All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mark Hounschell <markh@compro.net>
To: linux-kernel@vger.kernel.org
Cc: Mark Hounschell <dmarkh@cfl.rr.com>
Subject: Wireless network problem with Access Point selection using ifup
Date: Tue, 27 Nov 2012 10:04:12 -0500	[thread overview]
Message-ID: <50B4D66C.8070807@compro.net> (raw)

I have a wireless-N Linksys router centrally located in my home. I also 
have 2 each Linksys E1000 range extenders located at each end of the 
house. I have Desktop computers in fixed locations around the house. 
Using Network Manager all is well with all of them. I can see all the 
Access Points and select the one with the best signal. However network 
manager doesn't seem to be a valid option for all of them and I'm using 
traditional IFUP with static configurations.

When using IFUP, wrong Access Points are being chosen by whatever the 
underlying userland/kernel process is that is supposed to select it. It 
seems it wants to select the one with the weakest signal instead of the 
strongest.  I'm being told that when using ifup and WPA, as opposed to 
WEP, that userland has no choice in the selection of an Access Point by 
MAC address, only by SSID. All my APs have the same SSID.

I am using SuSE 12.2 Linux and 3.6.6 kernels on all these machines and 
the wireless hardware I am using on all these is are ASUS USB-300N adapters.

lsusb output: Bus 001 Device 004: ID 0b05:17ab ASUSTek Computer, Inc.

The kernel driver being used is rtl8192cu.


I don't understand why the developers of whatever the underlying 
userland/kernel process is that selects the AP, did not think that a 
user selection option by MAC address might be a good idea, as I'm told 
it is possible using WEP.  In any case, I am unable to get all this 
stuff working properly. Most of the time, the AP furthest away is 
selected. Often the router in the middle of the house is selected. Very 
very rarely is the closest AP chosen. When the furthest is chosen, I am 
unable to use the network. When the router in the middle of the home is 
selected, the connection comes to a crawl and often disassociates in the 
middle of use. When it does choose the correct AP, all is well until I 
reboot the machine.

The same thing happens with the desktop machine at the other end of the 
house but most of the time it selects the router in the middle of the 
house. It has never selected the closest AP.

Some sample dmesg and iwconfig output when the AP at the other end of 
the house was selected:


[   19.876977] wlan0: authenticate with c8:d7:19:7a:ab:1e
[   19.901616] wlan0: send auth to c8:d7:19:7a:ab:1e (try 1/3)
[   19.904099] wlan0: authenticated
[   19.916679] wlan0: associate with c8:d7:19:7a:ab:1e (try 1/3)
[   19.925839] wlan0: RX AssocResp from c8:d7:19:7a:ab:1e (capab=0x411 
status=0 aid=2)
[   19.925975] wlan0: associated
[   19.925988] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready



wlan0     IEEE 802.11bgn  ESSID:"hounschell-1"
               Mode:Managed  Frequency:2.437 GHz  Access Point: 
C8:D7:19:7A:AB:1E
               Bit Rate=150 Mb/s   Tx-Power=20 dBm
               Retry  long limit:7   RTS thr=2347 B   Fragment thr:off
               Encryption key:off
               Power Management:off
               Link Quality=66/70  Signal level=-44 dBm
               Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
               Tx excessive retries:0  Invalid misc:9   Missed beacon:0

This AP is at the other end of the house. The network was unusable after 
this boot.

Is this likely a kernel issue? If so, I would be willing to help with it 
if I can with my configuration.

Regards
Mark






             reply	other threads:[~2012-11-27 15:18 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-27 15:04 Mark Hounschell [this message]
  -- strict thread matches above, loose matches on Subject: below --
2012-12-06 14:10 Wireless network problem with Access Point selection using ifup Mark Hounschell
2012-12-06 15:03 ` Johannes Berg
2012-12-07 10:12   ` Mark Hounschell
2012-12-07 20:50     ` Mark Hounschell
2012-12-07 20:54       ` Johannes Berg
2012-12-07 21:06         ` Mark Hounschell
     [not found]           ` <50C5B65F.4040104@cfl.rr.com>
2012-12-10 10:26             ` Johannes Berg
2012-12-10 11:10               ` Mark Hounschell
2012-12-10 11:29                 ` Johannes Berg
2012-12-10 12:00                   ` Mark Hounschell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=50B4D66C.8070807@compro.net \
    --to=markh@compro.net \
    --cc=dmarkh@cfl.rr.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.