Intel 3945ABG wifi can't connect at all if 802.11a and 802.11n are both available
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Expired
|
Undecided
|
Unassigned |
Bug Description
Binary package hint: network-manager
I have a Dell Latitude D620, with a builtin Intel 3945ABG wifi adapter. I am running Maverick, and the wifi adapter is being controlled by the iwl3945 driver.
I can connect fine to 802.11g at work, and I can also connect fine to 802.11a at home (via a Netgear WNDR3300 access point using DD-WRT firmware) — *IF* the WNDR3300's 5-GHz radio is configured to run in "A-only" mode.
If I configure my home access point's 5-GHz radio to run in "AN-only" mode (i.e., connect either in 802.11a or 802.11n mode), I can no longer connect at all. The logs show an endless repetition of messages like the following:
Apr 9 00:20:44 rde-richw-2 NetworkManager[
Apr 9 00:20:44 rde-richw-2 kernel: [168861.354918] wlan0: authenticate with 00:1f:33:b6:cd:82 (try 1)
Apr 9 00:20:44 rde-richw-2 kernel: [168861.355625] wlan0: authenticated
Apr 9 00:20:44 rde-richw-2 kernel: [168861.355669] wlan0: associate with 00:1f:33:b6:cd:82 (try 1)
Apr 9 00:20:44 rde-richw-2 kernel: [168861.357320] wlan0: RX AssocResp from 00:1f:33:b6:cd:82 (capab=0x11 status=18 aid=0)
Apr 9 00:20:44 rde-richw-2 kernel: [168861.357328] wlan0: 00:1f:33:b6:cd:82 denied association (code=18)
Apr 9 00:20:44 rde-richw-2 kernel: [168861.357359] wlan0: deauthenticating from 00:1f:33:b6:cd:82 by local choice (reason=3)
Apr 9 00:20:54 rde-richw-2 NetworkManager[
Apr 9 00:20:54 rde-richw-2 NetworkManager[
My wife's Windows 7 machine, with a Linksys AE1000 dual-band wireless-N adapter, can connect to the Netgear AP without difficulty while the AP is in either "A-only" or "AN-only" mode. And when I reconfigured the AP back to "A-only" mode, my laptop was once again able to connect in 802.11a mode. So I'm assuming the problem is most likely in the iwl3945 wifi driver — or possibly a hardware or firmware flaw in the Intel 3945ABG wifi adapter — rather than a problem with my access point.
This isn't a showstopper for me right now (802.11a works just fine, and in contrast to some other reports, I'm not seeing any throughput problems on this system) — but the issue *WILL* eventually come to a head at some point when I really, really want to configure 802.11n at home (in addition to 802.11a), because doing that right now would force me to degrade my laptop to 802.11g in a townhouse complex with lots and lots of congestion in the 2.4-GHz band.
As I said, I'm currently running Maverick, and since this laptop is a "production" machine from work, I am not at liberty to install experimental development builds at leisure in order to try to track down the problem (hopefully someone else with similar hardware can do that).
ProblemType: Bug
DistroRelease: Ubuntu 10.10
Package: network-manager 0.8.1+git.
ProcVersionSign
Uname: Linux 2.6.35-28-generic i686
NonfreeKernelMo
Architecture: i386
CRDA: Error: [Errno 2] No such file or directory
Date: Tue Apr 12 11:12:21 2011
EcryptfsInUse: Yes
InstallationMedia: Ubuntu 9.10 "Karmic Koala" - Release i386 (20091028.5)
IpRoute:
171.66.136.0/22 dev eth0 proto kernel scope link src 171.66.139.93 metric 1
10.32.208.0/21 dev wlan0 proto kernel scope link src 10.32.208.162 metric 2
169.254.0.0/16 dev eth0 scope link metric 1000
default via 171.66.136.1 dev eth0 proto static
Keyfiles: Error: [Errno 2] No such file or directory
ProcEnviron:
PATH=(custom, user)
LANG=C
SHELL=/bin/csh
SourcePackage: network-manager
WifiSyslog:
Apr 12 11:09:50 rde-richw-2 dhclient: DHCPREQUEST of 10.32.208.162 on wlan0 to 171.64.7.111 port 67
Apr 12 11:09:50 rde-richw-2 dhclient: DHCPACK of 10.32.208.162 from 171.64.7.111
Apr 12 11:09:50 rde-richw-2 dhclient: bound to 10.32.208.162 -- renewal in 1260 seconds.
Changed in linux (Ubuntu): | |
status: | Incomplete → Confirmed |
Thank you for taking the time to report this bug and helping to make Ubuntu better. The issue that you reported is one that should be reproducible with the live environment of the Desktop CD of the development release - Natty Narwhal. It would help us greatly if you could test with it so we can work on getting it fixed in the next release of Ubuntu. You can find out more about the development release at http:// www.ubuntu. com/testing/. Thanks again and we appreciate your help.