Nas Type

[Roberto Greiner]

Peter Nixon wrote:
> On Fri 27 Jul 2007, Roberto Greiner wrote:
>> Hi,
>>
>> I was starting to look at checkrad, and found (based on
>> http://www.freeradius.org/radiusd/doc/Simultaneous-Use) that using
>> "other" as the NAS-type will actually check only radutmp instead of
>> looking at the actual NAS. Now, Could someone point me what would be the
>> proper NAS type to use for each of the devices below(or the proper
>> reference document to use)? I'm using the following NASes in my network:
>>
>> Monowall
>> pfSense
>> (3Com) Total Control
>> PopTop (in Linux)
>>
>> What I want to do is to use checkrad as one of the steps to make sure
>> that whoever appears as logged is really logged in, because I'm trying
>> to use Simultaneous-use check, and some of the above (notably monowall)
>> doesn't seem to be clearing properly sometimes.
> 
> As you have already found the docs you know the answer. The 3Com is obviously 
> type "tc". If its not on the list it's "other". 
> 
> However, if you write a patch to support the devices you mention, we would be 
> happy to include it in FreeRADIUS.
> 
> Cheers

I've re-checked the available options, and found that there is one nas
type for the Total Control, besides 'tc': "usrhiper". But there are a
few errors in the documentation speaking about it
(http://www.freeradius.org/radiusd/doc/Simultaneous-Use). The first is
the name itself. The page says "usrhyper", when the correct is
"usrhiper", with "i" instead of "y". The second is that it says that for
that option, the naspasswd file is not used, which is partially correct.
It can use naspasswd, and in that case the login name declared must be
SNMP, or it will fail.

For the other two devices (monowall and poptop), I don't know how to
proceed yet, since neither of them returns connected user information
through SNMP :-(

Thanks,

Roberto



-- 
  -----------------------------------------------------
                Marcos Roberto Greiner

   Os otimistas acham que estamos no melhor dos mundos
    Os pessimistas tem medo de que isto seja verdade
                                       Murphy
  -----------------------------------------------------