"Alan Litster" <[EMAIL PROTECTED]> wrote:
... just the RADIUS packet:
> 0x001c 014e 007a Q..B......8..N.z
> 0x0020 eb3c 1a7f 3e7b 5a29 d15e cabb d7be a4a4 .<..>{Z).^......
> 0x0030 0406 c0a8 0026 010c 3837 3033 3430 3530 .....&..87034050
> 0x0040 3632 3d06 0000 0000 1f1f 7369 703a 3837 62=.......sip:87
> 0x0050 3033 3430 3530 3632 4073 6970 2e63 616c [EMAIL PROTECTED]
> 0x0060 6c75 6b2e 636f 6d1a 3500 0019 da18 2f68 luk.com.5...../h
> 0x0070 3332 332d 636f 6e66 2d69 643d 3030 3061 323-conf-id=000a
> 0x0080 3861 6132 2d33 3133 3430 3064 392d 3230 8aa2-313400d9-20
> 0x0090 6565 6235 6161 2d34 6536 3338 eeb5aa-4e638
0x1c + 0x7a == 0x96
But the packet (with IP and UDP) is 0x9c bytes long. The last
RADIUS attribute extends 6 bytes past the end of the packet, as given
the RADIUS header.
The UDP header (which I didn't quote) says that the packet is 0x9c
bytes long, so it's OK. It looks to me like there's a bug in their
client which adds an integer attribute (0x04, or 0x3d), but does NOT
add its length to the RADIUS packet length.
Alan DeKok.
-
List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
