Passing NAS-Identifier for radius generated stop packets. (updated patch, please consider)

[Matthew Schumacher]

Matthew Schumacher wrote:
> List,
> 
> My radius system uses the NAS-Identifer attribute to make some decisions
> on how to log or record utmp records.  Everything works fine until I
> have a situation where freeradius generates a stop packet in the
> session_zap function.  Because the stop packet doesn't have the
> NAS-Identifier attribute it breaks the way I log things.
> 
> I started hacking together a patch that would pass the NAS-Identifier of
> the access-request packet, but it doesn't work.  Can someone take a look
> at my patch, or perhaps suggest a better work around?
> 
> Thanks,
> schu

Ok, I have a working patch now, but would really appreciate it if
someone else would look it over as I am not very good at C.

Also, since the current code doesn't set any NAS-Identifier on generated
stop packets I think this patch should be considered for inclusion as it
is fairly common to have people make auth and acct decisions based on
the NAS-Identifier attribute and the change wouldn't break or change
anything currently in place.

Thanks,

schu
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: nas_ident_session_zap_passthough.diff
URL: <http://lists.freeradius.org/pipermail/freeradius-devel/attachments/20060728/8758f390/attachment.ksh>