Since I had to fiddle with the settings at the router here is what finally worked for me:
RFC 2516 PPPoE
Multiplexing: LLC
QoS Type: UBR
PCR: default (4000)
SCR: default (4000)
VPI: 1
VCI: 32
DSL Modulation: ADSL2+
Username: 1und1/xxxx-yyy@online.de
Password: [your password]
Note: This works for ADSL based connections. VDSL is a different topic.
We have three Cisco 4410n (WAP4410N-E V02 to be precise) access points delivering wireless lan in our offices. All of them serve the same SSID on different channels. This way we don’t need a roaming controller and have a cheap, simple setup. This worked quite well in the past months. But now and then two of the three boxes stop working for no apparent reason. They can’t be reached over the wired interface and they don’t accept any clients (though connected clients still seem to work).
Using a syslog daemon I found many many of the following two lines:
User.Info host3 Apr 10 17:28:25 host3 syslog: Found beacon stuck, down and up the VAP interface.According to sources on the internet these messages are related to the problem.
Local0.Info host3 Apr 10 17:28:22 host3 kernel: ath_bstuck_tasklet: stuck beacon; resetting (bmiss count 36)
All access points are alreay up-to-date regarding the firmware release (2.0.6.1). Obviously Cisco doesn’t put any more efforts into a new firmware for this product.
I started experimenting with different settings. Some sources suggest increasing the beacon interval time. But this didn’t change anything. Then I compared the settings between the three boxes: all settings are the same besides the wifi channel. So I changed the channels and suddenly the error messages on the syslogd disappeared. As it turned out the errors start as soon as I configure an ap for wifi channel 9 or higher. I don’t understand how a different frequency makes a difference on things like dma and interrupts. So, lets see how things turn out in the next few days.