=== syn-ack is now known as Guest79638 | ||
=== Guest79638 is now known as syn-ack | ||
=== KB1JWQ is now known as TheOracle | ||
=== TheOracle is now known as KB1JWQ | ||
=== abhijitb is now known as bain | ||
zbenjamin | hi all | 15:40 |
---|---|---|
zbenjamin | i have a problem with the iwlagn driver + 03:00.0 Network controller: Intel Corporation PRO/Wireless 5300 AGN [Shiloh] Network Connection | 15:41 |
zbenjamin | i cannot connect to the AP | 15:41 |
zbenjamin | with WPA2 | 15:41 |
zbenjamin | dmesg output http://pastebin.com/XqHTQArQ | 15:42 |
zbenjamin | anyone can help with this? | 15:43 |
manjo | zbenjamin, why do you think it is a kernel problem ? | 15:43 |
zbenjamin | i think its related to the driver | 15:44 |
zbenjamin | also i got redirected here from #kubuntu | 15:44 |
zbenjamin | i know the password is ok, windows can connect fine | 15:45 |
manjo | zbenjamin, it could be a network manager issue | 15:47 |
zbenjamin | the same problem when i use wpa_supplicant without network-manager same problem with wcid | 15:48 |
zbenjamin | so i doubt its related to network manager | 15:48 |
manjo | zbenjamin, what version of Ubuntu are you using? | 15:49 |
manjo | karmic ? | 15:49 |
zbenjamin | karmic , kernel Linux version 2.6.31-20-generic (buildd@crested) (gcc version 4.4.1 (Ubuntu 4.4.1-4ubuntu8) ) | 15:49 |
zbenjamin | 64bit | 15:49 |
manjo | zbenjamin, kde ? | 15:50 |
zbenjamin | yes | 15:50 |
zbenjamin | but currently i try with wpa_supplicant directly | 15:50 |
zbenjamin | i could post wpa_supplicant output if that helps | 15:52 |
zbenjamin | oh can this be the problem? Could not configure driver to use managed mode | 15:52 |
zbenjamin | from wpa_supplicant | 15:52 |
manjo | zbenjamin, you should open a bug on this... but I know it is hard to do without working network :) | 15:54 |
cnd | zbenjamin: are you specifying the AP by mac address instead of SSID? | 15:55 |
cnd | if so, maybe try by ssid, since that's a more commonly used and tested scenario | 15:55 |
zbenjamin | no i use SSID | 15:55 |
zbenjamin | also the AP is not hidden | 15:56 |
cnd | zbenjamin: ok, I've just not seen "direct probe to AP" messages before | 15:56 |
cnd | zbenjamin: can you try another router to test? | 15:56 |
zbenjamin | cnd: yeah this is a new message after i installed the backports-modules package | 15:56 |
cnd | sometimes there are strange interactions between specific drivers and routers | 15:56 |
cnd | which backports-modules? | 15:57 |
zbenjamin | cnd: i tried with other routers and it works, i also have exactly the same hw router at home and it works without a problem | 15:57 |
zbenjamin | sec | 15:57 |
cnd | zbenjamin: same security type too? | 15:57 |
zbenjamin | linux-backports-modules-karmic | 15:57 |
zbenjamin | yes | 15:57 |
zbenjamin | but the router is not broken, i can connect to it from win without a problem | 15:58 |
cnd | zbenjamin: maybe this router has old firmware? | 15:58 |
zbenjamin | you think that could be a problem? | 15:58 |
cnd | zbenjamin: you may want to try the linux-backports-modules-wireless package | 15:58 |
zbenjamin | argh | 15:59 |
cnd | zbenjamin: fw could affect anything | 15:59 |
cnd | so yeah, I would ensure you are up to date on your router's fw | 15:59 |
zbenjamin | could not find linux-backports-modules-wireless | 15:59 |
zbenjamin | this one? linux-backports-modules-wireless-karmic-generic | 16:00 |
manjo | yes. | 16:00 |
zbenjamin | a reboot is enough? | 16:01 |
zbenjamin | or do i have to disable some drivers? | 16:01 |
manjo | reboot is fine | 16:01 |
zbenjamin | ok bib | 16:01 |
=== zbenjamin is now known as zbenjamin|away | ||
=== zbenjamin|away is now known as zbenjamin | ||
zbenjamin | back | 16:08 |
zbenjamin | i get this in messages: [ 230.810852] ADDRCONF(NETDEV_UP): wlan0: link is not ready | 16:08 |
zbenjamin | and this in dmesg [ 241.940662] wlan0: deauthenticating from 00:1a:2a:1a:fe:0c by local choice (reason=3) | 16:09 |
zbenjamin | that is totally weird | 16:09 |
manjo | so you are not able to connect still ? | 16:09 |
zbenjamin | no | 16:10 |
zbenjamin | never had so much problems before | 16:10 |
zbenjamin | whats that: | 16:12 |
zbenjamin | wpa_driver_nl80211_set_key: alg=0 addr=0x470a3c key_idx=3 set_tx=0 seq_len=0 key_len=0 | 16:12 |
zbenjamin | nl80211: set_key failed; err=-67 | 16:12 |
manjo | did this work previously ? | 16:15 |
zbenjamin | yes i was able to create connections to other APs | 16:15 |
manjo | zbenjamin, I have a feeling this is network manager related | 16:15 |
zbenjamin | it cannot, i start wpa_supplicant by hand | 16:15 |
zbenjamin | wpa_passphrase and stuff | 16:15 |
zbenjamin | that was the first thing i did, disable networkmanager | 16:16 |
zbenjamin | hm if i use -Dwext on wpa_supplicant it cannot find a AP at all | 16:16 |
zbenjamin | Try to find WPA-enabled AP | 16:17 |
zbenjamin | Try to find non-WPA AP | 16:17 |
zbenjamin | No suitable AP found. | 16:17 |
zbenjamin | sorry for the paste, i should use a pastebin | 16:17 |
zbenjamin | manjo: any other ideas? | 16:22 |
manjo | zbenjamin, no. you should open a bug with all this info in it. | 16:23 |
zbenjamin | :( | 16:24 |
salgado | cnd, load seems a bit high (load average: 2.04, 2.04, 1.92) but I can't find where that's coming from | 16:35 |
cnd | salgado: it's probabaly a process stuck in a loop | 16:36 |
cnd | bear with me a moment | 16:36 |
salgado | sure | 16:37 |
cnd | salgado: do the following: | 16:37 |
cnd | sudo sh -x "echo function > /sys/kernel/debug/tracing/current_tracer" | 16:38 |
cnd | sit for a few seconds | 16:38 |
cnd | then do: | 16:38 |
cnd | sudo sh -x "cat /sys/kernel/debug/tracing/trace > ~/trace" | 16:38 |
cnd | then "bzip2 ~/trace" | 16:39 |
salgado | sh: Can't open echo function > /sys/kernel/debug/tracing/current_tracer | 16:39 |
cnd | then upload ~/trace.bz2 to lp | 16:39 |
cnd | salgado: oops, I mean sh -c | 16:39 |
salgado | did that using sudo tee, though | 16:39 |
cnd | in both cases | 16:39 |
cnd | after you cat trace, you can turn off tracing by echoing "nop" to current_tracer | 16:40 |
salgado | ok, uploading | 16:41 |
salgado | cnd, http://launchpadlibrarian.net/43167185/trace.bz2 | 16:42 |
cnd | salgado: how many cpus are in the system? | 16:45 |
salgado | cnd, one cpu, 2 cores | 16:45 |
cnd | salgado: I'm not sure why your load avg is high, but it doesn't look related to this bug | 16:46 |
cnd | the trace doesn't show anything hung in a loop or anything related to graphics | 16:46 |
salgado | cnd, it seems to be doing tons of interrupts (700/s) and even more context switches (1800/s) | 16:52 |
cnd | salgado: I'm looking into mutex debugging right now, which will help us debug the core issue | 16:52 |
salgado | ok | 16:56 |
=== zbenjamin is now known as zbenjamin|away | ||
salgado | cnd, I need to step out for a bit, but I'll check back in 10 minutes or so, in case there's anything else I can do to help | 17:03 |
cnd | salgado: probably nothing for a bit | 17:04 |
cnd | I'm still researching | 17:04 |
salgado | ok, cool | 17:04 |
=== salgado is now known as salgado-brb | ||
=== zbenjamin|away is now known as zbenjamin | ||
zbenjamin | manjo: it works, now guess what i have done | 17:05 |
manjo | :) | 17:05 |
manjo | reinstalled network manager ? | 17:05 |
zbenjamin | NO removed the plug !!!! | 17:06 |
zbenjamin | using the normal switch did not work | 17:07 |
zbenjamin | ! | 17:07 |
zbenjamin | i had to remove the power plug, and now it magically works | 17:07 |
zbenjamin | took me only 8 hours to find :) | 17:07 |
zbenjamin | manjo: strange that it worked for windows | 17:08 |
=== yofel_ is now known as yofel | ||
manjo | you mean reboot the router ? | 17:09 |
manjo | or unplug the power cord from the laptop? | 17:09 |
zbenjamin | rebooting was not enough, i had to remove the power chord from the router | 17:09 |
zbenjamin | i used the power switch before | 17:10 |
zbenjamin | but that did not help | 17:10 |
manjo | that is a bit strange... | 17:10 |
zbenjamin | yes indeed | 17:10 |
zbenjamin | well maybe the router does only some suspend to ram thing who knows | 17:11 |
zbenjamin | ok i try to reconnect now lets see if it still works | 17:12 |
* zbenjamin hopes | 17:12 | |
=== zbenjamin is now known as zbenjamin|away | ||
=== zbenjamin|away is now known as zbenjamin | ||
zbenjamin | manjo: still works | 17:15 |
manjo | zbenjamin, cheers! what router is that ? | 17:16 |
zbenjamin | its a Speedport W700 | 17:16 |
zbenjamin | you get those from the t-online provider | 17:17 |
zbenjamin | http://www.findix.com/data/adpix/picture_L/router-t-com-speedport-w-700-v-194121.jpg | 17:18 |
zbenjamin | manjo: but it still has to be a problem with the driver, as i said it worked on windows | 17:20 |
zbenjamin | or the windows driver is aware of wrong behaviour from the router | 17:21 |
zbenjamin | or more error tolerant | 17:21 |
cnd | zbenjamin: had you tried windows after linux wasn't working? | 17:35 |
zbenjamin | cnd: yep | 17:46 |
zbenjamin | cnd: worked without a problem | 17:46 |
cnd | zbenjamin: I don't know then, but consumer routers are fairly flaky | 17:47 |
cnd | it's always good to try rebooting them | 17:47 |
zbenjamin | yeah well i tried to reboot it, using the power switch, but only removing the power chord fixed it | 17:48 |
zbenjamin | anyway thx for help guys | 17:51 |
cnd | glad to see your issue got resolved :) | 17:51 |
=== salgado-brb is now known as salgado | ||
salgado | cnd, is there any use in leaving the crashed laptop running or is it ok to restart it? | 17:59 |
cnd | salgado: you can restart it | 17:59 |
salgado | cool | 17:59 |
salgado | back in a bit | 17:59 |
cnd | hmmm... I'm getting errors on prepare-generic on the lucid sources, even after git clean -f -d, git reset --hard HEAD, and fdr clean | 18:05 |
cnd | apw, ogasawara any thoughts? | 18:05 |
ogasawara | cnd: hrm, what's the exact error? | 18:06 |
cnd | ogasawara: /home/cndougla/Canonical/ubuntu-lucid is not clean, please run 'make mrproper' | 18:07 |
jjohansen | cnd: run make mrproper and then do git checkout -f | 18:08 |
jjohansen | that happens with the kernel build system some times | 18:08 |
jjohansen | especially if you end up mixing fdr and regular kernel builds | 18:09 |
jjohansen | cnd: basically there are some files there that the build system is checking that git is ignoring so they don't show up in a git status | 18:10 |
cnd | jjohansen: ahhh, that makes sense | 18:10 |
jjohansen | the checkout -f will restore your debian directories etc that mrproper will destroy | 18:11 |
cnd | jjohansen: yep, it's working now, thanks! | 18:11 |
manjo | cnd it will be cool to doc that in the wiki | 18:24 |
manjo | kernel mainintanence starter or so | 18:25 |
cnd | manjo: yeah, I'll add it to my task list | 18:25 |
=== sconklin is now known as sconklin-lunch | ||
salgado | cnd, any news about that bug? or can you think of anything I could try as a workaround meanwhile? (it's happened two more times in the last 30 minutes...) | 19:32 |
cnd | salgado: I'm building a debug kernel right now | 19:32 |
cnd | I'll probably have it ready for you within the hour | 19:33 |
cnd | I'll post in the bug when it's ready | 19:33 |
salgado | cool. I just hope the bug doesn't go away when I install the debug kernel. :) | 19:33 |
cnd | heh, it shouldn't, but you never know :) | 19:34 |
cnd | it would be a really nasty bug if that were the case | 19:34 |
=== sconklin-lunch is now known as sconklin | ||
salgado | cnd, http://people.canonical.com/545039/ is a 404 | 20:55 |
cnd | salgado: oops: http://people.canonical.com/~cndougla/545039 | 20:56 |
salgado | that one works. :) | 20:57 |
salgado | cnd, I'm now on the karmic kernel, as the Lucid one wasn't allowing me to get any work done. luckily it will crash quickly once I switch to the new kernel you built | 20:58 |
cnd | ok | 20:59 |
* salgado restarts | 21:01 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!