Re: Next steps on the IPv6 Node requirements draft

2003-06-27 Thread Derek Fawcus
On Wed, Jun 25, 2003 at 07:30:44AM -0400, Ralph Droms wrote: Regarding reverse DNS entries ... there is a specific problem with reverse DNS entries for autoconfiguration addresses regarding update of the reverse entries by the client. [ snip ] There are solutions - disallow reverse DNS

Re: Next steps on the IPv6 Node requirements draft

2003-06-25 Thread Ralph Droms
Regarding reverse DNS entries ... there is a specific problem with reverse DNS entries for autoconfiguration addresses regarding update of the reverse entries by the client. The portion of the DNS namespace into which the host wants to insert its reverse DNS entry is owned by the network to which

Re: Next steps on the IPv6 Node requirements draft

2003-06-25 Thread Ralph Droms
I agree with kre - address configuration through DHCP (confrolled by 'M' bit) and autoconfiguration through advertised prefixes should be considered independent. An interface may well have both autoconfiguration addresses and addresses obtained through DHCP (and manually configured addresses, as

Re: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread Robert Elz
Date:Wed, 18 Jun 2003 17:23:05 -0700 From:Alain Durand [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] | So what if the M bit is set _and_ a prefix is advertized? | Should the node give up its stateless autoconfigured address in favor of | DHCP? No, it should

RE: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread john.loughney
13:57 To: Alain Durand Cc: Loughney John (NRC/Helsinki); [EMAIL PROTECTED] Subject: Re: Next steps on the IPv6 Node requirements draft Date:Wed, 18 Jun 2003 17:23:05 -0700 From:Alain Durand [EMAIL PROTECTED] Message-ID: [EMAIL PROTECTED] | So what

Re: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread Alain Durand
On Thursday, June 19, 2003, at 03:56 AM, Robert Elz wrote: It is generally harmless to own an extra address though, having the statelessly configured one, as well as a dhcp supplied one should not cause any harm. Not sure. Two reasons: - There may be filters in place, for example that only

RE: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread matthew . ford
-Original Message- From: Alain Durand [mailto:[EMAIL PROTECTED] Sent: 19 June 2003 17:07 On Thursday, June 19, 2003, at 03:56 AM, Robert Elz wrote: It is generally harmless to own an extra address though, having the statelessly configured one, as well as a dhcp supplied one

Re: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread itojun
- There are reverse DNS issues. They may point to 2 different names or more likely, the stateless autoconfigured address won't resolve to a name, where the DHCP one will. As default address selection does not (yet?) say to prefer the DHCP one, logs and/or (very)

Re: Next steps on the IPv6 Node requirements draft

2003-06-19 Thread Alain Durand
Itojun, I guess you missed the more likely part of my sentence. I agree we don't have any standard wrt DNS registration of autoconfigured addresses, and the current deployment practices show that it is a pain to do it either manually or with dnsupdate (you have a key distribution problem), so

RE: Next steps on the IPv6 Node requirements draft

2003-06-18 Thread john.loughney
could reference such a spec. John -Original Message- From: ext Peter Bell [mailto:[EMAIL PROTECTED] Sent: 18 June, 2003 02:09 To: Loughney John (NRC/Helsinki) Cc: [EMAIL PROTECTED] Subject: Re: Next steps on the IPv6 Node requirements draft Along with removing DES from

Re: Next steps on the IPv6 Node requirements draft

2003-06-18 Thread Alain Durand
[EMAIL PROTECTED] wrote: For those IPv6 Nodes that implement DHCP, those nodes MUST use DHCP upon the receipt of a Router Advertisement with the 'M' flag set (see section 5.5.3 of RFC2462). So what if the M bit is set _and_ a prefix is advertized? Should the node give up its stateless

Re: Next steps on the IPv6 Node requirements draft

2003-06-17 Thread Peter Bell
Along with removing DES from the SHALL list, it looks likely that AES will be added to the IPSec SHALL requirements, perhaps this draft should include AES. Peter. [EMAIL PROTECTED] wrote: Hi all, I've updated the draft on 4 major points that were discussed at the IETF. Roughly they cover