Alfa AWUS036H testing results with different driver sources

[linuxchuck]

great findings linuxchuck i learned a new cmd "watch" thats pimp, i'd like to make a small addition that ive noticed while trying to connect to my unencrypted AP this is what i have to do

ifconfig wlan0 up
iwconfig wlan0 essid "essid " notice the space, the driver im using likes to cut off the last character
then i do another

ifconfig wlan0 up
dhcpcd wlan0

and it magically works for me
using the older rtl8187_linux_26.1010.0622.2006/ patch some1 posted awhile ago via the btwiki HCL sniper or sumthin was his name i dunno but i sure cant wait for bt3 beta i'll have a renued lust to "hack that bitch up"

Heh, glad to hear you got a new technique from my post. Laziness tends to be a good motivator for such finds, and I'm plenty lazy. I became quite tired of having to continually run iwconfig just to see what was going on.

That's a strange issue on the last character of your essid being cut off. I did not see that problem on either of the newer drivers I tested. On both sets, I was able to associate without any issues.

I suppose I should have started with the default BT2 drivers, but after hanging in IRC with some of the aircrack-ng guys, and listening to their suggestions, I thought it would be best to jump straight to a newly compiled set. If we weren't so close to the BT3 release, I'd just fire up an old BT2 install, and give the built-in drivers a full test-run. But, since it's going to be released tomorrow, I just plan to wait for it, and give a full review on the new ones.

Thanks for the info, and I appreciate the comments!

[merlin051]

iI've experienced the driver cutting the last letter of the essid off, i was using the drivers from pilotsnipes and messing around with them trying to egt them to work, not sure what i did to produce the error but its these ones that do it.

[linuxchuck]

iI've experienced the driver cutting the last letter of the essid off, i was using the drivers from pilotsnipes and messing around with them trying to egt them to work, not sure what i did to produce the error but its these ones that do it.

That's good to know. Thanks for the verification.

I recall seeing those drivers while researching prior to purchasing my Alfa. I probably would have given them a try, but I almost always lean toward the "roll your own" standard in an attempt to learn as much as I can while doing it.

Glad to hear from you on this, thanks for the comment!

[ShadowKill]

Anyone able to give an update as to whether or not the Alfa is running stable under BT3?

[merlin051]

stable out of the box with BT3, had no problems with it.

[penguin_to_bits]

Is the Alfa truly the King of wireless?

Unfortunately I'm a bit clueless when it comes to drivers in Linux. As far as I know, there's two kinds of drivers:
1) Ones built into the kernel
2) Separate ones that you load as modules

Is that right? How do I go about installing and uninstalling drivers? Links are appreciated

Do many people modify the Linux kernel in order to add drivers? If so, why do they do this rather than just loading a module?

And just to show even more of my naivety: Is there only one Linux kernel? i.e. the one maintained by Linus Torvalds?

Does BT3 beta have the latest Linux kernel? If not, then why? Is there anything wrong with changing the kernel to a newer one? Also another thing I've been curious about... I always hear about people compiling the Linux kernel, but why isn't it distributed as a binary?

[Padawan]

Nice information!

Did you test WPA access? As far as I could see (because I have that Alfa), wpa_supplicant fails everytime I try to use it.
Another user on this forum reported to experience the same.

According to me, this is an important fact, because outside of testing, I don't want to use my AP in WEP mode.

[Padawan]

Anyone able to give an update as to whether or not the Alfa is running stable under BT3?

Stable here, yes. But wpa_supplicant doesn't work with this card.

[ShadowHatesYou]

Can you guys post a "dmesg | grep rtl8187" with your posts? It appears there's different versions of the rtl8187 radios. I own an awus036h, and I love it. Works almost perfectly without toying with it.

Anyways, my observations, from backtrack 2 with the default kernel:

rtl8187: Initializing module
rtl8187: Wireless extensions version 21
rtl8187: Initializing proc filesystem
rtl8187: Enabling 14 channels.
rtl8187: MAC chip version: 04
rtl8187: Card type: AD
rtl8187: Reported EEPROM chip is a 93c46 (1Kbit)
rtl8187: Card MAC address is 00:c0:ca:19:f4:75
rtl8187: RF Chip ID: 05
rtl8187: Card reports RF frontend Realtek 8225
rtl8187: WW:This driver has EXPERIMENTAL support for this chipset.
rtl8187: WW:use it with care and at your own risk and
rtl8187: WW:**PLEASE** REPORT SUCCESS/INSUCCESS TO andreamrl@tiscali.it
rtl8187: This seems a new V2 radio
rtl8187: PAPE from CONFIG2: 0
rtl8187: Driver probe completed
usbcore: registered new interface driver rtl8187

iwconfig essid blah will drop a character. I just generally repeat the last character.

bt rc.d # iwconfig wlan0 essid fail
bt rc.d # iwconfig wlan0 | grep ESSID
wlan0 802.11b/g ESSID:"fai"
bt rc.d #

Changing a channel via iwconfig fails, as described in a post above, but airmon-ng works just fine, takes a moment though.

bt rc.d # airmon-ng start wlan0 3


Interface Chipset Driver

wlan0 RTL8187 r8187 (monitor mode enabled)

bt rc.d #

I can inject and the like with no issues. I've verified this with wireshark on my 2200bg when booted natively. I'm going to be making an alfa-specific bt2final automation plugin(yay automating things to boot directly into!) in the next few days, when I get my next spurt of motivation. I'll play with other driver sources when I'm working on it and see what works best with what.


PS: wpa_supplicant works fine for me, although I generally use my 2200bg to do any real net traffic. Having two reliable wireless cards kicks ass like Bruce Lee on phenylcyclidine.

[Padawan]

I'm using BT3 beta. That may explain the different. I will try with BT2 later.