RE: [Mip4] Additional 3344bis error codes
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Mip4] Additional 3344bis error codes



Hi Henrik. 

I agree in general that explicit error codes would be nice since we are
clumping them under "Reason unspecified (128)" or "Admin prohibited
(129)".  I'll gather some of our error codes that should be assigned.

Comments below.

> -----Original Message-----
> From: mip4-bounces at ietf.org [mailto:mip4-bounces at ietf.org] On 
> Behalf Of Henrik Levkowetz
> Sent: Thursday, November 17, 2005 12:54 PM
> To: Charles E. Perkins; Mobile IPv4 Mailing List; Pete McCann
> Subject: [Mip4] Additional 3344bis error codes
> 
> Hi,
> 
>     I've had some feedback from the people at ipUnplugged, 
> who have run into some new error cases in deployments of 
> MIPv4 where dynamic home address allocation has been used. 

If these are directly related to dynamic HoA assignment based on RFC
2794, wondering if we should deal with them in 2794bis instead of
3344bis?  Any error codes not related to dynamic HoA assignment should
definitely be in 3344bis.

 
> After some iterations and experience with code discovering 
> and reporting on the error cases, it seems the following 4 HA 
> error codes would be appropriate:
> 
> 
> 	* home address already in use (as care-of address)

Isn't this an ambiguity issue with RFC 2794 when RRQ has NAI and HoA
that is set?  HA can either assign a new HoA in the RRP or reject with
error code.  The draft,
draft-paulkandasamy-mobileip-nai-based-home-address-00.txt, covers this
case by sending ADDR_ALLOC_FAILED error code.


> 	* care-of address already in use (as home address)

The CoA can be HoA when deregistering.  What is the condition for
registration?  The CoA and HoA should not be in the same subnet prefix
since technically MN is home?


> 	* home address is not unicast

Same as #1 in terms that should be covered in 2794bis.  If the HoA is
assigned by HA, this should never happen.  If MN registers with an HoA
that is used or invalid, then possibly we could use the same error code
as #1.  That's why it's best to debate the various error codes in the
scope of that draft.

> 	* care-of address is not unicast
> 

This may be related to #2 bullet in terms of having the same error code.
I'm not sure how CoA can be checked for unicast or not without knowing
the subnet prefix?  This info is not known in the RRQ.

Kent



> Could we consider adding these to the "Registration denied by 
> the home agent" error codes in Section 3.4?
> 
> 
> ( Oh, and incidentally - 3344bis refers to RFC 1700, as 
> reference [21], but RFC 1700 has been replaced by IANA's 
> online database, according to RFC 3232, so reference [21] 
> should probably point to the IANA web-site and specifically 
> to http://www.iana.org/assignments/mobileip-numbers .)
> 
> 
> 	Henrik
> 

--
Mip4 mailing list: Mip4 at ietf.org
    Web interface: https://www1.ietf.org/mailman/listinfo/mip4
     Charter page: http://www.ietf.org/html.charters/mip4-charter.html
Supplemental site: http://www.mip4.org/




Note: Messages sent to this list are the opinions of the senders and do not imply endorsement by the IETF.