- ipv6-wg - mailman.ripe.net

[ipv6-wg@ripe.net] Re: [address-policy-wg] Re: IANA to RIR IPv6 Allocation
by Wilfried Woeber, UniVie/ACOnet 25 Aug '04

25 Aug '04

>> My proposal would be that IANA would have to perform the subsequent >> allocation within this 30 day period, even if they might not be able to >> obtain advice or an expertise (for whatever reason) from any other party >> they might have decided to ask for advice. > >sure, if we can we do the same for rir to lir No problem with that, although I don't see the need for global coordination here. To me this sounds like a RIR-Customer service-contract aspect. Wilfried.

1 0

Re: [ipv6-wg@ripe.net] RE: [address-policy-wg] Re: IANA to RIR ipv6 Allocation
by Iljitsch van Beijnum 24 Aug '04

24 Aug '04

On 24-aug-04, at 4:48, Jeff Williams wrote: > Iljitsch and all, ^^^^^ No kidding... > None of this addresses the outstanding problems with Ipv6 or ipv6 > allocation concerns and existing as well as discussed problems. > They are "at least" the following: I don't think I get what you're talking about. > 1. ) Invalid or incorrect minimal allocation. What is invalid or incorrect here? > 2.) Still existing cost increases for allocations. What cost? > 3.) Still existing security/privacy "holes" in ipv6 What holes? There is no difference with IPv4. > 4.) Routing notification for new allocation practice or method. What on Earth is a "routing notification"? > 5.) Routing table maintenance Best practices and/or policy, Well, there is some of this but since the RIRs seem to have trouble living up to their own policies this is problematic. (See http://www.bgpexpert.com/archive2003q4.php ) > and enforcement of same. Yes, we don't do this in the internet. (And go fix the IPv4 table first if you feel so inclined.) > 6.) Dealing with "Dark" existing and/or future allocated addresses. Again, I have no idea what you're talking about. Now if we could have a mechanism to determine resolving DNS servers automatically in IPv6 and have visible IPv6 addresses for the root DNS servers we'd be well on our way. And of course we still need multihoming in IPv4 and the IETF has to quit rewriting the RFCs all the time. > Iljitsch van Beijnum wrote: I know what I wrote, so P L E A S E don't repeat it in its entirety!!! I'm really getting sick and tired of people who are too lazy to quote in a decent manner.

1 0

[ipv6-wg@ripe.net] DNS Weather Report 2004-08-24
by Daniel Roesen 24 Aug '04

24 Aug '04

DNS WEATHER REPORT for selected infrastructure zones ==================================================== Issue 2004-08-24 Zones analyzed and their SOA contacts: - . nstld(a)verisign-grs.com - arpa. nstld(a)verisign-grs.com - int. noc(a)icann.org - in-addr.arpa bind(a)arin.net - ip6.arpa. hostmaster(a)icann.org - ip6.int. hostmaster(a)ep.net Executive summary: not a single zone without any problems, but ip6.int situation improved significantly. The state of the root zone =========================== - m.root-servers.net is lagging behind the others... general SOA serial is 2004082201, m.root has 2004082101 The state of the ARPA zone ========================== - same SOA serial problem as in the root zone, m.root out-of-sync. Interestingly, all the same SOA serial numbers as the root zone (coincidence?) The state of the INT zone ========================= - in-zone NS RRset doesn't match the delegation NS RRset. in-zone RRset: delegation RRset: ns0.ja.net. ns0.ja.net. ns1.cs.ucl.ac.uk. ns1.cs.ucl.ac.uk. ns.icann.org. ns.icann.org. ns.isi.edu. ns.isi.edu. ns.uu.net. ns.uu.net. ns.ripe.net. z.ip6.int. - ns.isi.edu started to be auth again (was lame a few days ago), but is not in sync. current SOA serial of the INT TLD: 2004080300, ns.isi.edu has 2002080104 The state of the IN-ADDR.ARPA zone ================================== - same problem with m.root as with the root zone. m.root seems to lag exactly one day behind, judging from the SOA serial number. The state of the IP6.ARPA zone ============================== - in-zone NS RRset doesn't match the delegation NS RRset. in-zone RRset: delegation RRset: ns.apnic.net. ns.apnic.net. ns.icann.org. ns.icann.org. ns-sec.ripe.net. ns.ripe.net. tinnie.arin.net. tinnie.arin.net. Problem known, RIPE sent a reminder to ip6.arpa operators a week ago, obviously without any outcome. The state of the IP6.INT zone ============================= - ns.isi.edu (one of the INT TLD servers) feels authoritative for the IP6.INT zone, but is neither listed in the delegation NS RRset, nor in the in-zone NS RRset of IP6.INT. Luckily, ns.isi.edu carries ip6.int with the same SOA serial as the official servers, so introduces no operational problems so far. Regards, Daniel

1 0

[ipv6-wg@ripe.net] Re: IANA to RIR IPv6 Allocation
by Wilfried Woeber, UniVie/ACOnet 21 Aug '04

21 Aug '04

I've got one comment here on section 7. paragraph 4. "Subsequent Allocation": While the wording might be sufficiently clear for a native speaker by referring to "set of administrative procedures....", I am missing clear guidance for the IANA that there should _not_ be any incentive or responsibilty to consult with other groups, entities or third parties (irrespective of their, say, credibility for e.g. protocol development or technical specifications) which would require extending that 30 day period. My proposal would be that IANA would have to perform the subsequent allocation within this 30 day period, even if they might not be able to obtain advice or an expertise (for whatever reason) from any other party they might have decided to ask for advice. Wilfried ( https://cert.aco.net/ ) _________________________________:_____________________________________ Wilfried Woeber : e-mail: Woeber(a)CC.UniVie.ac.at UniVie Computer Center - ACOnet : Tel: +43 1 4277 - 140 33 Universitaetsstrasse 7 : Fax: +43 1 4277 - 9 140 A-1010 Vienna, Austria, Europe : RIPE-DB: WW144, PGP keyID 0xF0ACB369 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~:~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ...there's no place like 127.0.0.1 (or ::1/128 ?) From: David Kessens <david.kessens(a)nokia.com> To: Jeroen Massar <jeroen(a)unfix.org> Date: Fri, 20 Aug 2004 01:48:52 -0700 CC: ipv6-wg(a)ripe.net Subject: Re: [ipv6-wg(a)ripe.net] FW: [arin-council] IANA to RIR IPv6 Allocation Jeroen, On Fri, Aug 20, 2004 at 09:25:59AM +0200, Jeroen Massar wrote: > For people not on the arin-council (not likely ;) or the ppml@arin > lists: All policy discussions for the RIPE region are now conducted at the RIPE policy working group maillist: address-policy-wg(a)ripe.net David Kessens --- > -----Forwarded Message----- > From: "Sweeting, John" <John.Sweeting(a)teleglobe.com> > To: 'ppml(a)arin.net' <ppml(a)arin.net> > Subject: [ppml] FW: [arin-council] IANA to RIR IPv6 Allocation > Date: Thu, 19 Aug 2004 17:36:56 -0400 > > I am submitting the following proposal IAW ARIN's Internet Resource Policy > Evaluation Process. In the interest of all I would like to disclose up front > that the information used to prepare this template was provided by ARIN > staff. > > > ############################################ > Template: ARIN-POLICY-PROPOSAL-TEMPLATE-1.0 > > 1. Policy Proposal Name: Allocation of IPv6 Address Space by the Internet > Assigned Numbers Authority (IANA) Policy to Regional Internet Registries > > 2. Author > > a. name: John Sweeting > b. email: john.sweeting(a)teleglobe.com > c. telephone: 703-766-3042 > d. organization: ARIN Advisory Council > > 3. Proposal Version: 1 > > 4. Submission Date: 8/17/04 > > 5. Proposal type: new, global > > 6. Policy term: permanent > > 7. Policy statement: > > 1. Minimum Allocation Size. > > The minimum size of any allocation of IPv6 address space from IANA to an RIR > is prefix length /12. If this address space is not sufficient to meet the > needs of an RIR for a projected 18 month period, IANA shall allocate to that > RIR the address space for which it can provide justification. > > 2. Reservation of Unicast Address space. > > 2.1 IANA. By RFC 3513 the IANA has been allocated the range of IPv6 > addresses beginning at binary value 001 (prefix length /3) for its > allocations of unicast address space. In order to support regional > aggregation of IPv6 address space IANA shall establish a reservation of a > prefix length of /6 from this space for each established RIR and for each > emerging RIR. Allocations to each RIR will be from the appropriate > reservation. > > 2.2. RIRs. Each RIR may apply its own respective chosen allocation and > reservation strategy in order to meet the needs of its community and to > ensure the efficiency and efficacy of its work. Such reservations made by > an RIR will be considered as being allocated by that RIR when that RIR > applies for an allocation of address space from the IANA. > > 3. Initial Allocation. > > 3.1. Upon implementation of this policy IANA shall allocate to each > established RIR a /12 from the reservation established for each particular > RIR. > > 3.2. Upon recognition of an RIR by ICANN that RIR shall receive a /12 from > the reservation set aside for that RIR. > > > 4. Subsequent Allocation. > > An RIR shall be eligible for an allocation of at least a minimal allocation > from the IANA when its current holdings are less than 50% of its 18 month > requirement or when it has less than 180 days of holdings available. The > IANA shall evaluate the requested allocation using a set of administrative > procedures that are mutually agreed to by the IANA and the NRO. This set of > procedures shall be enacted within 30 days of the implentation of this > policy. > > > 5. Announcement of IANA allocations to the RIRs > > When address space is allocated to a RIR, the IANA will send a detailed > announcement to the receiving RIR. The IANA will also make announcements to > all other RIRs, informing them of the recent allocation. > > The IANA will make appropriate modifications to the "Internet Protocol V6 > Address Space" page of the IANA website and may make announcements to only > its own global announcement lists. The IANA announcements will be limited > to which address ranges, the time of allocation and to which Registry they > have been allocated. > > > 8. Rationale: > > The current IANA allocation policy for IPv6 is an interim policy that was > promulgated in 1999. Operational experience has demonstrated the current > minimum allocation size is too small; that the built in reservation system > that must be followed by the RIRs does not allow for efficient and effective > management of the resource by the RIR; and does not provide for an well > known evaluation criteria. This document does not stipulate performance > requirements in the provision of services by IANA to an RIR in accordance > with the policy. Such requirements should be specified by appropriate > agreements between the NRO and ICANN. > > > 9. Timetable for implementation: Thirty days after ratification by the ICANN > Board of Directors in accordance with the global policy development process _______________________________________________

5 6

[ipv6-wg@ripe.net] FW: [arin-council] IANA to RIR IPv6 Allocation
by Jeroen Massar 20 Aug '04

20 Aug '04

For people not on the arin-council (not likely ;) or the ppml@arin lists: -----Forwarded Message----- From: "Sweeting, John" <John.Sweeting(a)teleglobe.com> To: 'ppml(a)arin.net' <ppml(a)arin.net> Subject: [ppml] FW: [arin-council] IANA to RIR IPv6 Allocation Date: Thu, 19 Aug 2004 17:36:56 -0400 I am submitting the following proposal IAW ARIN's Internet Resource Policy Evaluation Process. In the interest of all I would like to disclose up front that the information used to prepare this template was provided by ARIN staff. ############################################ Template: ARIN-POLICY-PROPOSAL-TEMPLATE-1.0 1. Policy Proposal Name: Allocation of IPv6 Address Space by the Internet Assigned Numbers Authority (IANA) Policy to Regional Internet Registries 2. Author a. name: John Sweeting b. email: john.sweeting(a)teleglobe.com c. telephone: 703-766-3042 d. organization: ARIN Advisory Council 3. Proposal Version: 1 4. Submission Date: 8/17/04 5. Proposal type: new, global 6. Policy term: permanent 7. Policy statement: 1. Minimum Allocation Size. The minimum size of any allocation of IPv6 address space from IANA to an RIR is prefix length /12. If this address space is not sufficient to meet the needs of an RIR for a projected 18 month period, IANA shall allocate to that RIR the address space for which it can provide justification. 2. Reservation of Unicast Address space. 2.1 IANA. By RFC 3513 the IANA has been allocated the range of IPv6 addresses beginning at binary value 001 (prefix length /3) for its allocations of unicast address space. In order to support regional aggregation of IPv6 address space IANA shall establish a reservation of a prefix length of /6 from this space for each established RIR and for each emerging RIR. Allocations to each RIR will be from the appropriate reservation. 2.2. RIRs. Each RIR may apply its own respective chosen allocation and reservation strategy in order to meet the needs of its community and to ensure the efficiency and efficacy of its work. Such reservations made by an RIR will be considered as being allocated by that RIR when that RIR applies for an allocation of address space from the IANA. 3. Initial Allocation. 3.1. Upon implementation of this policy IANA shall allocate to each established RIR a /12 from the reservation established for each particular RIR. 3.2. Upon recognition of an RIR by ICANN that RIR shall receive a /12 from the reservation set aside for that RIR. 4. Subsequent Allocation. An RIR shall be eligible for an allocation of at least a minimal allocation from the IANA when its current holdings are less than 50% of its 18 month requirement or when it has less than 180 days of holdings available. The IANA shall evaluate the requested allocation using a set of administrative procedures that are mutually agreed to by the IANA and the NRO. This set of procedures shall be enacted within 30 days of the implentation of this policy. 5. Announcement of IANA allocations to the RIRs When address space is allocated to a RIR, the IANA will send a detailed announcement to the receiving RIR. The IANA will also make announcements to all other RIRs, informing them of the recent allocation. The IANA will make appropriate modifications to the "Internet Protocol V6 Address Space" page of the IANA website and may make announcements to only its own global announcement lists. The IANA announcements will be limited to which address ranges, the time of allocation and to which Registry they have been allocated. 8. Rationale: The current IANA allocation policy for IPv6 is an interim policy that was promulgated in 1999. Operational experience has demonstrated the current minimum allocation size is too small; that the built in reservation system that must be followed by the RIRs does not allow for efficient and effective management of the resource by the RIR; and does not provide for an well known evaluation criteria. This document does not stipulate performance requirements in the provision of services by IANA to an RIR in accordance with the policy. Such requirements should be specified by appropriate agreements between the NRO and ICANN. 9. Timetable for implementation: Thirty days after ratification by the ICANN Board of Directors in accordance with the global policy development process. 10. Meeting presenter: Elected AC Member or the President of ARIN. END OF TEMPLATE ################################################

4 5

[ipv6-wg@ripe.net] Re: 9/9/2004 IP6.INT Removal (Was: 9/9/2006 : ip6.int shutdown?)
by Daniel Roesen 20 Aug '04

20 Aug '04

On Wed, Aug 04, 2004 at 03:59:35AM +0000, bmanning(a)vacation.karoshi.com wrote: > > So the whole delegation thing is broken. The int. TLD servers delegate > > ip6.int to imag.imag.fr which feels it self authoritative, but isn't > > in the NS RRset of the ip6.int zone. And ip6.int is delegated to > > munnari.oz.au which doesn't know about this (anymore). > > so... this was apparently enough to get the attention it > deserved. the long postponed migration from imag.imag.fr. > to ns3.nic.fr. is now "done" and imag.imag.fr. has been removed > from the NS list in the INT zone. > and munnari.oz.au has been removed. - more cleanup in the > next week or two. Well, two weeks down the road and things got even worse: NS list summary for ip6.int. from parent (int.) servers == flag.ep.net. ns3.nic.fr. y.ip6.int. == z.ip6.int. $ dig @flag.ep.net. int. SOA +norec | fgrep status: ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 9740 $ dig @ns3.nic.fr int. SOA +norec | fgrep status: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 334 $ dig @y.ip6.int int. SOA +norec | fgrep status: ;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 17872 $ dig @z.ip6.int int. SOA +norec ;; connection timed out; no servers could be reached So we have 3 out of 4 servers totally broken. Looks like the importance of ip6.int is way overstated by some folks. Regards, Daniel

6 6

[ipv6-wg@ripe.net] RE: [sig-ipv6] Re: 9/9/2004 IP6.INT Removal (Was: 9/9/2006 :ip6.int shutdown?)
by Bound, Jim 20 Aug '04

20 Aug '04

> Can you identify those vendors which didn't react to a RFC > from 3 years ago? No that is up to those vendors product managers etc. What I know is from interoperability testing and cannot reveal. You missed mine and Bill's entire point. No one is against doing this it is a matter getting it into the production DNS code base. I think Bill should answer your other questions as you integrated my mail with Bill's. Your complaining here in IETF at least is irrelevant. And not the purpose of the IETF list IMO. I will not respond to other mail. If you want to discuss offline send me mail and I suggest you copy Bill. Ipv6.arpa is supported in the latest BIND releases and IMO all should be doing it, but I don't attribute it to being lazy at all. Again IPv6 is production code now vendors can't just change it as we do the public domain and university code base, it requires QA et al per the customer and that has cost and all changes to IPv6 at this point. /jim

2 1

[ipv6-wg@ripe.net] New IPv6 Address Block Allocated to RIPE NCC
by leo vegoda 18 Aug '04

18 Aug '04

Dear Colleagues, The RIPE NCC received the IPv6 address range 2001:4600::/23 from the IANA in August 2004. You may want to adjust any filters you have in place accordingly. More information on the IP space administered by the RIPE NCC can be found on our web site at: <https://www.ripe.net/ripe/docs/smallest-alloc-sizes.html> Regards, -- leo vegoda Registration Services Manager RIPE NCC

1 0

[ipv6-wg@ripe.net] 2.0.0.2.ip6.arpa broken
by Daniel Roesen 17 Aug '04

17 Aug '04

On Sat, Aug 14, 2004 at 09:17:56AM +0100, David Malone wrote: > 2.0.0.2.ip6.int has become completely unavailable in the last few > days, resulting in delays for people connecting from 6to4 addresses > via ssh and the like. 2.0.0.2.ip6.arpa is also broken in all colors. soa @ns.apnic.net. for ip6.arpa. has serial: 2004071900 soa @ns.icann.org. for ip6.arpa. has serial: 2004071900 soa @ns-sec.ripe.net. for ip6.arpa. has serial: 2004071900 dig @tinnie.arin.net. for SOA of parent (ip6.arpa.) failed ==> tinnie.arin.net IPv6 connectivity is broken: $ dig @69.25.34.195 ip6.arpa. SOA +short dns1.icann.org. hostmaster.icann.org. 2004071900 3600 1800 604800 10800 $ dig @2001:440:2000:1::22 ip6.arpa. SOA ; <<>> DiG 9.2.3 <<>> @2001:440:2000:1::22 ip6.arpa. SOA ;; global options: printcmd ;; connection timed out; no servers could be reached Found 0 NS and 0 glue records for 2.0.0.2.ip6.arpa. @ns.apnic.net. (AUTH) Found 0 NS and 0 glue records for 2.0.0.2.ip6.arpa. @ns.icann.org. (AUTH) Found 4 NS and 4 glue records for 2.0.0.2.ip6.arpa. @ns-sec.ripe.net. (AUTH) DNServers for ip6.arpa. === 3 were also authoritatve for 2.0.0.2.ip6.arpa. === 0 were non-authoritative for 2.0.0.2.ip6.arpa. ERROR: Found 2 diff sets of NS records === from servers authoritative for 2.0.0.2.ip6.arpa. WARNING: ns.apnic.net. claims to be authoritative for 2.0.0.2.ip6.arpa. == but no NS record at parent zone WARNING: ns.icann.org. claims to be authoritative for 2.0.0.2.ip6.arpa. == but no NS record at parent zone WARNING: ns-sec.ripe.net. claims to be authoritative for 2.0.0.2.ip6.arpa. == but no NS record at parent zone ==> so although the three remaining NS for ip6.arpa have the same SOA serial, they obviously have a different view on the NS RRset for 2.0.0.2.ip6.arpa... and all are authoritative for 2.0.0.2.ip6.arpa. Now looking at the NS RRset ns-sec.ripe.net returns for 2.0.0.2.ip6.arpa: == ns-apnic.6to4.nro.net. ns-arin.6to4.nro.net. ns-lacnic.6to4.nro.net. == ns-ripe.6to4.nro.net. soa @ns-apnic.6to4.nro.net. for 2.0.0.2.ip6.arpa. serial: 2004072901 dig @ns-arin.6to4.nro.net. for SOA of 2.0.0.2.ip6.arpa. failed soa @ns-lacnic.6to4.nro.net. for 2.0.0.2.ip6.arpa. serial: 2004072901 soa @ns-ripe.6to4.nro.net. for 2.0.0.2.ip6.arpa. serial: 2004072901 ==> ns-arin.6to4.nro.net == tinnie.arin.net, so we have the same IPv6 reachability problem again What a mess all over... why do ns.apnic.net and ns.icann.org return NXDOMAIN when queried for the NS RRset for 2.0.0.2.ip6.arpa, but ns-sec.ripe.net returning one? It looks like the whole 2.0.0.2.ip6.arpa is missing in the ip6.arpa zone, and ns-sec.ripe.net is only by chance carrying the 2.0.0.2.ip6.arpa zone, thus returning the NS RRset. I wonder where's the right place to report such global (multi-RIR/org) DNS problems to, if not v6ops. I see no "global IPv6 operations" list, can anyone enlighten me? Best regards, Daniel

2 4

[ipv6-wg@ripe.net] Re: [address-policy-wg] Policy for allocation of IPv6 address space from IANA to RIRs
by Daniel Roesen 13 Aug '04

13 Aug '04

On Wed, Aug 11, 2004 at 09:04:21AM +0100, Jon Lawrence wrote: > Yes, doubling does seem unwise. > It would make sense (to me anyway) that once xx% of a /12 is allocated then > another /12 is issued to the RIR. Indeed. I see no point in unconditional doubling. I'm pretty confident that /12 blocks are large enough to serve a RIR long enough so that ordering a new /12 is not hampering anything. My suggestion would be to set aside a /8 per RIR (perhaps also a DNS reverse delegation for that) and allocate /12s to the RIRs upon their request. A RIR qualifies for a new /12 block as soon as nn% usage of the current /12 block is reached. nn might be 50% or more. As Randy suggests, the percentage should be low enough so that the RIRs can get new space without delaying allocation to LIRs (as it happens nowadays). Best regards, Daniel

3 2