RE : : Using DIAMETER_MISSING_AVP for optional AVP

"MORAND Lionel RD-CORE-ISS" <[email protected]>
Newsgroups gmane.ietf.aaa
Message-ID <7DBAFEC6A76F3E42817DF1EBE64CB026018B33F4@FTRDMEL2.rd.francetelecom.fr>
Hi Mikko,

> > Based on this definition, it is only applicable to AVP
> > "required by the command code definition". It is not clear if
> > we could reuse the same result code for an optional AVP.
> 
> In my opinion here "command code definition" refers to the 
> whole definition of the command, not just the ABNF. Some 
> parts of the definition are often presented in "free" text. 
> For example, Diameter Application command definition might 
> state that Pet-Poodle AVP must be present in the message in 
> case it is Friday the 13th. :)

[LM] Maybe but it is not clear in the RFC3588
 
> > My assumption was that a "SHOULD" was possible here because 
> you could 
> > have relyed on the command code ABNF descritpion to find 
> the missing 
> > AVP.
> 
> So if the command-code ABNF definition defines three required 
> AVPs and you receive DIAMETER_MISSING_AVP without Failed-AVP 
> how do you know which one was missing? You need to know the 
> message contents and the command-code definition (including 
> the rules for "conditional" AVPs) if you want to know what 
> AVP was missing. It doesn't matter at all whether the missing 
> AVP is required or optional.

[LM] Based on the RFC 3588:

6.1.2.  Sending a Request

   When sending a request, originated either locally, or as the result
   of a forwarding or routing operation, the following procedures MUST
   be followed:

   -  the Hop-by-Hop Identifier should be set to a locally unique value

   -  The message should be saved in the list of pending requests.

Receivig DIAMETER_MISSING_AVP without any Failed_AVP AVP, I'm able to
know what AVP is missing just by comparing the pending request with the
command code ABNF description. I don't need extra information if this
error message is only sent for missing "required" AVP.

By the way, this is not a discussion about 'conditional AVPs'. This
concept doesn't exist in the base protocol! We should only deal with
optional AVPs, whatever the use of these optional AVPs.

BR,

Lionel
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.