Re: [RADIATOR] (P)EAP flow

2014-02-19 Thread Heikki Vatiainen
On 02/17/2014 05:16 PM, Garry Shtern wrote:

 Would it make sense not modify Radiator behavior to only send reject if
 the OpenSSL returns mismatch rather than unexpected record?

Then there would need to be a correct request coming in later that
allows the authentication to continue? That is, if the request is not
rejected and can not be challenged, then the option would be to wait for
the real request?

 This way if
 there is a packet loss or intermittent client issues, the client doesn't
 get kicked off the net.

I would say it might be a better idea to see how to minimise the number
of unexpected messages. Would that be an option to explore?

Thanks,
Heikki

 Thanks.
 
 
 
 Sent with Good (www.good.com)
 
 
 -Original Message-
 *From: *Heikki Vatiainen [h...@open.com.au mailto:h...@open.com.au]
 *Sent: *Monday, February 17, 2014 02:22 PM Coordinated Universal Time
 *To: *radiator@open.com.au
 *Subject: *Re: [RADIATOR] (P)EAP flow
 
 On 02/14/2014 07:17 PM, Garry Shtern wrote:
 I have noticed that if Radiator receives a midstream EAP exchange
 message, it responds back with a CHALLENGE.
 
 I would expect something like this with PEAP.
 
 ERR: EAP TLS error: -1, 1, 8465,  13062: 1 - error:140940F5:SSL
 routines:SSL3_READ_BYTES:unexpected record
 
 Then an Access-Reject is sent back to the client.
 
 I am trying to understand
 what exactly happens at this point.  Does the Supplicant respond to the
 challenge with a brand new exchange or just retransmits whatever packet
 it sent before?  If it’s the latter, is there any way to force a
 supplicant to re-start the negotiation, perhaps with a crafted CHALLENGE?
 
 The supplicant probably restarts, but that's only because it got an
 unexpected response. I most cases I would expect that a midstream EAP
 message results as a some sort of error on Radiator side.
 
 Thanks,
 Heikki
 
 --
 Heikki Vatiainen h...@open.com.au
 
 Radiator: the most portable, flexible and configurable RADIUS server
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
 TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
 NetWare etc.
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator
 


-- 
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

Re: [RADIATOR] Cisco NX-OS TACACS+ problems

2014-02-19 Thread Heikki Vatiainen
On 02/19/2014 04:40 PM, Caporossi, Steve G. wrote:
 We upgraded to version 5.2(9) last weekend and our problem appears to be 
 solved.
 
 Thanks for keeping this on your radar.

Good to hear. Thanks for letting us know the problem was solved.

Maybe NX-OS devices Alexander mentioned are still using a version of
NX-OS that does not have the patch? A quick look tells there are not as
many different software trains as there are/were for IOS, but there are
plenty of minor releases still.

Thanks,
Heikki

-- 
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator


Re: [RADIATOR] (P)EAP flow

2014-02-19 Thread Heikki Vatiainen
On 02/19/2014 04:48 PM, Garry Shtern wrote:

 Actually, I was thinking perhaps if the Radiator is getting unexpected packet 
 from the supplicant to challenge the supplicant to restart the negotiation.  
 If that is possible, then the reject would only be sent if Radiator got all 
 the packets during the exchange but OpenSSL rejected this because of 
 certificate, negotiation or handshake errors.

Hmm, I don't think restarting an ongoing EAP authentication is possible.

Also, the incoming EAP messages are fed to OpenSSL as they come, not
collected together first, and the authentication continues based on what
OpenSSL returns. In other words, the TLS part, for example PEAP phase 1,
is mostly the EAP supplicant talking to OpenSSL libraries and Radiator
sending appropriate RADIUS messages based on what the SSL libraries return.

Thanks,
Heikki

 As for minimizing of unexpected messages, I am definitely with you on this 
 one.
  
 -Original Message-
 From: Heikki Vatiainen [mailto:h...@open.com.au] 
 Sent: Wednesday, February 19, 2014 9:35 AM
 To: Garry Shtern; 'radiator@open.com.au'
 Subject: Re: [RADIATOR] (P)EAP flow
 
 On 02/17/2014 05:16 PM, Garry Shtern wrote:
 
 Would it make sense not modify Radiator behavior to only send reject 
 if the OpenSSL returns mismatch rather than unexpected record?
 
 Then there would need to be a correct request coming in later that allows the 
 authentication to continue? That is, if the request is not rejected and can 
 not be challenged, then the option would be to wait for the real request?
 
 This way if
 there is a packet loss or intermittent client issues, the client 
 doesn't get kicked off the net.
 
 I would say it might be a better idea to see how to minimise the number of 
 unexpected messages. Would that be an option to explore?
 
 Thanks,
 Heikki
 
 Thanks.



 Sent with Good (www.good.com)


 -Original Message-
 *From: *Heikki Vatiainen [h...@open.com.au mailto:h...@open.com.au]
 *Sent: *Monday, February 17, 2014 02:22 PM Coordinated Universal Time
 *To: *radiator@open.com.au
 *Subject: *Re: [RADIATOR] (P)EAP flow

 On 02/14/2014 07:17 PM, Garry Shtern wrote:
 I have noticed that if Radiator receives a midstream EAP exchange 
 message, it responds back with a CHALLENGE.

 I would expect something like this with PEAP.

 ERR: EAP TLS error: -1, 1, 8465,  13062: 1 - error:140940F5:SSL 
 routines:SSL3_READ_BYTES:unexpected record

 Then an Access-Reject is sent back to the client.

 I am trying to understand
 what exactly happens at this point.  Does the Supplicant respond to 
 the challenge with a brand new exchange or just retransmits whatever 
 packet it sent before?  If it’s the latter, is there any way to force 
 a supplicant to re-start the negotiation, perhaps with a crafted CHALLENGE?

 The supplicant probably restarts, but that's only because it got an 
 unexpected response. I most cases I would expect that a midstream EAP 
 message results as a some sort of error on Radiator side.

 Thanks,
 Heikki

 --
 Heikki Vatiainen h...@open.com.au

 Radiator: the most portable, flexible and configurable RADIUS server 
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald, 
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, 
 TLS, TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP, 
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS, 
 NetWare etc.
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator

 
 
 --
 Heikki Vatiainen h...@open.com.au
 
 Radiator: the most portable, flexible and configurable RADIUS server 
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald, 
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS, TTLS, 
 PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP, DIAMETER etc. Full 
 source on Unix, Windows, MacOSX, Solaris, VMS, NetWare etc.
 


-- 
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

[RADIATOR] Delayed Stop Record and Active Sessions

2014-02-19 Thread rohan.henry @cwjamaica.com
Hello,

How can fix an issue where the DeleteQuery statement in my Sessions DB
config deletes the row for a new active session because of a delayed Stop
record?

Scenario:

1. A session is up (and row entered in the database for active session)
2. The session is dropped because of a premature disconnection (eg. modem
line cable unplugged) but Stop record is delayed.
3. New session is created after modem line cable is restored (and after
DeleteQuery statement removes database row for previous session)
4. The delayed Stop record finally comes in - the DeleteQuery statement now
removes the row for the active session (An unwanted behavior).

How do I compensate for the delayed Stop record that is causing active
session database records to be deleted?

Rohan
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

[RADIATOR] User-Password vs Cleartext-Password

2014-02-19 Thread Grant Spradling
I’m running Radiator in AuthBy FREERADIUSSQL mode off of a FreeRADIUS sql user 
database.

FreeRADIUS authenticates users when the Cleartext-Password attribute is used 
for the user password, however when Radiator authenticates against the same 
database, I get:

Access Rejected for bob: Check item Cleartext-Password expression ‘mypass’ does 
not match ‘’ in request

If I change the default to return attribute to User-Password instead of 
Cleartext-Password, the user can authenticate.

I’m using PAP and if I print the password from the request (%P) it matches the 
one stored in the database.

Thanks for your help,

Grant Spradling


___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator


Re: [RADIATOR] User-Password vs Cleartext-Password

2014-02-19 Thread Alan Buxey
Because you are using a freeradius-ism

Cleartext-Password is an internal attribute of freeradius.  The real attribute 
in the request is User-Password ... Use that on other servers. 

Alan
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

Re: [RADIATOR] EAP TLS issues routines:SSL3_READ_BYTES:tlsv1 alert access denied

2014-02-19 Thread Heikki Vatiainen
On 02/19/2014 10:08 PM, Jeffrey Smith wrote:

 Wed Feb 19 10:59:58 2014: ERR: EAP PEAP TLS read failed:  13601: 1 -
 error:14094419:SSL routines:SSL3_READ_BYTES:tlsv1 alert access denied

Here's one more possibility from the list archives:
http://www.open.com.au/pipermail/radiator/2004-August/009982.html

I agree with Alan that the AP client probably does not care but the
other client does.

In addition to what has already been suggested, I'd check the Radiator
certificate to see the Extended Key Usage (EKU) is there.

http://support.microsoft.com/kb/814394

Thanks,
Heikki


-- 
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator


Re: [RADIATOR] Delayed Stop Record and Active Sessions

2014-02-19 Thread Heikki Vatiainen
On 02/19/2014 09:22 PM, rohan.henry @cwjamaica.com wrote:

 How can fix an issue where the DeleteQuery statement in my Sessions DB
 config deletes the row for a new active session because of a delayed
 Stop record?

A quick idea: Do you think the DeleteQuery could be changed to include
Acct-Session-Id in the query. That is, the NAS-Port, etc, and
Acct-Session-Id must match the existing entry.

If the session has been replaced, the delete will not match any rows
because the new entry on the row it would otherwise match has a
different session id that belongs to the new session.

Please let us know how this works.
Thanks,
Heikki


 Scenario:
 
 1. A session is up (and row entered in the database for active session)
 2. The session is dropped because of a premature disconnection (eg.
 modem line cable unplugged) but Stop record is delayed.
 3. New session is created after modem line cable is restored (and after
 DeleteQuery statement removes database row for previous session)
 4. The delayed Stop record finally comes in - the DeleteQuery statement
 now removes the row for the active session (An unwanted behavior).
 
 How do I compensate for the delayed Stop record that is causing active
 session database records to be deleted?


-- 
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator


Re: [RADIATOR] EAP TLS issues routines:SSL3_READ_BYTES:tlsv1 alert access denied

2014-02-19 Thread Jeffrey Smith
Heikki,
  Thanks for the links.  I did come across that in my Googling.  My
certificate reports:

X509v3 Extended Key Usage:
TLS Web Server Authentication, TLS Web Client Authentication

It doesn't mention the OID specifically just the text as given.  The cert
is from RapidSSL as an aside.  Other clients treat this differently as
well.  An android device will successfully auth according to the debug logs
but never connects to the AP as it seems to timeout. And a Mac OSX device
just authenticates successfully over and over and over again, per the debug
logs, without connecting.  Its really bothersome that all the devices
aren't behaving the same way, since I have the feeling if I can find a way
to fix it for one the others will continue to fail.

Given that, I'm at a loss on how to continue to debug this issue.  Do you
have any other suggestions or can I provide any more logs?

Alan,
  To make sure I'm on the same page with you, I'm guessing by supplicant
you mean the wireless client (in this case a Windows 7 laptop)? There's no
configuration that pops up immediately on that one.  I tell it to connect
to the network and it pops up a username / password dialog no other options
to set.

I'm under the impression that no certs need to be installed on clients for
this to function correctly, is that the case?

Thanks,
Jeff Smith
Network Engineer
Neonova Network Services
(919) 460-3330
d...@neonova.net


On Wed, Feb 19, 2014 at 3:32 PM, Heikki Vatiainen h...@open.com.au wrote:

On 02/19/2014 10:08 PM, Jeffrey Smith wrote:

 Wed Feb 19 10:59:58 2014: ERR: EAP PEAP TLS read failed:  13601: 1 -
 error:14094419:SSL routines:SSL3_READ_BYTES:tlsv1 alert access denied

Here's one more possibility from the list archives:
http://www.open.com.au/pipermail/radiator/2004-August/009982.html

I agree with Alan that the AP client probably does not care but the
other client does.

In addition to what has already been suggested, I'd check the Radiator
certificate to see the Extended Key Usage (EKU) is there.

http://support.microsoft.com/kb/814394

Thanks,
Heikki


--
Heikki Vatiainen h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

Re: [RADIATOR] EAP TLS issues routines:SSL3_READ_BYTES:tlsv1 alert access denied

2014-02-19 Thread Sami Keski-Kasari
Hello Jeff,

I think that Android and MACOSX problems will be solved if you add
configuration parameter  AutoMPPEKeys to outer handler.

It is needed so that encryption keys to WLAN connection can be calculated.

In windows case:
Because client is sending that alert message it is hard to say exact
reason without seeing your client configuration.

Do you have your CA certificate installed in your windows machine?
You probably need to go to the wireless settings and check what CA
certificates are accepted for your connection.

Best Regards,
 Sami


On 02/19/2014 11:02 PM, Jeffrey Smith wrote:
 Heikki,
   Thanks for the links.  I did come across that in my Googling.  My
 certificate reports:
 
 X509v3 Extended Key Usage:
 TLS Web Server Authentication, TLS Web Client Authentication
 
 It doesn't mention the OID specifically just the text as given.  The cert
 is from RapidSSL as an aside.  Other clients treat this differently as
 well.  An android device will successfully auth according to the debug logs
 but never connects to the AP as it seems to timeout. And a Mac OSX device
 just authenticates successfully over and over and over again, per the debug
 logs, without connecting.  Its really bothersome that all the devices
 aren't behaving the same way, since I have the feeling if I can find a way
 to fix it for one the others will continue to fail.
 
 Given that, I'm at a loss on how to continue to debug this issue.  Do you
 have any other suggestions or can I provide any more logs?
 
 Alan,
   To make sure I'm on the same page with you, I'm guessing by supplicant
 you mean the wireless client (in this case a Windows 7 laptop)? There's no
 configuration that pops up immediately on that one.  I tell it to connect
 to the network and it pops up a username / password dialog no other options
 to set.
 
 I'm under the impression that no certs need to be installed on clients for
 this to function correctly, is that the case?
 
 Thanks,
 Jeff Smith
 Network Engineer
 Neonova Network Services
 (919) 460-3330
 d...@neonova.net
 
 
 On Wed, Feb 19, 2014 at 3:32 PM, Heikki Vatiainen h...@open.com.au wrote:
 
 On 02/19/2014 10:08 PM, Jeffrey Smith wrote:
 
 Wed Feb 19 10:59:58 2014: ERR: EAP PEAP TLS read failed:  13601: 1 -
 error:14094419:SSL routines:SSL3_READ_BYTES:tlsv1 alert access denied
 
 Here's one more possibility from the list archives:
 http://www.open.com.au/pipermail/radiator/2004-August/009982.html
 
 I agree with Alan that the AP client probably does not care but the
 other client does.
 
 In addition to what has already been suggested, I'd check the Radiator
 certificate to see the Extended Key Usage (EKU) is there.
 
 http://support.microsoft.com/kb/814394
 
 Thanks,
 Heikki
 
 
 --
 Heikki Vatiainen h...@open.com.au
 
 Radiator: the most portable, flexible and configurable RADIUS server
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
 TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
 NetWare etc.
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator
 
 
 
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator
 


-- 
Sami Keski-Kasari sam...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
NetWare etc.
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator


Re: [RADIATOR] EAP TLS issues routines:SSL3_READ_BYTES:tlsv1 alert access denied

2014-02-19 Thread Jeffrey Smith
Sami,
  Thanks for the AutoMPPEKeys, that did in fact fix OSX and Android.  I'm
hoping that the cert doesn't need to be installed on the windows clients as
this is for a widespread WISP solution for end users.

I did find one other oddity that may or may not also be certificate
related.  For ChromeOS it gets back an EAP MSCHAP-V2 Authentication failure
for the user:

Wed Feb 19 13:12:28 2014: DEBUG: Handling request with Handler
'TunnelledByPEAP=1', Identifier ''

Wed Feb 19 13:12:28 2014: DEBUG: internal Deleting session for
testu...@neonova.net, 137.118.48.15, 0

Wed Feb 19 13:12:28 2014: DEBUG: Handling with Radius::AuthMassGeneric:

Wed Feb 19 13:12:28 2014: DEBUG: Handling with EAP: code 2, 20, 70, 26

Wed Feb 19 13:12:28 2014: DEBUG: Response type 26

Wed Feb 19 13:12:28 2014: DEBUG: Reading users file
/usr/local/raddb/users/ppp/neonova.net

Wed Feb 19 13:12:28 2014: DEBUG: Radius::AuthMassGeneric looks for match
with testu...@neonova.net [testu...@neonova.net]

Wed Feb 19 13:12:28 2014: DEBUG: Radius::AuthMassGeneric ACCEPT: :
testu...@neonova.net [testu...@neonova.net]

Wed Feb 19 13:12:28 2014: DEBUG: EAP Failure, elapsed time 0.115332

Wed Feb 19 13:12:28 2014: DEBUG: EAP result: 1, EAP MSCHAP-V2
Authentication failure

Wed Feb 19 13:12:28 2014: DEBUG: AuthBy MassGeneric result: REJECT, EAP
MSCHAP-V2 Authentication failure

Wed Feb 19 13:12:28 2014: INFO: Access rejected for d...@neonova.net: EAP
MSCHAP-V2 Authentication failure

But I'm not seeing what is causing the Auth Failure.  I'm at Trace level
6.  Increasing that number doesn't appear to garner anymore data.

Thanks,
Jeff Smith
Network Engineer
Neonova Network Services
(919) 460-3330
d...@neonova.net


On Wed, Feb 19, 2014 at 4:14 PM, Sami Keski-Kasari sam...@open.com.auwrote:

Hello Jeff,

I think that Android and MACOSX problems will be solved if you add
configuration parameter  AutoMPPEKeys to outer handler.

It is needed so that encryption keys to WLAN connection can be calculated.

In windows case:
Because client is sending that alert message it is hard to say exact
reason without seeing your client configuration.

Do you have your CA certificate installed in your windows machine?
You probably need to go to the wireless settings and check what CA
certificates are accepted for your connection.

Best Regards,
 Sami


On 02/19/2014 11:02 PM, Jeffrey Smith wrote:
 Heikki,
   Thanks for the links.  I did come across that in my Googling.  My
 certificate reports:

 X509v3 Extended Key Usage:
 TLS Web Server Authentication, TLS Web Client
Authentication

 It doesn't mention the OID specifically just the text as given.  The cert
 is from RapidSSL as an aside.  Other clients treat this differently as
 well.  An android device will successfully auth according to the debug
logs
 but never connects to the AP as it seems to timeout. And a Mac OSX device
 just authenticates successfully over and over and over again, per the
debug
 logs, without connecting.  Its really bothersome that all the devices
 aren't behaving the same way, since I have the feeling if I can find a way
 to fix it for one the others will continue to fail.

 Given that, I'm at a loss on how to continue to debug this issue.  Do you
 have any other suggestions or can I provide any more logs?

 Alan,
   To make sure I'm on the same page with you, I'm guessing by supplicant
 you mean the wireless client (in this case a Windows 7 laptop)? There's no
 configuration that pops up immediately on that one.  I tell it to connect
 to the network and it pops up a username / password dialog no other
options
 to set.

 I'm under the impression that no certs need to be installed on clients for
 this to function correctly, is that the case?

 Thanks,
 Jeff Smith
 Network Engineer
 Neonova Network Services
 (919) 460-3330
 d...@neonova.net


 On Wed, Feb 19, 2014 at 3:32 PM, Heikki Vatiainen h...@open.com.au wrote:

 On 02/19/2014 10:08 PM, Jeffrey Smith wrote:

 Wed Feb 19 10:59:58 2014: ERR: EAP PEAP TLS read failed:  13601: 1 -
 error:14094419:SSL routines:SSL3_READ_BYTES:tlsv1 alert access denied

 Here's one more possibility from the list archives:
 http://www.open.com.au/pipermail/radiator/2004-August/009982.html

 I agree with Alan that the AP client probably does not care but the
 other client does.

 In addition to what has already been suggested, I'd check the Radiator
 certificate to see the Extended Key Usage (EKU) is there.

 http://support.microsoft.com/kb/814394

 Thanks,
 Heikki


 --
 Heikki Vatiainen h...@open.com.au

 Radiator: the most portable, flexible and configurable RADIUS server
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
 TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
 NetWare etc.
 ___
 radiator mailing list
 

Re: [RADIATOR] SIP2 + Fortigate setup

2014-02-19 Thread Chad Roseburg
Thanks Heikki ~ there is an option to change the authentication scheme. I
changed it to PAP as you suggest.

Now it appears as though the fortigate is sending the password encrypted
...Ex:

Test credentials:
user: 29030pretend
pass: gulash

Server output excerpt:
DEBUG: SIP2 send '2300020140219141804AO|AA29030pretend|ACterminal
password|AD�$.%�6Է!H�'

In looking at the docs, I see several encryption/decrypt options ...what do
I include in my config to allow Radiator to decrypt
this password?

Thank you!

Chad





On Sat, Feb 15, 2014 at 12:32 AM, Heikki Vatiainen h...@open.com.au wrote:

 On 02/15/2014 02:42 AM, Chad Roseburg wrote:
  I have an evaluation version of Radiator 4.12.1. I need to set up a web
  captive portal on a Fortigate 60D that uses SIP2 authentication.
 
  The SIP2 part works ...tests successful:

 Hello Chad,

 radpwtst uses PAP with the options you have specified and sends
 User-Password which can be then used with AuthBy SIP2.

 However, it looks like the Fortigate is trying to do MS-CHAP instead of
 PAP. With MS-CHAP there is not password, only a challenge and response,
 and for this reason it does not work.

 Presence of MS-CHAP-Challenge without User-Password indicates MS-CHAP is
 tried. There should be a MS-CHAP-Response too with the attributes, but
 maybe you have left that out. These two attributes are used by MS-CHAP.

 See if there's 'Authentication Scheme', I think this is the option in
 Fortigate, or something similar that has been set to MS-CHAP or defaults
 to MS-CHAP. There should be an option to switch it to PAP.

 Please let us know if the above helps.

 Thanks,
 Heikki


  Ex.
  perl radpwtst -noacct -user 29030pretend -password secrets
  sending Access-Request...
  OK
 
  On RADIUS server I see:
  -
  Fri Feb 14 16:07:47 2014: DEBUG: SIP2 send '2300020140214
   160747AONCRL|AA29030pretend|ACterminal password|ADsecrets|'
  Fri Feb 14 16:07:47 2014: DEBUG: SIP2 read '24  00020140214
 160727AEJOE SMITH|AA29030pretend|BLY|CQY|AFGreetings. |AONCRL|'
  Fri Feb 14 16:07:47 2014: DEBUG: Radius::AuthSIP2 ACCEPT: : 29030pretend
  [29030pretend]
  Fri Feb 14 16:07:47 2014: DEBUG: AuthBy SIP2 result: ACCEPT
 
  But the second part is that I need to connect the fortigate to the
  RADIUS server. I add the fortigate as a client in the config using IP
  and a 'Secret'
 
  Here's some edited output when I test from the fortigate using the same
  creds:
  Fri Feb 14 16:23:44 2014: DEBUG: SIP2 send '2300020140214
   162344AONCRL|AA29030pretend|ACterminal password|AD|'
  Fri Feb 14 16:23:44 2014: DEBUG: SIP2 read '24  00020140214
 162323AEJOE SMITH|AA29030pretend|BLY|CQN|AFGreetings. |AONCRL|'
  Fri Feb 14 16:23:44 2014: DEBUG: Radius::AuthSIP2 REJECT: Bad password:
  29030002429839 [29030002429839]
  Fri Feb 14 16:23:44 2014: DEBUG: AuthBy SIP2 result: REJECT, Bad password
 
  It looks like it's not sending the password. Also, at the top of the
  transmission there's mention of a MS-CHAP-Challenge:
  Attributes:
  NAS-Identifier = Fortinet_RTR
  MS-CHAP-Challenge =
  b1372381464165145.9229163j129220M
  Acct-Session-Id = 0021
  Connect-Info = test
  Fortinet-Vdom-Name = root
 
  This is the Client config:
  Client 192.x.x.99
  Secret  secretspass
  DupInterval 0
  /Client
 
  Thanks for any advice!
 
  --
  Chad
 
 
  ___
  radiator mailing list
  radiator@open.com.au
  http://www.open.com.au/mailman/listinfo/radiator
 


 --
 Heikki Vatiainen h...@open.com.au

 Radiator: the most portable, flexible and configurable RADIUS server
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
 TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
 NetWare etc.
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator




-- 
Chad Roseburg
Automation Dept.
North Central Regional Library
___
radiator mailing list
radiator@open.com.au
http://www.open.com.au/mailman/listinfo/radiator

Re: [RADIATOR] SIP2 + Fortigate setup

2014-02-19 Thread Hugh Irvine

Hello Chad -

You don’t need to do anything special - Radiator will process the password 
automatically.

If you are using a flat file for your user records you should add an entry like 
this:



# flat file user definitions

29030pretend  User-Password = gulash



hope that helps

regards

Hugh


On 20 Feb 2014, at 09:42, Chad Roseburg croseb...@ncrl.org wrote:

 Thanks Heikki ~ there is an option to change the authentication scheme. I 
 changed it to PAP as you suggest. 
 
 Now it appears as though the fortigate is sending the password encrypted 
 ...Ex:
 
 Test credentials:
 user: 29030pretend
 pass: gulash
 
 Server output excerpt:
 DEBUG: SIP2 send '2300020140219141804AO|AA29030pretend|ACterminal 
 password|AD�$.%�6Է!H�'
 
 In looking at the docs, I see several encryption/decrypt options ...what do I 
 include in my config to allow Radiator to decrypt
 this password?
 
 Thank you!
 
 Chad
 
 
 
 
 
 On Sat, Feb 15, 2014 at 12:32 AM, Heikki Vatiainen h...@open.com.au wrote:
 On 02/15/2014 02:42 AM, Chad Roseburg wrote:
  I have an evaluation version of Radiator 4.12.1. I need to set up a web
  captive portal on a Fortigate 60D that uses SIP2 authentication.
 
  The SIP2 part works ...tests successful:
 
 Hello Chad,
 
 radpwtst uses PAP with the options you have specified and sends
 User-Password which can be then used with AuthBy SIP2.
 
 However, it looks like the Fortigate is trying to do MS-CHAP instead of
 PAP. With MS-CHAP there is not password, only a challenge and response,
 and for this reason it does not work.
 
 Presence of MS-CHAP-Challenge without User-Password indicates MS-CHAP is
 tried. There should be a MS-CHAP-Response too with the attributes, but
 maybe you have left that out. These two attributes are used by MS-CHAP.
 
 See if there's 'Authentication Scheme', I think this is the option in
 Fortigate, or something similar that has been set to MS-CHAP or defaults
 to MS-CHAP. There should be an option to switch it to PAP.
 
 Please let us know if the above helps.
 
 Thanks,
 Heikki
 
 
  Ex.
  perl radpwtst -noacct -user 29030pretend -password secrets
  sending Access-Request...
  OK
 
  On RADIUS server I see:
  -
  Fri Feb 14 16:07:47 2014: DEBUG: SIP2 send '2300020140214
   160747AONCRL|AA29030pretend|ACterminal password|ADsecrets|'
  Fri Feb 14 16:07:47 2014: DEBUG: SIP2 read '24  00020140214
 160727AEJOE SMITH|AA29030pretend|BLY|CQY|AFGreetings. |AONCRL|'
  Fri Feb 14 16:07:47 2014: DEBUG: Radius::AuthSIP2 ACCEPT: : 29030pretend
  [29030pretend]
  Fri Feb 14 16:07:47 2014: DEBUG: AuthBy SIP2 result: ACCEPT
 
  But the second part is that I need to connect the fortigate to the
  RADIUS server. I add the fortigate as a client in the config using IP
  and a 'Secret'
 
  Here's some edited output when I test from the fortigate using the same
  creds:
  Fri Feb 14 16:23:44 2014: DEBUG: SIP2 send '2300020140214
   162344AONCRL|AA29030pretend|ACterminal password|AD|'
  Fri Feb 14 16:23:44 2014: DEBUG: SIP2 read '24  00020140214
 162323AEJOE SMITH|AA29030pretend|BLY|CQN|AFGreetings. |AONCRL|'
  Fri Feb 14 16:23:44 2014: DEBUG: Radius::AuthSIP2 REJECT: Bad password:
  29030002429839 [29030002429839]
  Fri Feb 14 16:23:44 2014: DEBUG: AuthBy SIP2 result: REJECT, Bad password
 
  It looks like it's not sending the password. Also, at the top of the
  transmission there's mention of a MS-CHAP-Challenge:
  Attributes:
  NAS-Identifier = Fortinet_RTR
  MS-CHAP-Challenge =
  b1372381464165145.9229163j129220M
  Acct-Session-Id = 0021
  Connect-Info = test
  Fortinet-Vdom-Name = root
 
  This is the Client config:
  Client 192.x.x.99
  Secret  secretspass
  DupInterval 0
  /Client
 
  Thanks for any advice!
 
  --
  Chad
 
 
  ___
  radiator mailing list
  radiator@open.com.au
  http://www.open.com.au/mailman/listinfo/radiator
 
 
 
 --
 Heikki Vatiainen h...@open.com.au
 
 Radiator: the most portable, flexible and configurable RADIUS server
 anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald,
 Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS,
 TTLS, PEAP, TNC, WiMAX, RSA, Vasco, Yubikey, MOTP, HOTP, TOTP,
 DIAMETER etc. Full source on Unix, Windows, MacOSX, Solaris, VMS,
 NetWare etc.
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator
 
 
 
 -- 
 Chad Roseburg
 Automation Dept.
 North Central Regional Library
 ___
 radiator mailing list
 radiator@open.com.au
 http://www.open.com.au/mailman/listinfo/radiator


--

Hugh Irvine
h...@open.com.au

Radiator: the most portable, flexible and configurable RADIUS server 
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald, 
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP,