RE: Issue: Attribute concatenation/splitting

["Nelson, David" <dnelson@enterasys.com>]

Pasi Eronen writes...

> Yes, this would work as well: an attribute length 255 would mean that
> the data item continues to the next attribute; attribute length less
> than 255 would indicate the final piece of this data item.

Making the RADIUS application layer fragmentation and reassembly
agnostic to specific attribute ID indeed has some attractive properties.

> The only catch might be that RFC 2865 prohibits string/text attributes
> with length 2 (=string length 0). But on the other hand, at least
> draft-aboba-radext-fixes talks about User-Name attributes with string
> length 0, so perhaps this could be used anyway...?

We are documenting this sort of "implementation discrepancies", but IMO,
we ought not to encourage them.  If RFC 2865 says MUST or MUST NOT, I
think we need to honor those requirements.

I'll offer another idea, although I'm not sure it is one that I would
prefer.  We can take advantage for the tagged attribute format used for
the tunnel attributes [RFC 2868].  All RADIUS attributes containing
fragments of a single Diameter rule would have the same tag value.


--
to unsubscribe send a message to radiusext-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/radiusext/>