Holger Schurig | 1 Mar 2010 12:08
Picon

Re: [PATCH] wpa_supplicant 0.6.X: fix FTBFS on Debian GNU/kFreebsd

Shouldn't you do this rather for wpasupplicant 0.7.1?

By staying at this old version, you thwart usage of cfg80211.
Kel Modderman | 1 Mar 2010 10:41
Picon
Favicon

Re: [PATCH] wpa_supplicant 0.6.X: fix FTBFS on Debian GNU/kFreebsd

On Monday 01 March 2010 21:08:21 Holger Schurig wrote:
> Shouldn't you do this rather for wpasupplicant 0.7.1?

What, like the patch in the other email?

> 
> By staying at this old version, you thwart usage of cfg80211.
> 

Debian squeeze is going to be frozen soon and for a few years. Do you think
it should be frozen with an early release of a development branch of
wpa_supplicant or a mature release of the current stable branch which has
gone through all the "user -> find bug -> report bug -> fix bug" cycles for
a long length of time?

What happens if there is some serious bug found in 0.7.1 release 1/2 years from
now, is Jouni or others likely to help me/Debian fix it? What if a fix for the
problem has been committed but is non-trivial to backport to 0.7.1? Is it less
likely to get help with the current stable release 1/2 years from now?

If Linux, and all its modules, is frozen at 2.6.32, how much benefit is it to
users to be using version 0.7.1 for the next couple of years? Do these benefits
override the advantage of having a version which has been used for a long time
and seems to work well enough now but maybe doesn't support the really new
features?

Thanks, Kel.
Holger Schurig | 1 Mar 2010 14:21
Picon

Re: [PATCH] wpa_supplicant 0.6.X: fix FTBFS on Debian GNU/kFreebsd

> What, like the patch in the other email?

Ah, I didn't get that the other patch was for wpa-supplicant 0.7.x.

However, even Debian unstable is still at wpa_supplicant 0.6.10, and AFAIK it 
still doesn't have a package for the crda, despite the fact that CRDA works 
flawlessly since months (I know about svn://svn.debian.org/pkg-wpa, but this 
is in limbo since more than half-a-year).

To be clear: I'm not requesting YOU to package this, in bug 536502 you said 
that you don't intend to do so.

Yet using this old wpasupplicant event for Debian unstable shows again that 
Debian's wireless support isn't that up to par. And because I bemoan this 
fact, I wrote my original message, which I regret now. Was kinda uninformend.

Best regards,
Holger

--

-- 
http://www.holgerschurig.de
Kel Modderman | 1 Mar 2010 14:07
Picon
Favicon

Re: [PATCH] wpa_supplicant 0.6.X: fix FTBFS on Debian GNU/kFreebsd

On Monday 01 March 2010 23:21:56 Holger Schurig wrote:
> > What, like the patch in the other email?
> 
> Ah, I didn't get that the other patch was for wpa-supplicant 0.7.x.
> 
> 
> However, even Debian unstable is still at wpa_supplicant 0.6.10, and AFAIK it 

There were serious questions in my previous email which I've considered in
selecting the 0.6.X branch of wpa_supplicant for the next stable release of
Debian. Debian unstable is the starting point for packages intended for the
next release while the package archive is unfrozen.

Please do offer suggestions as to why 0.7.X is a better candidate.

> still doesn't have a package for the crda, despite the fact that CRDA works 
> flawlessly since months (I know about svn://svn.debian.org/pkg-wpa, but this 
> is in limbo since more than half-a-year).

CRDA and wireless-regdb are proving difficult to integrate into the
distribution.

> 
> To be clear: I'm not requesting YOU to package this, in bug 536502 you said 
> that you don't intend to do so.
> 
> Yet using this old wpasupplicant event for Debian unstable shows again that 
> Debian's wireless support isn't that up to par. And because I bemoan this 
> fact, I wrote my original message, which I regret now. Was kinda uninformend.
> 
(Continue reading)

ikiran k | 1 Mar 2010 14:34
Picon

Radius Servers

Does hostapd support multiple radius server per station. What does it take to support multiple radius server per station?

Thanks & Regards,
Kiran

_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap
Holger Schurig | 2 Mar 2010 13:21
Picon

Re: [PATCH] wpa_supplicant 0.6.X: fix FTBFS on Debian GNU/kFreebsd

> Please do offer suggestions as to why 0.7.X is a better candidate.

Oh, there are some reasons.

WEXT is E.O.L.

0.6.x can only do WEXT. And WEXT is 802.11n stupid.

WEXT cannot do mac80211's mesh (okay, that's not an issue for wpa_supplicant)

Wireless developers are moving away from WEXT, e.g Luis just now wrote 
"Anyway, it would be a good thing to forget about iwconfig now.".

It's already difficult to backport some fixes from 0.7.x driver_wext.c to 
0.6's driver_wext.c

But hey, Joini said that 0.6.x continues to be the stable version, so all of 
my arguing is bogus if you're targeting Debian Stable. I just hope that once 
the next Debian stable is released, Debian unstable will get something newer 
:-)

--

-- 
http://www.holgerschurig.de
R C | 3 Mar 2010 01:11
Picon
Favicon

100,000 WPA supplicant requests

Hello,

I have a requirement to generate a 100,000 WPA supplicant requests. Here is my requirement in more detail:

I want to test the scalability of my server software. This server software and a Free Radius Server sit on the same linux box. I want to see how many client requests the Radius server can handle via EAP-TLS. And, in turn, I know how my server software can handle these requests simultaneously. We are looking at sending 100,000 wpa supplicant requests simultaneously to the free radius server.

1. Considering the limitation of the resources, I was thinking about how i can get so many (100,000) wpa supplicant sessions up and running at the same time. Do you have any input on this?

2.  Is there a free wpa supplicant (peer) that generates multiple separate sessions at the same time?

3. Can one interface be used to send out multiple simultaneous wpa supplicant requests? If yes, how many wpa client sessions can i generate on 1 interface?

4. Can you please tell me how good WPA supplicant and Hostapd fit this test with FreeRadius Server?

5. Are they good for scalability testing or should I look out for other open source wpa supplicants/NASs?

6. Do you know of any other supplicants and AP that might do the work for me or are known to do a better job in scalability tests like this?

Thank you very much for your time.

Regards,
RC.

_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap
R C | 3 Mar 2010 01:12
Picon
Favicon

Fw: 100,000 WPA supplicant requests

Hello,

I have a requirement to generate a 100,000 WPA supplicant requests. Here is my requirement in more detail:

I want to test the scalability of my server software. This server software and a Free Radius Server sit on the same linux box. I want to see how many client requests the Radius server can handle via EAP-TLS. And, in turn, I know how my server software can handle these requests simultaneously. We are looking at sending 100,000 wpa supplicant requests simultaneously to the free radius server.

1. Considering the limitation of the resources, I was thinking about how i can get so many (100,000) wpa supplicant sessions up and running at the same time. Do you have any input on this?

2.  Is there a free wpa supplicant (peer) that generates multiple separate sessions at the same time?

3. Can one interface be used to send out multiple simultaneous wpa supplicant requests? If yes, how many wpa client sessions can i generate on 1 interface?

4. Can you please tell me how good WPA supplicant and Hostapd fit this test with FreeRadius Server?

5. Are they good for scalability testing or should I look out for other open source wpa supplicants/NASs?

6. Do you know of any other supplicants and AP that might do the work for me or are known to do a better job in scalability tests like this?

Thank you very much for your time.

Regards,
RC.


_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap
Lewis Adam-VNQM87 | 3 Mar 2010 11:31

RE: Mutual TLS authentication in handshake phase of EAP-TTLS

Hello again.

  Is there anybody who can advise me as to what I can do next? It would
be really
useful to know whether mutual TLS authentication is supported for TTLS.
I don't
know if the question is valid or not appropriate for this forum. I
appreciate that
the functionality of eapol_test is not as critical as that of the
wpa_supplicant
but I would hope the answer would still be of general interest.

Regards,
Adam Lewis

> -----Original Message-----
> From: Lewis Adam-VNQM87 
> Sent: Thursday, February 25, 2010 11:42 AM
> To: 'hostap <at> lists.shmoo.com'
> Subject: Mutual TLS authentication in handshake phase of EAP-TTLS
> 
> Hi,
> 
>   apologies if this question has been answered elsewhere - I 
> looked but couldn't see anything, even in the "Mutual 
> EAP-TTLS Authentication" thread.
> 
> I am currently looking at the eapol_test code to see if I can 
> use it as a RADIUS client. I have ran eapol_test with various 
> EAP-TLS and EAP-TTLS configuration files, testing 
> successfully with a freeRADIUS server. Looking at the 
> EAP-TTLS RFC 5281, I have read the following: 
> 
> In EAP-TTLS, the TLS authentication may be mutual; or it may 
> be one-way, in which only the server is authenticated to the client.
> 
> My question is, does eapol_test currently allow mutual TLS 
> authentication for EAP-TTLS? If so, how do I configure it (or 
> the configuration files) to do so? I believe the tunnelled 
> protocol can also be TLS but I want to avoid this as I need 
> to have the ability to verify users rather than the client 
> (e.g. by doing user/password checks).
> 
> I'd appreciate any help you can give.
> 
> Regards,
> Adam Lewis.
> 
> 
Alessandro Nasorri | 3 Mar 2010 12:45
Picon
Favicon

(unknown)

 
Hello. I'm using wpa_supplicant as access requestor (AR) in a TNC scenario.
I know that TTLS method support TNC
Does anyone knows if wpa_supplicant client manage  PEAP as EAP method with TNC support.
I received a NAK from the wpa_supplicant when he receives an EAP_TYPE_TNC as method and then an MD5 method is negotiated.
Looking the code (src\eap\eap_peer.c) seems that TNC is supported also for PEAP.
 
# See /usr/doc/wpa_supplicant-0.5.10/wpa_supplicant.conf.sample
# for many more options that you can use in this file.
# This line enables the use of wpa_cli which is used by rc.wireless
# if possible (to check for successful association)
ctrl_interface=/var/run/wpa_supplicant
# By default, only root (group 0) may use wpa_cli
ctrl_interface_group=0
eapol_version=1
ap_scan=1
fast_reauth=1
# WPA protected network, supply your own ESSID and WPAPSK here:
network={
ssid="example 802.1X network"
key_mgmt=IEEE8021X
eap=PEAP
identity="usertest"
#phase2="auth=MSCHAPV2"
phase2="autheap=TNC"
password="password"
priority=10
}
# Plaintext connection (no WPA, no IEEE 802.1X),
# nice for hotel/airport types of WiFi network.
#network={
# key_mgmt=NONE
# priority=0
#}
 
 
Thanks a lot
Alessandro

Chiacchiera con i tuoi amici via Webcam su Messenger. Videochiamali!
_______________________________________________
HostAP mailing list
HostAP <at> lists.shmoo.com
http://lists.shmoo.com/mailman/listinfo/hostap

Gmane