I alos note that the device for inout gets constantly reset to the S/PDIF option: I keep chooise the "regular" one (Front Microphone), but every time I close and open the volume control app, the input device is set to the S/PDIF one | 13:11 |
| nessita | also* | 13:11 |
| === fmasi_lunch is now known as fmasi | ||
| nessita | diwic, want me to try with "options snd-hda-intel position_fix=2" noe? | 13:12 |
| nessita | now* | 13:12 |
| nessita | puf, I'm typing pretty bad today | 13:12 |
| diwic | nessita, what have you tried so far? both "position_fix=1", "position_fix=2" and "tsched=0" is worth a try | 13:13 |
| diwic | (and a reboot between each one) | 13:13 |
| nessita | diwic, right, so far only tried the first one, but I wanted to mention the "pattern" (audio does not break until I open the app in the sound indicator, and input device gets "reset") | 13:14 |
| nessita | diwic, will try the other two | 13:14 |
| * nessita reboots | 13:14 | |
| nessita | diwic, hello again! well, seems like position_fix=2 fixed it (well, at least I tried all I tried before and the audio playblack kept working) | 13:27 |
| nessita | will comment on the bug | 13:28 |
| diwic | nessita, all right, interesting | 13:29 |
| nessita | diwic, FYI, the app below the sound indicator keeps "resetting" the input device to the S/PDIF one, but changing it to the Front microphone allows me to use the mic (otherwise mic does not work) | 13:30 |
| diwic | nessita, you mean the gnome sound settings? | 13:31 |
| nessita | diwic, the app title is "Sound", but I think is that, yes | 13:32 |
| nessita | is the section under All settings -> Sound | 13:32 |
| === psivaa is now known as psivaa-afk | ||
| nessita | diwic, ping again, when you can. So, audio died, not sure when. | 13:53 |
| nessita | diwic, any more ideas what can I try? | 13:54 |
| diwic | nessita, well, the tsched=0 one. Remove the nonworking position_fix first | 13:55 |
| nessita | diwic, ack, will do | 13:56 |
| nessita | diwic, just to be sure, the line I have in /etc/pulse/default.pa is "load-module module-udev-detect use_ucm=0" | 14:00 |
| nessita | (the wiki says "load-module module-udev-detect" without the use_ucm=0) | 14:00 |
| === fmasi is now known as fmasi_afk | ||
| === fmasi_afk is now known as fmasi | ||
| diwic | nessita, you can keep the use_ucm=0 too, it doesn't matter much | 14:01 |
| nessita | ack | 14:01 |
| diwic | for your hardware | 14:01 |
| nessita | ok, so, reverted options snd-hda-intel position_fix=2 and added the tsched=0, rebooting | 14:01 |
| rtg | ppisati, can you install https://launchpad.net/~canonical-kernel-team/+archive/ppa/+build/4806723 on calxeda as a test ? | 14:09 |
| ppisati | rtg: i'll do | 14:09 |
| nessita | diwic, so far working, will report back if audio playback stops working | 14:11 |
| diwic | ok | 14:11 |
| diwic | nessita, you're using standard stereo profile, right? Nothing surround related? | 14:12 |
| nessita | diwic, standard stereo setup, and in particular nothing changed in my setup since precise (where audio worked with no glitches) | 14:13 |
| nessita | diwic, not sure if it's worth noting that the "only" thing that breaks is audio playback, mic keeps working without issues (even if I can not hear anything) | 14:13 |
| diwic | ok | 14:13 |
| diwic | nessita, yeah, every time we try to improve the stack for some set of users, it's easy to break for another, as we don't have the ability to test > 1000 machines every time something is changed | 14:14 |
| rtg | apw, overlayfs is the only thing standing between me and switching over to 3.11-rc2 in the archive. Is that readdir issue still on your todo list ? | 14:15 |
| apw | rtg, yes on todays indeed | 14:15 |
| rtg | or, are you gonna make _me_ learn it ? | 14:15 |
| apw | heh no, i wouldn't do that | 14:15 |
| rtg | prolly wouldn't hurt | 14:16 |
| rtg | well, it would _hurt_ a lot, but I should learn it | 14:16 |
| apw | rtg, next time | 14:17 |
| ppisati | [ 12.075144] RAMDISK: Couldn't find valid RAM disk image starting at 0. | 14:23 |
| ppisati | [ 12.082461] VFS: Cannot open root device "UUID=608a7f64-20e8-400f-97b0-fb34d06bf327" or unknown-block(0,0): error -6 | 14:23 |
| ppisati | [ 12.092984] Please append a correct "root=" boot option; here are the available partitions: | 14:23 |
| ppisati | rtg: ^ | 14:23 |
| ppisati | rtg: uhmm.... | 14:23 |
| rtg | ppisati, FIIK | 14:24 |
| ppisati | rtg: that's weird | 14:24 |
| ppisati | rtg: it's the same node as yesterday | 14:24 |
| ppisati | rtg: so that UUID should be valid | 14:25 |
| rtg | ppisati, missing device driver ? It _is_ a completely different kerenel version | 14:25 |
| ppisati | rtg: well, there're two weird things: | 14:25 |
| ppisati | 1) [ 3.035132] RAMDISK: Couldn't find valid RAM disk image starting at 0. | 14:26 |
| ppisati | 2) the uuid thing | 14:26 |
| ppisati | actually 2 could be a consequence of 1 | 14:26 |
| rtg | ppisati, as I am totally distracted by some other issues, I'll leave this problem in your capable hands. | 14:26 |
| rtg | :) | 14:27 |
| ppisati | rtg: ack | 14:27 |
| rtg | at least the tool chain seems functional | 14:27 |
| apw | rtg, just pushed a rules change to master-next and unstable to make a new udeb for use in the seeds | 14:35 |
| rtg | apw, did you catch the unstable rebase I pushed just a few minutes ago ? | 14:36 |
| apw | yeah i fetched 10s before | 14:36 |
| rtg | apw, cool | 14:36 |
| apw | and when i pushed it was linear | 14:36 |
| apw | infinity, i've commited that udeb change for saucy both ... for whenever we upload that | 14:37 |
| infinity | apw: Feel the urge to push that to Ben's branch too? | 14:38 |
| apw | infinity, will do in a bit indeed | 14:38 |
| infinity | (If not, I'll talk to him about it $later) | 14:38 |
| hallyn | apw: hey - have you used git-send-email through smtp.canonical.com recently? I thought it used to work (with the expected obvious arguments) but i'ts failing me right now... wondering if htere are any new tips and tricks | 14:56 |
| apw | hallyn, i think i did it a couple of days ago, nothing different then that i know of | 14:57 |
| hallyn | hm, ok. actually maybe this is a bug in the saucy git-send-email. it also won't honor --smtp-debug | 14:57 |
| apw | hallyn, oh, no i have switched to gmail | 14:57 |
| hallyn | just keeps telling me to use --smtp-debug :) | 14:57 |
| hallyn | oh, right. | 14:58 |
| hallyn | i've been thinking maybe i'll give in and do that. | 14:58 |
| hallyn | though i figure they'll just disable imaps access soon now that jabber and rss are gone :) | 14:58 |
| apw | i think we are paying for it, so they had better not | 14:59 |
| hallyn | all right - thanks. i guess i'll just send by hand and get on with actual work | 14:59 |
| hallyn | ah - was specifying port 587 without thinking - should be 465 (or just unspecified) | 15:03 |
| rtg | apw, infinity: so you're gonna use linux-udebs-generic in the debian installer ? | 15:08 |
| apw | rtg, in the seeds | 15:09 |
| apw | rtg, so a simple fix for overlayfs for the moment might be to revert the removal of readdir | 15:14 |
| apw | and then we can work on the replacement or wait for it from upstream overlayfs if it is coming | 15:15 |
| apw | anyhow i am working on it | 15:15 |
| === cyphermox_ is now known as cyphermox | ||
| nessita | diwic, so, bad news. Audio playback no longer works. More info: as per the debugging I was first requested in the bug, I had set pulseaudio to autospawn = no, so I could gather more verbose logging (which is attached to the bug). Noticing that the latest attempt may have worked, I resetted that setting (ie, I removed the .pulse/client.conf) and restarted the computer. Now, opening more than one thing that access pulseaudio (mplay | 15:16 |
| nessita | diwic, another debugging info is that now, mplayer can not play a video, it gets absolutely blocked | 15:16 |
| nessita | (ie, the first video frame appears and then nothing else happens) | 15:17 |
| diwic | nessita, okay. I'm afraid I don't have time to help you right now, and I don't have any good ideas either. | 15:17 |
| nessita | diwic, do you think that moving to saucy could fix this? | 15:17 |
| diwic | nessita, you could also try https://wiki.ubuntu.com/Audio/UpgradingAlsa/DKMS | 15:18 |
| rtg | apw, is the removal atomic ? I thought it touched a bunch of file system drivers, | 15:18 |
| diwic | nessita, that would be even better to try than saucy | 15:18 |
| apw | rtg, he added the new interface, moved everything over to the new one, and removed the old | 15:18 |
| apw | so in theory we could revert the last bit and be ok for now, maybe | 15:18 |
| apw | i'll have a play with it | 15:18 |
| rtg | ok | 15:18 |
| apw | ahh no he may have removed the infrastructure we use as welll ... hmmm well see | 15:19 |
| nessita | diwic, will try that, thanks. | 15:19 |
| nessita | diwic, shall I remove the "load-module module-udev-detect tsched=0" tweak? | 15:20 |
| diwic | nessita, yes, remove it | 15:20 |
| nessita | ack | 15:20 |
| nessita | diwic, so, if you have more ideas what else could I try, I will appreciate those, since even after updating dkms my audio playback dies. I know you mentioned you can't help me debug now, but perhaps you can indicate another person that could help or guide me? | 15:37 |
| nessita | I have pulseaudio set to be verbose (at system level), I keep getting these 3 lines in syslog: | 15:38 |
| nessita | Jul 19 12:38:37 pulseaudio[2636]: last message repeated 10 times | 15:38 |
| nessita | Jul 19 12:38:37 dali pulseaudio[2636]: [alsa-sink] ratelimit.c: 15 events suppressed | 15:38 |
| nessita | Jul 19 12:38:37 dali pulseaudio[2636]: [alsa-sink] alsa-sink.c: Underrun! | 15:38 |
| diwic | nessita, yeah, the "Underrun!" is unusual actually | 15:39 |
| diwic | nessita, does it happen only with low latency - sound settings, mumble, skype etc, if you only use say some music player, does it still crash? | 15:40 |
| nessita | diwic, I use sonata which is a GUI client for mpd, the music server. I never got those to work since I updated to raring | 15:40 |
| ppisati | rtg: what's happended to linux-lts-raring - 3.8.0-27.40~precise2? i see there's a new one in the pipe (~precise3), any chance i can get the previous one? | 15:40 |
| diwic | nessita, I wonder if the "Underrun!" is due to wrong position reports or system scheduling issues | 15:41 |
| nessita | diwic, I even tried uninstalling those two (sonata and mpd), and using rhythmbox, but audio playback eventually dies no matter what | 15:41 |
| rtg | ppisati, infinity re-uploaded to use an orig tarball. dunno if the original still exists. | 15:41 |
| diwic | nessita, hmm, okay | 15:41 |
| diwic | nessita, for further troubleshooting I guess you would need a special pulseaudio build with DEBUG_TIMING on | 15:42 |
| nessita | diwic, I'm not sure about "wrong position reports or system scheduling issues", I don't know what those mean :-) | 15:42 |
| nessita | happy to try saucy this weekend, though I see pulseaudio version (mayor) has not changed | 15:43 |
| diwic | nessita, we have PA 4.0 in a PPA, but I doubt that would make a difference | 15:48 |
| === rtg is now known as rtg-afk | ||
| diwic | nessita, https://launchpad.net/~ubuntu-audio-dev/+archive/pulse-testing <-- PA 4.0 is here, and you're welcome to test it, but I doubt it'll help | 15:49 |
| nessita | diwic, your doubts makes me doubtful :-) | 15:49 |
| nessita | diwic, not sure if this helps, but is the output of pulseaudio log in syslog when I try to play a video (audio playback does not work and video speed is off, frames are shown too fast, something like 1.5x) https://pastebin.canonical.com/94733/ | 15:49 |
| nessita | without understanding the output that much, lines that call my attention are "Scheduling delay of 6.33ms, you might want to investigate this to improve latency..." | 15:50 |
| nessita | "Have to rewind 65276 bytes on render memblockq" | 15:50 |
| diwic | nessita, yeah, scheduling messages with >= 20 ms is not optimal, I saw you had one of those, but not enough to cause constant problems | 15:54 |
| nessita | diwic, since you're busy, in order to stop interrrupting you, is there someone else I can ping for debugging further? | 15:55 |
| diwic | nessita, not really...unfortunately | 15:57 |
| nessita | diwic, would it be better if I ping you again next week? I kind of need audio to work (meetings and all) | 15:59 |
| diwic | nessita, what time zone are you in? | 16:01 |
| nessita | diwic, GMT-3 (is 1pm for me now( | 16:01 |
| nessita | )) | 16:01 |
| diwic | nessita, hmm, that sounds like Brazil or something | 16:02 |
| nessita | diwic, yeah, Argentina | 16:02 |
| diwic | nessita, anyway, yeah, try pinging me on Monday. You could try TheMuso too, but he's in Australia so I doubt you and him will find a time that works for you both | 16:03 |
| nessita | diwic, will do, thanks for your time today | 16:04 |
| * nessita reboots one more time | 16:05 | |
| diwic | nessita, yw | 16:06 |
| nessita | diwic, one more interrupt (last one!). When rebooting the computer (first reboot since I installed the new dkms), computer will not shutdown, had to sysrq+b to have it rebooting. Error shown in screen was http://ubuntuone.com/5crDl5mdauSvTF0BjnQrQk | 16:19 |
| * apw breaks out a beer | 16:21 | |
| apw | rtg-afk, t | 16:22 |
| apw | rtg-afk, t | 16:22 |
| apw | rtg-afk, this is going to take some porting me thinks, now do remember if aufs is working that will be used anyhow, so we don't need overlayfs per-se. so you might upload it to the PPA so that the dkms people can start their grinding | 16:22 |
| === fmasi is now known as fmasi_afk | ||
| jsalisbury | apw, rtg-afk, do you think this one should be hot listed: bug 1202887 | 16:33 |
| ubot2` | Launchpad bug 1202887 in linux-manta (Ubuntu) "'binder: RLIMIT_NICE not set' when using binder from the ubuntu side" [Medium,New] https://launchpad.net/bugs/1202887 | 16:33 |
| ppisati | ubuntu@c16:~$ uname -a | 16:36 |
| ppisati | Linux c16 3.10.0-4-generic #13~precise1-Ubuntu SMP Thu Jul 18 23:37:03 UTC 2013 armv7l armv7l armv7l GNU/Linux | 16:36 |
| ppisati | ubuntu@c16:~$ lsb_release -a | 16:36 |
| ppisati | No LSB modules are available. | 16:36 |
| ppisati | Distributor ID: Ubuntu | 16:36 |
| ppisati | Description: Ubuntu 12.04.2 LTS | 16:36 |
| ppisati | Release: 12.04 | 16:36 |
| ppisati | Codename: precise | 16:36 |
| ppisati | ubuntu@c16:~$ | 16:36 |
| ppisati | rtg-afk: ^ | 16:36 |
| ppisati | rtg-afk: the new vmlinuz was too big and was overwriting part of initrd (that resulted in garbage) | 16:36 |
| ppisati | rtg-afk: shuffling the things around, made it boot | 16:36 |
| ppisati | rtg-afk: but i still need to tweak it to make it automatic now | 16:37 |
| * ppisati -> EOW | 16:40 | |
| === rtg-afk is now known as rtg | ||
| rtg | jsalisbury, I'll have a look at bug #1202887. No need to hot list it. | 18:02 |
| ubot2` | Launchpad bug 1202887 in linux-manta (Ubuntu) "'binder: RLIMIT_NICE not set' when using binder from the ubuntu side" [Medium,New] https://launchpad.net/bugs/1202887 | 18:02 |
| jsalisbury | rtg, thanks | 18:03 |
| apw | jsalisbury, we should keep an eye yes | 19:42 |
| rtg_ | apw, do you understand that bug ? 'cause I'm not seeing the problem. | 19:43 |
| rtg_ | on the other hand, nice seems unreasonably complicated. | 19:44 |
| rsalveti | rtg: apw: bug 1203173, in case you guys got a maguro as well | 20:09 |
| ubot2` | Launchpad bug 1203173 in touch-preview-images "[maguro] broadcom dongle host driver sometimes fails to load the correct mac address" [Medium,Confirmed] https://launchpad.net/bugs/1203173 | 20:09 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!