> > > > > Hi, Graeme: > > > > > > > > > > Just a brainstrom (probably stupid) idea.. On what arch are you trying > > > > > freeradius? 17152 is 0x4300.. Just wondered if radius meant 0x0043 instead. > > > > > > > > Linux i386 based. > > > > > > > > and the test packets are coming from the same machine.. ... shirley endianness > > > > isn't the problem? > > > > > > freeradius gives the same error for both paket length and radius > > > length field errors. perhaps your radius client (I guess it's not the > > > freeradius client?) builds mysterious pakets? > > > > uh, I'm lying. it says length > and that means the length field of the > > radius client. > > read: .... radius packet.
AHHHHHHRRRRGGGHHHH :) Indeed! looks like a bug in my code that writes the length to the packet. It is writing it Little-Endian.. DAMN! I guess icRADIUS isn't too fussy about a few things. checking the TCPDUMP and pumpin git through bc with ibase=16 gives the same value as freeradius (well off by one :) ). Thanks for your help -- ----- Graeme Hinchliffe (BSc) Core Internet Systems Designer Zen Internet (http://www.zen.co.uk) ICQ 3842605 (link) Sales : 0870 6000 971 Fax : 0870 6000 972 - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
