The IESG has a number of comments on this document. Most of them are
of an editorial nature, and I'll let Rich summarize those in a
separate note.

One issue was substantive and requires WG feedback. The -06 document
specifies that public addresses are to be preferred over temporary by
default. My recollection is that the WG originally had temporary
addresses being preferred, but more recently switched this under
concerns that there were cases where the use of temporary addresses
might harm an application. Thus, the conservative thing to do would be
to prefer public addresses but allow the application (via an API) to
reverse this preference.

The IESG is concerned that if temporary addresses are not enabled by
default, they won't see widespread use in practice. This would
significantly decrease the overall impact that RFC 3041 will have on
reducing the threat it attempts to address.  What the IESG would
prefer seeing is that temporary addresses be given preference by
default, but also include more guidance text in the document to make
it clear that there are cases where the use of temporary addresses by
default may not be appropriate.

The following text has been proposed to address this issue:

Proposed new wording:

   Rule 7: Prefer public addresses.
   
   If SA is a temporary address and SB is a public address, then prefer 
   SA. Similarly, if SB is a temporary address and SA is a public 
   address, then prefer SB. 
   An implementation MUST support a per-connection configuration 
   mechanism (for example, an API extension) to reverse the sense of 
   this preference and prefer public addresses over public 
   addresses. 

   In order for temporary addresses to have their desired effect, they
   must become widely used in practice. The above rule is designed to
   ensure widespread usage and enablement of temporary addresses. It
   is believed that in many cases, temporary addresses can be used in
   place of public addresses without problems. However, there are some
   classes of applications that may fail due to the relatively short
   lifetime of temporary addresses (compared with public addresses) or
   due to the possibility of the reverse lookup of a temporary address
   either failing or returning a randomized name.  Implementations in
   which it is believed that such applications exist MAY reverse the
   sense of this rule and by default prefer public addresses over
   temporary addresses.

   It should also be noted that a simple "prefer" or "don't prefer"
   configuration switch with regards to the default selection of
   temporary addresses is not sufficiently flexible in many cases. In
   practice, it is expected that both client and server applications
   will run on the same devices simultaneously. Client applications
   will generally be able to use temporary addresses, while servers
   will generally need to use public addresses. One possible heuristic
   for distinguishing these cases is to assume that an application
   that invokes a passive open (as its first network usage) is a
   server, while an application that first invokes an active open be
   assumed to be a client.

(Note also, that the above text changes a "may" to MUST with regards
the need to provide a way for individual applications to change the
default via, e.g, a socket option)

Is the above a reasonable way forward? (Individual responses, even if
they are along the lines of "fine with me" would be appreciated, as
this document is needed soon to meet a 3GPP deadline).

Thomas
--------------------------------------------------------------------
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 [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to