Hello List,

I am trying to test DHCPv6 Authentication with IA_NAs. When I add the
following packet...

packet=Ether(dst=dstMAC,src=srcMAC)
/IPv6(src=srcIP,dst=dstIP,hlim=1)
/UDP()
/DHCP6_Request(trid=transactionID)
/DHCP6OptClientId(optlen=14,duid=DUID_LLT(lladdr=clientID))
/DHCP6OptIA_NA(optlen=40,iaid=1,T1=4294967295,T2=4294967295,ianaopts=
 [DHCP6OptIAAddress(addr=requestedAddress,optlen=28)])
/DHCP6OptElapsedTime(optlen=2,elapsedtime=300)
/DHCP6OptOptReq(optlen=4,reqopts=[23,24])
/DHCP6OptServerId(optlen=14,duid=DUID_LLT(lladdr=serverID))



...scapy sends out a packet that has 4 extra bytes after the
DHCP6OptIAAddress option. I believe this is the iaid field in that option,
b/c when I change the iaid field these extra bytes change accordingly.

However, wireshark sees these bytes as an unknown DHCPv6 option (or other
option if the iaid was high enough), thus breaking how wireshark sees these
packets. Also, looking at the RFC 3315 (
http://www.networksorcery.com/enp/rfc/rfc3315.txt)
I cannot find anything
that would say that each individual IA_NA needs a separate iaid.

I can get around this by just using the raw packet header...but it's
irritating that the header doesn't create the correct packet.

Thanks