=== bigbash is now known as zz_bigbash | ||
=== bjf is now known as bjf[afk] | ||
=== bladernr_ is now known as bladernr_afk | ||
=== smb` is now known as smb | ||
apw | sconklin, i have just fiddled with your jenkins XML doc adding some information on using python unittest | 12:01 |
---|---|---|
apw | sconklin, which i believe is commonly used in this context from the results produced by other tests | 12:02 |
tgardner | jsalisbury, are you gonna start a bisect on bug #917962 ? | 13:15 |
ubot2 | Launchpad bug 917962 in linux "BUG: scheduling while atomic: swapper/3/0/0x10000100" [High,Confirmed] https://launchpad.net/bugs/917962 | 13:15 |
=== tgardner is now known as tgardner-afk | ||
=== bladernr_afk is now known as bladernr_ | ||
* ogasawara changes locations. back in 20. | 14:20 | |
=== tgardner-afk is now known as tgardner | ||
* tgardner is rebooting tangerine at the top of the hour for a kernel update | 14:52 | |
jsalisbury | tgardner, yes, I will bisect bug 917962 | 14:55 |
ubot2 | Launchpad bug 917962 in linux "BUG: scheduling while atomic: swapper/3/0/0x10000100" [High,Confirmed] https://launchpad.net/bugs/917962 | 14:55 |
ogasawara | tgardner: I'm gonna prep an upload, anything else you want shoved in? | 15:10 |
tgardner | ogasawara, nothing leaps to mind | 15:13 |
tgardner | oh, forgot to reboot tangwerine | 15:13 |
* ogasawara makes sure to use gomeisa | 15:14 | |
ogasawara | tgardner: I might be off the grid by the time the upload finishes. It's an ABI bumper, could you or apw do the linux-meta upload tomorrow for me? | 15:15 |
tgardner | ogasawara, no problem. drop Andy a note since he'll be on earlier then me. | 15:15 |
ogasawara | tgardner: ack | 15:16 |
tgardner | ogasawara, I'm updating your chroots on gomeisa. they should be done in a couple minutes | 15:17 |
ogasawara | tgardner: ok thanks | 15:17 |
jsalisbury | ogasawara, tgardner, is it OK to build test kernels on tyler? | 15:24 |
tgardner | jsalisbury, sure | 15:24 |
jsalisbury | tgardner, cool, just wanted to check. | 15:24 |
tgardner | ogasawara, gomeisa updated | 15:26 |
ogasawara | tgardner: cool, thanks | 15:26 |
* ogasawara kicks off test builds | 15:26 | |
pgraner | apw, ping | 15:37 |
tgardner | pgraner, apw is out until tomorrow | 15:51 |
tgardner | sforshee, did you get the DSDT table? | 15:52 |
sforshee | tgardner, yes, thanks. Haven't gotten around to looking at it yet. | 15:52 |
tgardner | sforshee, np | 15:52 |
tgardner | smb, nice write up: https://wiki.ubuntu.com/Kernel/Reference/Orchestra | 15:56 |
smb | tgardner, Thanks, hoped it could be helpful and actually remind me as I tend to forget. :) | 15:56 |
bjf | smb, i'm working on orchestra+koan+jenkins and will add to your doc. very nice start | 15:57 |
tgardner | smb, bjf: nagios is installed with the orchestra server. I wonder what stats it keeps ? | 15:59 |
smb | bjf, cool. The other idea was to have the tools (in that case hopefully helpful) in seperate docs and then put them together under a starting point | 15:59 |
bjf | smb, yes, i wasn't going to polute your doc so much as maybe add pointers from it | 16:00 |
bjf | tgardner: don't know, that package install a *lot* of other packages | 16:00 |
smb | tgardner, I seem to see some nagios messages on those machines I had install via orchestra, but have not looked closer | 16:00 |
tgardner | smb, bjf: may SpamapS knwos | 16:01 |
tgardner | maybe* | 16:01 |
smb | bjf, Oh I wasn't so much afraid of pollution. Just liked the idea of separation because that way my brain tends to be more single tasked :) | 16:02 |
smb | errr | 16:02 |
smb | that does not make sense | 16:02 |
kirkland | tgardner: smb: damn, nice write up :-) | 16:02 |
=== JanC_ is now known as JanC | ||
bjf | smb: understand. i don't want it to turn into a doc like some of our "build" docs which are so hard to understand | 16:02 |
* smb nods | 16:03 | |
smb | kirkland, thanks | 16:03 |
kirkland | smb: fyi, there's also https://help.ubuntu.com/community/Orchestra | 16:04 |
smb | kirkland, Ah cool, I will go trhough that | 16:05 |
* smb bookmarks | 16:05 | |
tgardner | smb, whenever you update one of the cobbler config files, don't you need a 'sudo cobbler sync' ? | 16:05 |
smb | tgardner, Good point. I not sure when and when not. | 16:06 |
tgardner | smb, I think anytime you want your changes deployed | 16:07 |
pgraner | tgardner, ok thx | 16:07 |
tgardner | pgraner, he sent email to c-k-t that he was out early to do to the opera (or some such shit) | 16:08 |
tgardner | pgraner, maybe he was jealous of your experience :) | 16:08 |
smb | tgardner, Right, on the other hand a lot of those are local text files and I don't really understand toward where that sync works to understand why I need it | 16:09 |
smb | But probably it restart various daemons or builds other files | 16:09 |
pgraner | tgardner, don't rub it in | 16:14 |
tgardner | :) | 16:14 |
* smb leaves for appointment | 16:22 | |
BetaBrain | hi all | 17:28 |
BetaBrain | http://en.wikipedia.org/wiki/Comparison_of_open-source_wireless_drivers are confused with this kind of damn realtek driver (rtl8187 - r8187) from when we have a quarrel one day and I absolutely do not find the solution. this chipset (RTL8187) is still poorly supported by linux. but in practice it works very well because realtek linux drivers not provide decent features it provides to be able to write from scratch. What can I do? I have | 17:28 |
BetaBrain | looked everywhere with open source drivers I have tried with this result http://rtl8180-sa2400.sourceforge.net https://rtl-wifi.svn.sourceforge.net/svnroot/rtl-wifi/ but compiles it and then reading http://forums.fedoraforum.org/archive/index.php/t-151020.html here and that some expert tell me how thanks: Delta | 17:28 |
BetaBrain | Sorry for my bad English | 17:28 |
=== yofel_ is now known as yofel | ||
Gerald | ohsix, hi...what is causing this problem? http://openpaste.org/f67E0f34 | 20:30 |
Gerald | i refering to the build | 20:34 |
geri | can someone help out with my problem refering to dynamic kernel module support...? | 21:32 |
geri | http://openpaste.org/cB0Ae2B9 | 21:33 |
niksoft | Hi | 21:43 |
elfurbe | Anyone know the status of TRIM support for md devices offhand? | 21:43 |
elfurbe | It looks like it was added to dm a while back | 21:43 |
geri | hi elfurbe can u help out with my problem refering to dynamic kernel module support...? http://openpaste.org/cB0Ae2B9 | 21:46 |
=== bladernr_ is now known as bladernr_afk | ||
elfurbe | I definitely cannot, geri | 21:46 |
elfurbe | I'm here looking for help myself | 21:47 |
niksoft | elf, have you looked at: http://www.ocztechnologyforum.com/forum/showthread.php?82648-software-RAID-LVM-TRIM-support-on-Linux | 21:48 |
elfurbe | I did, but he's not doing mirrored lvs, just striped | 21:49 |
niksoft | no raid 1 is mirror... | 21:50 |
elfurbe | He was doing raid0 | 21:50 |
elfurbe | I mean | 21:50 |
niksoft | i see, it is deceiving, he does mirror for boot | 21:50 |
elfurbe | raid1 for boot | 21:50 |
elfurbe | which is fine, I don't think you can boot from lvm, I'm not entirely sure | 21:51 |
elfurbe | Yeah :D | 21:51 |
geri | hi niksoft ... can u help on this? | 21:54 |
niksoft | hold on for a sec geri, i'll look at it (though i am not a kernel dev, i still should be able to figure this out, i think | 22:03 |
sforshee | geri, is hello.ko file present in the build directory after your build fails? | 22:04 |
geri | no | 22:04 |
geri | im not sure what im mission from the description... ;( | 22:07 |
sforshee | geri, you're looking in /var/lib/dkms/hello/0.1/build ? | 22:09 |
geri | i pasted it | 22:09 |
sforshee | i mean for the .ko file | 22:10 |
geri | there is not .ko created | 22:10 |
geri | whats this path here? /lib/modules/$(KVERSION)/build | 22:13 |
geri | thats set in the makefile | 22:13 |
sforshee | geri, can you paste your dkms.conf, and also check that you have the linux-headers package installed for your kernel | 22:13 |
geri | ok mom | 22:14 |
geri | http://openpaste.org/28463Dd1 | 22:14 |
geri | here | 22:14 |
geri | i manually build the linux kernel | 22:15 |
sforshee | okay, but you have to have the necessary files in /lib/modules/$(KVERSION)/build to build modules against your kernel | 22:15 |
sforshee | the headers package normally provides this | 22:15 |
geri | i have a folder /lib/modules/$(KVERSION)/build | 22:16 |
geri | there is a build kernel and source directory | 22:16 |
niksoft | Would anyone be able (or willing) to help me with a bonding issue? More specifically 802.1ad link aggregation that is not balancing tx (at all) | 22:18 |
geri | any idea sforshee? | 22:22 |
sforshee | I think you should try a stock kernel. If it works there then it's something about your custom build. Maybe some missing infrastructure under that /lib/.../build directory, it needs to have the makefile and headers in place for what's in your makefile to work | 22:24 |
geri | wait i have a problem in the makefile | 22:24 |
geri | i forgot to add the tab | 22:24 |
geri | now it says: http://openpaste.org/43A44CFF | 22:25 |
geri | :) | 22:25 |
sforshee | and make.log says what? | 22:26 |
geri | http://openpaste.org/6AFB5Eac | 22:26 |
geri | oh | 22:27 |
niksoft | actually hold on my issue may be a tool issue | 22:27 |
niksoft | i swear i will kick some spirent devs if it is | 22:27 |
sforshee | geri, I think your /lib/module/.../build directory doesn't have all the required stuff | 22:29 |
geri | ok it builds now | 22:29 |
geri | it compiles against the linux header files? | 22:29 |
sforshee | yes, the #includes have to be satisfied from somewhere | 22:30 |
geri | ok it works fine now | 22:33 |
geri | i patched a linux kernel file in drivers/usb/class/cdc-acm.c...could i apply this method to avoid recompiling of the complete kernel? | 22:34 |
geri | i only compile this single file or how does that work? | 22:34 |
sforshee | yes, so long as you set up dkms.conf to install to /updates it will override the modules shipped with the kernel | 22:35 |
sforshee | you have to compile whatever files are needed for the module ... | 22:35 |
sforshee | in this case it looks like only that file would be needed | 22:35 |
=== chrisccoulson_ is now known as chrisccoulson | ||
niksoft | anyone know what the linux implementation of 802.3ad uses for load balancing? | 22:50 |
niksoft | i am guessing not round robin, but what is the hash based on? | 22:50 |
geri | sforshee, what should i set the package name and package version to? | 22:54 |
geri | shoudl i modify the source direct in the kernel path? | 22:56 |
sforshee | really they can be whatever you want, I'd probably use cdc-acm and 0.1, incrementing the version each time you update the package | 22:56 |
sforshee | I don't know what you mean | 22:57 |
niksoft | nm, so digging in the documentation, by default 802.3ad uses an xor balancing method based on the 2 macs | 22:57 |
geri | can i modify the cdc-acm.c direct in the kernel source path? | 22:57 |
sforshee | you can modify it in the kernel source tree, but it's what gets into your dkms source package that matters | 22:57 |
zpmorgan | Is the kernel's ASPM overheating regression supposed to be fixed by now in Oneiric? | 22:58 |
geri | sforshee, any idea what i did wrong? http://openpaste.org/DE031c9e | 23:06 |
geri | ok try to fix it | 23:10 |
geri | so it build now | 23:11 |
geri | sforshee, modprobe will automatically overwrite the kernel module? | 23:18 |
geri | how can i reset modprobe? | 23:18 |
geri | sforshee, one last question will apt-get source linux-2.6.35-generic include the linux header files? | 23:56 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!