Agreed. Since the OP was presented with conflicting info I simply provided a little "training" to let the OP know that I actually know WTF I'm talking about (sometimes) and that /29 will work. When all else fails, try it and see what happens!
----- Original Message -----
From: freeradius-users-bounces+ggatten=waddell.com@lists.freeradius.org <freeradius-users-bounces+ggatten=waddell.com@lists.freeradius.org>
To: FreeRadius users mailing list <freeradius-users@lists.freeradius.org>
Sent: Fri Aug 21 02:15:31 2009
Subject: Re: NAS IPs
Gary Gatten wrote:
> If I understand you correctly – I respectfully submit you are
> incorrect. When using VLSM / classless masks to define a NETWORK, you
> are correct (I haven’t checked the math – just assuming you are
> correct). However, when defining ACL type stuff, the proposed /29 would
> be perfectly OK. That said, I’m speaking strictly of my experience with
> routers, firewalls, etc. If FR supports this I can’t say for sure.
>
>
>
> ------------------------------------------------------------------------
>
>
> you can not to use /29 for 8 IPs because 3 bits have exactly 8
> combination ,..x000 is the network and ...x111 is the difusion ,then
> really you have 6 IPs with /29, you need /28 16 combinations minus
> two,min 14 IPs for 8 numbers.
>
I think we should stop the TCPIP training.
This is the freeradius list.
The user wants to know if /29 can be used.
The answer is yes.
--
Johan Meiring
Cape PC Services CC
Tel: (021) 883-8271
Fax: (021) 886-7782
-
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html