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

Link Summary Nack issue



Hi,

The current LMP draft (draft-ietf-ccamp-lmp-02.txt) provides an error_code
object to indicate what type of error is being reported in a LinkSummaryNack
message. The error being reported is generic and does not provide a correlation
between the included data link objects and their corresponding errors
respectively.

Problem: 
The LinkSummaryNack message points to a generic error, but does not point to
what is incorrect in a particular data link. This forces a reexamination of the
received data link objects and trying to figure out what is mismatched in each
of them.

Proposal:
Extend the Data Link object to contain an error_code entry. This can be similar
to the "error codes" defined in the draft and can point to multiple
inconsistencies in a particular data link.

Example: (from the draft, modified with the proposed change) 

14.13.  DATA_LINK Class 
    
   Class = 17.  
    
   o    IPv4, C-Type = 1 
    
    0                   1                   2                   3 
    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
   |     Flags     |                   (Reserved)                  | 
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
   |                   Local_Interface_Id (4 bytes)                | 
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
   |                   Remote_Interface_Id (4 bytes)               | 
   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 
   |                   ERROR CODE (4 bytes)                        | This field
is added
  
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+                              

The above ERROR CODE field is added to all C-Types for the DATA_LINK Class.

Advantages:
Efficiency. Eliminates unnecessary comparisons at the receiving end; Helps
isolate the incorrect parameter in the data link object.

Comments welcome.

Thanks in advance,
Srini Datari.