/srv/irclogs.ubuntu.com/2010/10/08/#ubuntu-kernel.txt

=== Sgiomlaireachd is now known as away_Sgiomlairea
* away_Sgiomlairea is now away - Reason : dinner00:10
=== jjohansen is now known as jj-afk
=== away_Sgiomlairea is now known as Sgiomlaireachd
* Sgiomlaireachd is no longer away : Gone for 2 hours 34 minutes 41 seconds02:44
=== amitk-afk is now known as amitk
=== ivoks is now known as ivoks-afk
lucentping?07:30
JFolucent, is there anyone in particular you want to ping? If you have a question, just ask. 07:33
lucentJFo: thanks, when I disconnect a firewire device from one of my controllers, all controllers stop working correctly07:37
lucentdmesg is kind of quiet though, I only see the storage layer complain07:37
lucenthow do I get more info about firewire? 07:38
lucentcrickets.  hence the ping?08:00
laglucent: Can you paste me the output of lsmod at paste.ubuntu.com please?08:47
lucentlag: one moment, will do08:48
lucentlag: http://paste.ubuntu.com/508598/08:49
laglucent: Obviously with the Firewire controller up and working08:49
lagOkay, that's good08:50
lagWait one08:50
lucentI have a ricoh chipset that's built into the laptop, and an expresscard with a TI chipset08:51
lucentsince using the new firewire stack my problems are data corruption on either one or both, after removing a storage device from either firewire bus08:52
lucentold firewire stack never showed these problems for me.08:52
laglucent: Issue: echo 1 | sudo tee -a /sys/module/firewire_ohci/parameters/debug08:52
lucentlag: okay, no output08:53
lagTry: echo -1 | sudo tee -a /sys/module/firewire_ohci/parameters/debug08:54
lucentoutput "-1"08:54
lagIf that doesn't work paste me the result of "ls /boot/config-`uname -r`"08:56
lucent/boot/config-2.6.35-22-generic08:56
lagYes, cat that file and paste me the output08:57
lucentoh okay! 08:57
lagBy the way, when you issued "echo -1 | sudo tee -a /sys/module/firewire_ohci/parameters/debug" did you then unplug your Firewire device and check your logs?08:57
lucentlag: http://paste.ubuntu.com/508602/08:58
lagAll that command does is place the Firewire module into debug mode08:58
lucentand no sir, I didn't know that I should... 08:58
lagSorry, I assumed too much :)08:58
lagThat's what you need to do08:58
lagLet me check your .config08:58
lagThat's good - you have CONFIG_FIREWIRE_OHCI_DEBUG=y08:59
lucentlag: I am willing and motivated to reproduce this bug again with firewire debug on, take logs, whatever but I'm just needing some help like you have said to get that information :)09:00
lagdmesg should be sufficient 09:01
lucentdmesg at http://paste.ubuntu.com/508603/09:02
lagAnd now you have your output :)09:02
lagIs there anything else I can help you with? ;)09:03
lucentthat's fantastic09:04
lucentin theory should I be able to unload the firewire modules ?09:04
lucentI've tried and it led to kernel panic09:04
lagAre you able to paste the panic?09:05
lucentI don't know that it will panic again if I do so again this time, but is firewire supposed to be unloadable at runtime?09:05
lucent"so... would it be a bug if I did get a panic?" is my question 09:05
lagYes09:05
lagAny panic is a bug09:05
lucentgood to know, then.09:06
lucentthank you for the advice. I will go try to catch this bug and report 09:06
lagThank you09:06
lagInsure to attach the pastes you've sent me09:07
lagThey would be helpful09:07
JFowell, I updated my Maverick box and everything that was working nicely has gone to crap now10:29
* ikepanhc try to remember when I pushed the upgrade button...10:36
lagJFo: Such as?10:36
ikepanhcah, I updated 7hr ago...10:36
JFogeneral sluggishness10:39
JFodiagnosing now10:39
=== smb` is now known as smb
JFolooking like it may be rhythmbox badness10:41
JFoas that is the last of the items to test across all of my trials10:42
JFodoing a final test to see10:45
JFoyep, rhythmbox is effing hosed10:49
JFoplaying through a song and suddenly it stops playing. go to the interface and it appears to be set to play, but clicking the pause button causes the song to start for a second then go to the next song10:50
JFowhile looking, for all intent and purpose as if it were paused10:50
JFowhich explains some of the other oddities I have been seeing tonight10:51
=== diwic is now known as diwic_afk
lagJFo: I thought you used musica11:17
tgardnersmb, can you look at bug #644452 ? It appears that rfkill is the root of this guys problem. Had the HP platform driver changed from Lucid to Maverick?11:22
ubot2Launchpad bug 644452 in linux (Ubuntu) "Atheros AR928X not connecting on wireless HP G60-468CA (affects: 10) (dups: 3) (heat: 253)" [Undecided,New] https://launchpad.net/bugs/64445211:22
smbtgardner, Let me have a look. I haven't followed that driver11:23
JFolag, pete uses musica11:27
lagAh, okay11:27
=== diwic_afk is now known as diwic
avruserHi, people!12:00
avruserI have a problem with debugging kernel crash on 8.04.4.12:04
avruserI got crash dump with kdump, and trying to debug it with crash utility.12:05
avrusercrash utility exit with error:12:06
avruserplease wait... (gathering module symbol data)a)12:06
avrusercrash: invalid kernel virtual address: 4a2001  type: "pgd page"12:06
lagavruser: What kind of crash are you trying to debug?12:09
avruserKernel panic - not syncing: Fatal exception in interrupt12:13
lagOkay, can you past your dmesg at paste.ubuntu.com please?12:14
avruserok12:14
avruserhttp://paste.ubuntu.com/508708/12:20
lagI don't see a panic?12:20
avruserthis is current dmesg12:21
lagPerhaps I should have been more clear :)12:22
lagCan you paste the panic?12:22
lagIf it's not in the current dmesg, it will most likely be in /var/log/kern.log or /var/log/messages12:23
lag(or both)12:23
avruserok12:24
lagThanks12:24
avruserPanic was not logged in any log files, but at the moment of crash on the monitor appears this messages:12:32
avruserCode: Bad EIP value12:32
avruserEIP: [<0000 0000>] 0x0 SS:ESP 0068:c043bf1412:32
avruserKernel panic - not syncing: Fatal exception in interrupt12:32
avruserhttp://paste.ubuntu.com/508716/12:36
avruserpart of kern.log at time of crash12:36
lagavruser: It's hard to say whether that last line is related to the crash12:44
lagavruser: I need the lines prior to "Code: Bad EIP value" to make any sense of it12:45
lagavruser: Did it dump the at any point? A list of function names12:45
lagavruser: Have you changed any hardware recently? A Bad EIP value can be related to dodgy hardware; NICs, ATA Disks etc12:47
avruserI was used bluetooth dongle12:48
lagTry and use it again12:48
lagSee if you can replicated the issue - narrow it down to a piece of hardware12:48
lagBad RAM can cause the EIP issue too12:48
avruserI can replicate this crash12:49
lagHow?12:49
lags/Did it dump the at any point/Did it dump the stack at any point12:50
avruserIt happens then i pass voice through the usb bluetooth dongle12:50
lagShouting at it?12:51
lagI'd kernel panic if you shouted at me too :)12:51
avruser:)12:52
lagI'd need more information if I was going to rectify this12:52
lagWhat were the lines preceding "Code: Bad EIP value"12:52
avruserOK, I need to connect monitor to my server12:53
lagavruser: Is there any solid reason why your server is still running Hardy?12:57
avruser'cause LTS13:10
lagSo is Lucid :)13:10
avruserI need very stable system13:10
avruserhardy got many patches13:11
lagDid you install dahdi yourself?13:11
avruseryes13:11
avruserI got some trace from my screen:)13:11
lagGood, good13:12
apwavruser, the number of patches is pretty much proportional to age of the release13:12
lagavruser: I think people would be even more inclined to help you if you didn't have such an old OS, despite the fact that we're still supporting it13:13
lagBesides, I think your problem my lie in dahdi13:13
lagI've just installed it13:13
lagGoing to have a play13:13
avruserI experience this problem only when i make call through the chan_mobile13:14
avruserthis module use bluetooth stack13:15
lagHow can I do that?13:16
lagDo I need specific hardware to do that?13:16
lagI have a bluetooth enabled phone and a dongle?13:16
avruserThis enough13:16
avruserHave you chan_mobile compiled?13:17
lagI also have dahdi installed13:17
avruserFrom asterisk-addons 1.6.213:17
lagIn the repos? Or from source?13:17
lagJust installed asterisk-1.6.213:18
avruser8.04.4 have not asterisk-1.6 and addons-1.6 in off repo13:20
avruserI have installed from source13:20
lagI can't see addons13:20
lagBut the asterisk-1.6 is in the Lucid repos13:20
lagI've just installed it13:21
lagNow how do I reproduce this bug?13:21
lagConnected to Asterisk 1.6.2.5-0ubuntu1 currently running on computer (pid = 15415)13:22
avruserDo you have chan_mobile?13:23
avruserasterisk -r13:23
lagI did that13:23
lagI have CLI now13:23
lagcomputer*CLI> channel 13:24
lagoriginate  redirect   request 13:24
avrusermodule show like chan_mobile.so13:24
lagWhere?13:24
avruserin CLI13:24
lagcmd?13:24
avruseryes13:24
lagsystem1*CLI> chan_mobile.so13:25
lagNo such command 'chan_mobile.so' (type 'core show help chan_mobile.so' for other possible commands)13:25
avruser'module show like chan_mobile.so' command13:25
avruserwhat version of kernel?13:26
lagIt would help if the modules were listed in lexicographical order13:27
lagchan_phone.so                  Linux Telephony API Support              013:27
avruserDid you installed asterisk-addons?13:28
laghttp://paste.ubuntu.com/508753/13:28
lagNo13:28
lagIt wasn't there13:28
laghttp://paste.ubuntu.com/508754/13:30
lagPerhaps asterisk-mobile13:30
lagavruser: Nope13:31
avruseryes, maybe 13:31
lagIt's still not there13:32
lagAre you sure it's not been renamed to -phone?13:32
avruserdid you installed asterisk-mobile?13:33
lagYes13:33
avruserrestart asterisk?13:34
lagWhy don't you email me the module?13:34
lag/usr/lib/asterisk/modules/chan_mobile.sh13:34
lag/usr/lib/asterisk/modules/chan_mobile.so13:34
avruserno problem13:34
avruserlag: can we complete this tomorrow?13:35
lagWe can continue on Monday13:35
lagI might be around tomorrow13:35
lagBut I can't promise 13:35
avruserThanks:)13:36
lagCan you email me the commands that you use to replicate this bug13:36
avrusersure13:37
avruserGood Bye!13:38
lucenthow do I get the full dmesg output?  man page for dmesg hints at -s option, however I can find no example of this14:01
=== diwic is now known as diwic_Afk
=== diwic_Afk is now known as diwic_afk
* ericm|ubuntu reconnects14:17
vanhoofogasawara: \o/15:06
vanhoofogasawara: you made my week15:06
vanhooflol15:06
vanhoofprogress++15:06
ogasawaraheh15:06
ogasawaravanhoof:  cking did all the leg work testing :) , but it indeed sounds promising15:07
lagIs there a (ex or currently) release manger around that can spare me a couple of minutes? 15:08
ogasawaralag: as in the entire distro release manager? or kernel release manager? or ....?15:08
cooloneylag, kernel release manager = ogasawara15:09
lagAnyone with release manager permissions on LP :)15:10
lagYes, I know that cooloney :)15:10
ogasawaralag: ah, that might be a different story in you need LP permissions somewhere.15:10
ogasawaralag: what do you need changed in LP?15:10
lagbug 2207015:11
ubot2Launchpad bug 22070 in linux-source-2.6.20 (Ubuntu) (and 2 other projects) "Logitech Quickcam Messenger is not usable (affects: 15) (dups: 3) (heat: 144)" [Undecided,Won't fix] https://launchpad.net/bugs/2207015:11
ogasawarawhoa, 2.6.2015:11
ogasawaralag: and what do you need changes with that bug?15:12
ogasawaras/changes/changed/15:12
lagJust looking15:12
vanhoofogasawara: well he's here with a machine in hand :)15:12
lagI need the Nominated for Mavericks removing 15:13
lagAnd the Lucids removing from linux-source-2.6.15 (Ubuntu) and linux-source-2.6.20 (Ubuntu)15:13
ogasawaralag: ah, so that's a quirk in launchpad that you can't remove the nomination once it's been added.  You can only decline it now.15:14
lagThat seems silly15:14
ogasawaralag: it is, but I can decline them for you.15:14
lagI would have only like to add the nominated for on on line too, but it added them all :(15:15
lagThat's fine :)15:15
ogasawaralag: yah, also another quirk that's annoying15:15
lagJust leave The top Lucid one15:15
ogasawaralag: ack15:15
lagThanks Leann15:15
* ogasawara wonders if I decline a nomination for one, if it'll decline it for all15:15
lagAnother stupid quirk?15:16
ogasawarawell that's stupid, it does decline it for all15:17
lagGrrrr15:18
lagWhat about if you allow the top Lucid one?15:18
ogasawaralag: ok, so I approved the Lucid nomination which of course approved it for all the packages15:18
lagThen deline the others?15:18
lagDoh!15:18
lagWhaaaa, now what?15:19
ogasawaralag: so then I just marked the Lucid nominations for linux-source-2.6.15 and linux-source-2.6.20 as Inavlid15:19
lagAh, okay15:19
ogasawarawhat a painful process15:19
lagNo kidding15:19
lagOnce you've declined it, can you change it again?15:20
lagI remember why I nominated it for Maverick now15:21
ogasawaralag: yep, I can approve it15:21
lagNo I can split them and mark Maverick as Fix Released 15:21
=== diwic_afk is now known as diwic
lagCan you approve Maverick please15:21
lagI think I can then change then to invalid etc after15:21
lags/then/them15:21
ogasawaralag: sure, although it's sorta implied it's already Fix Released for Maverick15:22
ogasawaralag: as the linux task itself always reflects the actively developed release, ie Maverick at the moment15:22
ogasawarawow, so I approved it and it propagates the status automatically.15:24
* lag tuts15:25
lagOh, you mean Fix Released15:25
ogasawaralag: right15:26
lagWeird 15:26
ogasawaralag: so for example the linux-source-2.6.15 package being Fix Released for Maverick makes no sense15:26
lagYep, just seen/changed that15:27
lagNeither does the "Status tracked in Maverick"15:28
ogasawarayah, that's confusing too15:28
lagOkay, I think that's correct now15:28
lagWhy does the "linux-source-2.6.15 (Ubuntu)" have to be different?15:29
lagAnd has no dropdown?15:29
ogasawaralag: I think that's due to the nominations now, eg each nomination has a dropdown15:30
lagNone of them are the same15:31
lagTop one is "Status tracked...", but is white15:31
lagThe second one is the same, but yellow15:32
lagAnd the last one is while with a dropdown15:32
lagAnd is Won't fix15:32
ogasawaralag: the yellow highlight is due to the way the url is crafted15:32
ogasawaralag: eg https://bugs.edge.launchpad.net/ubuntu/+source/linux-source-2.6.15/+bug/22070 points to the linux-source-2.6.15 package and thus highlights it15:32
ubot2Launchpad bug 22070 in linux-source-2.6.20 (Ubuntu Maverick) (and 8 other projects) "Logitech Quickcam Messenger is not usable (affects: 15) (dups: 3) (heat: 142)" [Undecided,Invalid]15:32
lagAh, okay15:37
lagbug 2207015:38
ubot2Launchpad bug 22070 in linux-source-2.6.20 (Ubuntu Maverick) (and 8 other projects) "Logitech Quickcam Messenger is not usable (affects: 15) (dups: 3) (heat: 140)" [Undecided,Invalid] https://launchpad.net/bugs/2207015:38
lagBut the bug relates to linux-source-2.6.2015:39
lagOdd15:39
ogasawaralag: I'm guessing the bot just picks up the last package in the set of bug tasks15:45
lagStupid bot :)15:45
=== bjf[afk] is now known as bjf
=== ivoks-afk is now known as ivoks
=== amitk is now known as amitk-afk
=== diwic is now known as diwic_afk
tgardnerapw, bug #64224117:20
ubot2Launchpad bug 642241 in initramfs-tools (Ubuntu) "package initramfs-tools 0.92bubuntu78 [modified: usr/sbin/update-initramfs] failed to install/upgrade: subprocess installed post-installation script returned error exit status 1 (affects: 1) (heat: 269)" [Undecided,New] https://launchpad.net/bugs/64224117:20
tgardnerbug #64242117:24
ubot2Launchpad bug 642421 in linux (Ubuntu) "Maverick could not load /lib/modules/2.6.35-22-generic/modules.dep (affects: 41) (dups: 2) (heat: 206)" [Medium,Confirmed] https://launchpad.net/bugs/64242117:24
=== diwic_afk is now known as diwic
=== ivoks is now known as ivoks-afk
=== diwic is now known as diwic_afk
=== ppetraki is now known as ppetraki-lunch
mpoirierfriends, how do I generate the [PATCH 0/0] such as this one http://paste.ubuntu.com/508973/ when submitting multipile patches ?19:22
bjfmpoirier, have you tried "git format-patch --cover-leter" ?19:23
mpoirierbjf: will this command generate the information I have in the paste bin ?19:26
bjfmpoirier, try it19:26
bjfmpoirier, i am assuming that you use git format-patch to create your patches?19:27
mpoirieryes.19:27
mpoirierNow I need to generate the final 0/0.19:27
bjfmpoirier, just add --cover-letter to the options you are already use19:27
mpoirierok, I'll try.19:28
mpoirierbjf: git format-patch --cover-letter HEAD~6..HEAD will do the trick.19:47
mpoirierthanks for the tip.19:47
bjfmpoirier, git format-patch --cover-letter -619:48
bjfmpoirier, also: git format-patch --cover-letter HEAD~6...19:48
mpoirierbjf: indeed, they all seem to have the same effect.19:50
=== ppetraki-lunch is now known as ppetraki
ogasawarakees: hrm, for the day 0 Maverick kernel upload which is primarily CVE fixes, I was originally going to upload to -proposed and then get it immediately promoted to -updates.  But does it need to get to the -security pocket as well?20:01
=== ivoks-afk is now known as ivoks
keesogasawara: yup. but we can pocket-copy it there too.20:01
ogasawarakees: ok cool20:01
keesogasawara: this early in the cycle I'm not worried about build dep problems due to building in the regular queue vs security queue20:02
keesfor anything other than a 0day SRU, I'd be nervous :)20:02
* ogasawara nods20:02
=== ivoks is now known as ivoks-afk
=== yofel_ is now known as yofel

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!