[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ipsec-tools-devel
Subject:    Re: [Ipsec-tools-devel] Ipsec-tools-devel Digest, Vol 11, Issue 7
From:       Matthew Grooms <mgrooms () shrew ! net>
Date:       2007-04-19 18:25:31
Message-ID: 4627B41B.7030808 () shrew ! net
[Download RAW message or body]

On 4/19/2007, "Andy Tang" <andy.atang@gmail.com> wrote:
 > >Thank you for the advice.
 > >Isn't that the authentication_algorithm line in the sainfo is for 
the ESP?
 > >How does one distinguish if the auth algorithm should be used by ESP 
and not
 > >AH?

If you specify an authentication_algorithm, racoon will attempt to 
negotiate hmac algorithm for ESP.  Notice, there is a non_auth option 
which I assume would be used to tell racoon to disable auth for ESP. To 
be honest, I have never tried to disable message authentication.

 > >I don't know how to disable the AH in QM?  I could not find such 
reference
 > >in man racoon.conf.

I don't know what QM is, but the AH algorithm is specified when adding 
an IPSEC policy to the kernel using setkey. Both your policies below 
also use transport mode which would be used for encrypting traffic only 
between the two IPSEC peers and not tunneling between two private 
networks. Is this what you want?

 > >
 > >The setkey -DP is showing something like this now:
 > >setkey -DP
 > >192.168.15.102[any] 192.168.15.103[any] any
 > >        in prio def ipsec
 > >        esp/transport//require
 > >        ah/transport//require
 > >        created: Apr 18 23:14:24 2007  lastused:
 > >        lifetime: 0(s) validtime: 0(s)
 > >        spid=248 seq=1 pid=3459
 > >        refcnt=1
 > >192.168.15.103[any] 192.168.15.102[any] any
 > >        out prio def ipsec
 > >        esp/transport//require
 > >        ah/transport//require
 > >        created: Apr 18 23:14:24 2007  lastused: Apr 18 23:14:37 2007
 > >        lifetime: 0(s) validtime: 0(s)
 > >        spid=241 seq=2 pid=3459
 > >        refcnt=2
 > >

When policies include multiple protocols, they are considered additive. 
In other words, your not using ESP or AH, your using ESP and AH. This 
causes racoon to expect the Vista box to negotiate a SA bundle and may 
explain why you are getting the protocol mismatch error.

I believe this was discussed recently on the list but my memory is 
fuzzy. It seems that there was a stock setkey wrapper script in a Linux 
distro ( maybe fedora ) that specified AH and ESP when auto building 
policies. Unfortunately, this not what a user wants the vast majority of 
the time. ( see setkey man page for more details ) Your best bet is to 
find the FC6 tool or script that specifies AH and ESP and remove AH.

Any FC6 admins out there that can point to documentation on the "right 
way" to configure IPSEC policy? In FreeBSD, you specify these in the 
/etc/ipsec.conf file.

I have a FC6 test box collecting dust that I could look at but I don't 
have access to it at the moment. If you can't figure this out, let me 
know and I will try to look at it later and give you some better advice.

 > >The weird thing is that I updated the /etc/racoon/psk.txt with a new
 > >preshared key.
 > >Then I ran "killall racoon" and restart the racoon daemon.
 > >When I started to initiate an IPsec session, I notice that the MM 
does not
 > >complete successful, when I looked at the racoon.log, I saw:
 > >
 > >2007-04-18 23:01:26: DEBUG: pfkey X_SPDDUMP failed: No such file or
 > >directory
 > >

I really can't comment too much on this. I would assume it means racoon 
couldn't open a PF_KEY socket. Are you starting it as root?

-Matthew

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Ipsec-tools-devel mailing list
Ipsec-tools-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ipsec-tools-devel
[prev in list] [next in list] [prev in thread] [next in thread]