[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IPv6 Advanced Socket API extension for Mobile IP



Samita

Some comments on the draft

In the following section, the some of the members of the data
structures are not properly alligned within the structure.

2.1.3  Home Address Test Init (HoTI) Message
  - cookie is defined as array of 32 bit unsigned integers, but
    does not allign on 32 bit boundry

2.1.4  Care-of Address Test Init (CoTI) Message
  - Same as 2.1.3

2.1.5  Home Address Test (HOT) Message
  - cookie and keygen tocken are defined as arrays of 32 bit unsigned
    integers, but do not allign of 32 bit boundry

2.1.6  Care Of Address Test (COT) Message
   - same as 2.1.5

2.1.9  Binding Error Mobility Message
   - IPv6 address 'ip6mhbe_homeadr' does not allign on 64 bit boundry.

The recomendation is to change the type of all these structure members
to uint8_t thus reducing alignment requirement to 8 bits.


Section 2.2 assignes the value 62 to be the mobility header.  According
to IANA, valud 62 is assigned to CFTP.  According to same document,
the value 55 is currently assigned to IP Mobility.  I am not very familiar 
with Mobile IPv4, but I did not think it required a new protocol value.


In Section 3 you describe how to set the Home Address option through the
ancilary data.  However, you do not address what happens if the user
specifies multiple options, include HAO, in the same ansilary data
(same header).  I would suggest adding a restriction that to correctly
set the HAO in the destination header, the option SHOULD be specified alone
in a given ancilary data object.

In Section 4.1, the draft describes how to use the IPV6_CHECKSUM
option on mobility headers.  The draft seems to turn off "kernel"
checksumming by default on the IPPROTO_MH socket.  I belive
that the checksumming should be on by default (similar to ICMPv6)
with the ability to turn it off by the socket.  The reason is that
most Mobile IPv6 implementation are supported in the "kernel" and
already do checksumming.  The RAW socket interface can take advantage
of that.  So, at this poing NOT doing checksumming is an change
to the expected behavior.  To minimize that change, we can default to
do checksumming with the ability to turn it off and let the applcation
do it.


That's about it for now.
-vlad


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Vladislav Yasevich		Tru64 UNIX Network Group
Hewlett Packard 		Tel: (603) 884-1079
Nashua, NH 03062		ZKO3-3/T07


--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to majordomo@sunroof.eng.sun.com
--------------------------------------------------------------------