Good thread.  That was quick research into the Privacy draft Tim!

It sounds like we are all pretty much in agreement that:

*) generating private link-local addresses is a bad idea, and
neither the RFC or new Draft say to do it

*) generating private ULA's does make sense, just like private
global's makes sense (if desired by local administrators)

*) I see in the Draft where it says local administrators should
be able to disable privacy extensions by prefix, so privacy
addresses could be generated for, say, global but not ULA-scope
addresses, or as local administrators deem appropriate.

I like choices.  Thanks.

John Spence
----------------------------------------------------
John Spence, CCSI, CCNA, CISSP
Native6, Inc.
IPv6 Training and Consulting
[EMAIL PROTECTED]
(wk) 206-682-0275
www.native6.com
----------------------------------------------------
 

>-----Original Message-----
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
>Behalf Of [EMAIL PROTECTED]
>Sent: Wednesday, January 04, 2006 12:23 PM
>To: [EMAIL PROTECTED]
>Cc: [email protected]
>Subject: (no subject)
>
>Accidentally left original subject: out of original reply; 
>sorry about that. Comments in-line:
>
>-----Original Message-----
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On 
>Behalf Of Christian Huitema
>Sent: Wednesday, January 04, 2006 3:20 AM
>To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
>Cc: [email protected]
>Subject: RE: (no subject)
>
>Hosts are not supposed to make any distinction between ULA and 
>global scope addresses. 
>
>-> "not supposed to" seems a bit strong. Section 4.5 of RFC 
>4193 says "Application and other higher level protocols CAN 
>(capitalization mine) treat Local IPv6 addresses in the same 
>manner as other types of global unicast addresses." Again, in 
>section 1 "-In practice, applications MAY (capitalization 
>mine) treat these addresses like global scoped addresses." 
>Also, "In some cases, it is better for nodes and applications 
>to treat them differently from global unicast addresses.
>Hosts autoconfigure ULA addresses if the RA advertises and ULA
prefix. 
>
>-> 'if' being the operative word (they could also be assigned 
>via DHCPv6 or manually).
>
>Thus, hosts that are programmed to generate RFC 3041 addresses 
>for global scope addresses will do the same for ULA.
>
>-> I just read draft-ietf-ipv6-privacy-addrs-v2-04.txt***, and 
>see that it includes references to ULAs. It also refers to the 
>ULA spec as informative, which was at the time also a draft. 
>If the draft*** becomes an RFC (which I expect it will), thus 
>obsoleting RFC 3041, it is then it would be appropriate to say 
>hosts "will do the same for ULA". At present (RFC 3041, not 
>RFC 4193) it does not mention ULAs. It's only appropriate to 
>cite drafts as "works in progress".
>
>Best Regards,
>
>Tim Enos
>1Sam16:7
>
>> -----Original Message-----
>> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf
>Of
>> [EMAIL PROTECTED]
>> Sent: Tuesday, January 03, 2006 8:14 PM
>> To: [EMAIL PROTECTED]
>> Cc: [email protected]
>> Subject: (no subject)
>> 
>> Hi John, please see my comments in-line:
>> 
>> -----Original Message-----
>> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf
>Of
>> John
>> Spence
>> Sent: Monday, January 02, 2006 12:23 PM
>> To: [email protected]
>> Subject: Are privacy extensions, RFC 3041,defined for non 
>global-scope 
>> addresses?
>> 
>> 
>> I re-read the document, and it certainly focuses on the 
>privacy needs 
>> of global-scope addresses.  I did not find a place where it
said it 
>> was not defined for ULA or link-local scope addresses.
>> 
>> -> AFAICS, RFC 3041 deals only with global-scope addresses. 
>The stated
>> goals (2-4) explicitly refer to global-scope addresses.
>> 
>> Is that the intent - not defined for non global-scope
addresses?
>> Or I am reading that into it?
>> 
>> -> I think it's reasonable to conclude the mechanism defined
in RFC
>3041
>> is not defined for non global-scope addressses. ULAs to my
knowledge 
>> didn't exist at the time 3041 was written (RFC 3041 in January
2001,
>RFC
>> 4193 not until October 2005). Even though there is an extant
draft
>meant
>> to update 3041 [draft-ietf-ipv6-privacy-addrs-v2-04.txt], it
has yet
>to
>> become an RFC itself.
>> 
>> -> If by some stretch RFC 3041 was meant for link-local scope
>addresses,
>> it seems that would be suboptimal. At least as often as the 
>temp link- 
>> local unicast address changed, the node would have to 
>(un)subscribe to
>the
>> corresponding solicited-node multicast group(s). That could
lead to 
>> reduced performance. I'd also wonder about the affect
temporary
>link-local
>> addresses would have on a router's neighbor cache, and/or any
>connectivity
>> dependent upon the accuracy of cache entries... How might this
affect
>ND
>> itself (not a leading question BTW)?
>> 
>> Thanks.
>> 
>> -> Best regards,
>> 
>> Tim Enos
>> 1Sam16:7
>> 
>> ----------------------------------------------------
>> John Spence, CCSI, CCNA, CISSP
>> Native6, Inc.
>> IPv6 Training and Consulting
>> [EMAIL PROTECTED]
>> ----------------------------------------------------
>> 
>> 
>> 
>>
-----------------------------------------------------------------
---
>> IETF IPv6 working group mailing list
>> [email protected]
>> Administrative Requests:
https://www1.ietf.org/mailman/listinfo/ipv6
>>
-----------------------------------------------------------------
---
>> 
>> 
>>
-----------------------------------------------------------------
---
>> IETF IPv6 working group mailing list
>> [email protected]
>> Administrative Requests:
https://www1.ietf.org/mailman/listinfo/ipv6
>>
-----------------------------------------------------------------
---
>
>----------------------------------------------------------------
----
>IETF IPv6 working group mailing list
>[email protected]
>Administrative Requests:
https://www1.ietf.org/mailman/listinfo/ipv6
>----------------------------------------------------------------
----
>
>
>----------------------------------------------------------------
----
>IETF IPv6 working group mailing list
>[email protected]
>Administrative Requests:
https://www1.ietf.org/mailman/listinfo/ipv6
>----------------------------------------------------------------
----


--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to