Another working USB WiFi adapter:

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Another working USB WiFi adapter:

Craig Skinner-3
For the archives,

A friend loaned me this while I gave an Unbound & NSD talk in a pub:

TP-Link TL-WN722N "High-Gain 150Mbps"

athn1 at uhub0 port 2 "ATHEROS USB2.0 WLAN" rev 2.00/1.08 addr 2
athn1: AR9271 rev 1 (1T1R), ROM rev 13, address c0:4a:00:1f:f6:3e

$ uname -mrsv
OpenBSD 5.6 GENERIC.MP#299 i386

Cheers!
--
Great Glasgow oak cask aged beer: Innis & Gunn
The most popular British bottled beer sold in Canada,
and second most popular in sunny Sweden. (Superb country.)
http://www.InnisAndGunn.Co.UK

Reply | Threaded
Open this post in threaded view
|

Re: Another working USB WiFi adapter:

Stefan Sperling-5
On Thu, Sep 03, 2015 at 08:53:11AM +0100, Craig Skinner wrote:

> For the archives,
>
> A friend loaned me this while I gave an Unbound & NSD talk in a pub:
>
> TP-Link TL-WN722N "High-Gain 150Mbps"
>
> athn1 at uhub0 port 2 "ATHEROS USB2.0 WLAN" rev 2.00/1.08 addr 2
> athn1: AR9271 rev 1 (1T1R), ROM rev 13, address c0:4a:00:1f:f6:3e
>
> $ uname -mrsv
> OpenBSD 5.6 GENERIC.MP#299 i386
>
> Cheers!

I'm surprised it works for you at all on 5.6. It might actually crash
your machine quite easily because of bugs in the driver. And scanning
is rather broken with this device in 5.6 (won't see more than 8 APs).

Support for this device was improved a lot for 5.7. See the changelog at
http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys/dev/usb/if_athn_usb.c

There are still some issues, though. Some people are seeing frequent
device timeouts which I can't reproduce. And hostap mode seems to be broken.

If you have time to do some tests (with -current), I'd be interested to
know if hostap works. To test, set it up for hostap mode, try to assoiate
to it, assign IPs, and ping across.
Does the device send beacons in hostap mode? Beacons should be visible
from another computer with a wireless interface in monitor mode on the
same channel, and tcpdump -y IEEE802_11_RADIO -v

Do you see frequent or occasional device timeouts when using the device
normally? If so, can you figure out which conditions trigger them?

Reply | Threaded
Open this post in threaded view
|

Re: Another working USB WiFi adapter:

Craig Skinner-3
Hi Stefan,

On 2015-09-03 Thu 10:50 AM |, Stefan Sperling wrote:
>
> I'm surprised it works for you at all on 5.6. It might actually crash
> your machine quite easily because of bugs in the driver. And scanning
> is rather broken with this device in 5.6 (won't see more than 8 APs).
>

A friend brought this adapter to the pub with him, so we tried it.

It seemed to work without problems for 1.5 hours of several ssh sessions
to DNS master & slave hosts & web surfing.

Yes, for a variety of reasons, this netbook needs to be on -current,
the todo list guestimate is October....

I'll try to report back later.
--
My University of Edinburgh dance partners perform 'The Braes of Busby':
http://www.scottish-country-dancing-dictionary.com/video/braes-of-busby.html
The driving Strathspey rhythm & step is unique to Scottish folk music/dancing.

Reply | Threaded
Open this post in threaded view
|

Re: Another working USB WiFi adapter:

Henrique N. Lengler
In reply to this post by Stefan Sperling-5
On Thu, Sep 03, 2015 at 10:50:02AM +0200, Stefan Sperling wrote:

> On Thu, Sep 03, 2015 at 08:53:11AM +0100, Craig Skinner wrote:
> > For the archives,
> >
> > A friend loaned me this while I gave an Unbound & NSD talk in a pub:
> >
> > TP-Link TL-WN722N "High-Gain 150Mbps"
> >
> > athn1 at uhub0 port 2 "ATHEROS USB2.0 WLAN" rev 2.00/1.08 addr 2
> > athn1: AR9271 rev 1 (1T1R), ROM rev 13, address c0:4a:00:1f:f6:3e
> >
> > $ uname -mrsv
> > OpenBSD 5.6 GENERIC.MP#299 i386
> >
> > Cheers!
>
> I'm surprised it works for you at all on 5.6. It might actually crash
> your machine quite easily because of bugs in the driver. And scanning
> is rather broken with this device in 5.6 (won't see more than 8 APs).
>
> Support for this device was improved a lot for 5.7. See the changelog at
> http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/sys/dev/usb/if_athn_usb.c
>
> There are still some issues, though. Some people are seeing frequent
> device timeouts which I can't reproduce. And hostap mode seems to be broken.
>
> If you have time to do some tests (with -current), I'd be interested to
> know if hostap works. To test, set it up for hostap mode, try to assoiate
> to it, assign IPs, and ping across.
> Does the device send beacons in hostap mode? Beacons should be visible
> from another computer with a wireless interface in monitor mode on the
> same channel, and tcpdump -y IEEE802_11_RADIO -v
>
> Do you see frequent or occasional device timeouts when using the device
> normally? If so, can you figure out which conditions trigger them?

I use this device daily (have two of them), and have some problems with it.
Frequent device timeouts as you said, already changed the router, wifi
channel, and still the problem.
Sometimes programs takes some time to connect, like when I'm
browsing the web and open irc, it takes some time to connect with the
server.
Usually timeouts happens when I do a new connection, like I never had a
timeout when only watching a video on youtube, it happens like when I open a new
site.
It has been difficult to me to figure out what causes this error, and
try to reproduce it, it looks totally random, some days I get a lot of
timeouts, others I get no one.
--
Regards

Henrique Lengler

Reply | Threaded
Open this post in threaded view
|

Re: Another working USB WiFi adapter:

Stefan Sperling-5
On Fri, Sep 04, 2015 at 02:52:19PM -0300, Henrique Lengler wrote:
> It has been difficult to me to figure out what causes this error, and
> try to reproduce it, it looks totally random, some days I get a lot of
> timeouts, others I get no one.

A "device timeout" means that an outgoing packet could not be transmitted
during a period of 5 seconds. Assuming the driver works correctly, what
you're seeing could simply be a matter of heavy contention for the medium.

How many other wifi networks are "active" around you when the timeout occurs?
Perhaps monitor traffic on your channel (from another box with a wireless
interface in monitor mode on the same channel) continously, and check if
bursts of traffic from other networks coincide with the athn device giving up.