Date: Mon, 28 Nov 2011 17:20:33 +0100 From: =?ISO-8859-1?Q?Gustau_P=E9rez?= <gperez@entel.upc.edu> To: Adrian Chadd <adrian.chadd@gmail.com> Cc: freebsd-wireless@freebsd.org Subject: Re: Big delays between an AR5416 AP and its clients Message-ID: <4ED3B4D1.3020701@entel.upc.edu> In-Reply-To: <CAJ-Vmo=P%2BWAArZ1Kidh947yHuftZTBwF=Vx5Kn36-LKtD4grPg@mail.gmail.com> References: <3d4d2d249836fafc6acd885693c02198@webmail.entel.upc.edu> <CAJ-Vmo=1k=oKP-vDUH=BLdyeVzGnQvNmJ6TwGcdQjwpgJWt52w@mail.gmail.com> <4ECFCDAB.1080407@entel.upc.edu> <CAJ-VmokQTpgxMo=gmj0LsYhD6t9mmoHXG_sbhZdFNW5_OSGbmg@mail.gmail.com> <a9d1d8f91deb80a89bed98c1ffdb23fd@webmail.entel.upc.edu> <CAJ-Vmo=P%2BWAArZ1Kidh947yHuftZTBwF=Vx5Kn36-LKtD4grPg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 28/11/2011 11:08, Adrian Chadd wrote:
> Just run athstats a few times whilst you're pinging in the background
> and you see spikes of latency.
>
> In fact, try doing something like "ping -i 0.1<ip>", that way it
> sends out 10 pings a second. That'll hopefully bump up the counters
> even further.
>
> Thanks,
>
>
> Adrian
I see little increase in the counters. I hope you can see what
happens from them, because I have no idea.
I did 5 runs of 300 pings with 0,1 seconds before each one. The
client and the iphone are half a meter far from the AR5416 box.
Before each run I collected the counters with athstats, I labeled
each file as:
athstats_1client_pinging_iphone_assoc_before_runX.txt
where X is the run. Then I collected the state of the card while
pinging the AP with the AR5416, I labeled eash file as:
athstats_1client_pinging_iphone_assoc_runX.txt
I collected the info while pinging with:
athstats -i 1 | tee athstats_1client_pinging_iphone_assoc_runX.txt
I'm also adding the output of the ping command. The file is:
ping_0.1_delays.txt
As you may see, the problem has a frequency of about 8,5 seconds.
That is to say that after 8,5 seconds the delays fall one order of
magnitude. However the minimum value of the delays is about 10ms which
is 3 orders of magnitude bigger than the normal values (which are of
0,1ms).
You can fetch the results from:
https://www-entel.upc.edu/gus/ath_delay_values.zip
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4ED3B4D1.3020701>