AP performance (again): txpower regulation

Adrian Chadd adrian at freebsd.org
Fri Sep 9 12:08:43 UTC 2011


.. I forgot to cc- the list.

Also to clarify:

* short retries are "I sent an RTS but didn't hear a CTS."
* long retries are "I sent the frame but didn't hear an ACK."

Since the default athstats doesn't spit out the ofdm and cck errors, try this:

athstats -o input,output,altrate,short,long,xretry,crcerr,crypt,phyerr,rssi,ofdm,cck,rate
1



Adrian

On 9 September 2011 20:06, Adrian Chadd <adrian at freebsd.org> wrote:
> Right.
>
> xretry means "exceeded retries". Ie, all the chip retries failed.
>
> On the TX side of things - look at how high your long retry count is.
> That means the chip TXed the frame and it didn't hear an ACK back.
>
> Let's focus on RX for now, as TX may just be a little screwy.
>
> Try a 20mbit UDP test stream from the client -> hostap -> server:
>
> server:
>
> iperf -su -i 1
>
> iperf -c <server ip> -ub 20m -t 30
>
> Then see what the receive rate is like.
>
>
>
> adrian
>


More information about the freebsd-wireless mailing list