[Prism54-devel] [Bug 55] Spurious disassociations in prismGT STA <-> prism2.5 AP setup

[bugzilla-daemon@mcgrof.com]

http://prism54.org/cgi-bin/bugzilla/show_bug.cgi?id=55





------- Additional Comments From vda@port.imtp.ilyichevsk.odessa.ua  2004-03-03 21:13 -------
>I have the same problem (also caused by weak signal), under even moderate
>traffic it reassociates every few seconds. Increasing number of retries helps
>alot, 13 or 14 seem to work best for me, ie:
>
>iwconfig <iface> retry limit 14

Thanks! Seems like this helps with disassoc problem,
but performance is mediocre at best.

However, my signal is NOT weak. More precisely, other
STAs have comparable or _lower_ signal level, yet work
mostly at 11Mbits, while prism54 sends frames mostly
at 2Mbit, and with packet loss.

These are stats from AP:

00:04:e2:64:15:e5: last_rx: silence=156 signal=175 rate=10
tx[1M]=2        tx[2M]=22       tx[5.5M]=357    tx[11M]=5421
rx[1M]=576      rx[2M]=6173     rx[5.5M]=101    rx[11M]=0

Above is prism54. Mostly sends at 2Mbit (btw: packet loss ~10%)

00:05:5d:fa:1a:e3: last_rx: silence=156 signal=169 rate=110
tx[1M]=0        tx[2M]=0        tx[5.5M]=0      tx[11M]=7482
rx[1M]=0        rx[2M]=15       rx[5.5M]=0      rx[11M]=626

DLink DWL520. Signal level is lower, but it is solid at 11Mbit.
Below: more DWLs, all doing fine mostly at 11Mbits.

00:05:5d:fa:56:20: last_rx: silence=156 signal=171 rate=110
tx[1M]=0        tx[2M]=0        tx[5.5M]=0      tx[11M]=515974
rx[1M]=5393     rx[2M]=5735     rx[5.5M]=18725  rx[11M]=521155

00:05:5d:fa:56:36: last_rx: silence=156 signal=182 rate=110
tx[1M]=0        tx[2M]=2        tx[5.5M]=323    tx[11M]=933886
rx[1M]=149206   rx[2M]=31658    rx[5.5M]=41172  rx[11M]=667589

00:05:5d:fa:58:2b: last_rx: silence=156 signal=188 rate=110
tx[1M]=0        tx[2M]=0        tx[5.5M]=84     tx[11M]=1000026
rx[1M]=43868    rx[2M]=80278    rx[5.5M]=168466 rx[11M]=670682

In other words, I'm in deep, hm, interoperability problem ;)



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.