Keyboard Shortcuts
Thread View
j: Next unread messagek: Previous unread messagej a: Jump to all threadsj l: Jump to MailingList overview
Dear SIG members
The proposal "Proposal to modify 'end site' definition and allow end sites to receive IPv6 allocations" has been sent to the Policy SIG for review. It will be presented at the Policy SIG at APNIC 23 in Bali, Indonesia, 26 February - 2 March 2007. You are invited to review and comment on the proposal on the mailing list before the meeting.
The proposal's history can be found at:
http://www.apnic.net/policy/proposals/prop-045-v001.html
PS. The deadline for proposals for policy SIG is on 29th of this month. Please submit your proposal (Policy / Informational) before the 29th.
Regards,
Kenny Huang Policy SIG huangk@alum.sinica.edu
________________________________________________________________________
prop-045-v001: Proposal to modify "end site" definition and allow end sites to receive IPv6 allocations ________________________________________________________________________
Author: Jordi Palet Martinez, Consulintel
Version: 1
Date: 22 January 2007
SIG: Policy
Introduction ------------ This policy modification is intended to provide a solution for portable assignments required by entities which are not multihomed.
Summary of the current problem ------------------------------ Often, some organisations need to make internal assignments. Their networks may be made up of a number of sites that each has their own L2 infrastructure. In some cases, organisations may have a small number of sites, but still need their own block so that they can avoid future renumbering, if they change their upstream provider, or have other administrative or policy reasons for that.
One example might be a large university that has several campuses and faculties, each requiring IPv6 addresses. It may have one or several upstream providers. The university will most likely need to be able to assign IPv6 addresses from the same block to its sites and, at the same time, be able to use one or several upstreams. The university network behaves like an internal university ISP to each of the end sites. Existing policy 035 only solve the problem for this university if they are multihomed.
Situation in other RIRs ----------------------- A similar proposal has also been submitted to RIPE NCC, LACNIC and AfriNIC regions and updated versions are under preparation.
Details ------- The following policy changes are proposed for APNIC-089, "IPv6 Address Allocation and Assignment Policy":
1. Definition of "end site"
The definition of "end site" in section 2.9 should be broadened to include a wider range of end users. This definition should be expanded to include end users that have a legal relationship with the service provider. Such legal relationships would allow end users that are part of the service provider or legally associated with the provider to be considered "end sites". For example, the different faculties or campuses of a university could be considered to be "end sites" under this proposed new definition.
2. Initial allocation criteria
The following changes are proposed in section 5.1.1 of APNIC-089:
a) Allow end sites to apply for an allocation
b) Expand the criteria of the types of sites an organisation can provide IPv6 connectivity to include:
- sites within its own organisations - sites at related organisations
Pros/Cons ---------
Advantages:
The difficulty encountered in receiving IPv6 address space by some big entities that have a need to use IPv6 is a clear barrier for its deployment.
Disadvantages:
One possible effect of this proposal would be a growth of global routing tables. This is only to be expected when new allocations are made possible under this proposal.
Effect on APNIC --------------- There may be a small increase in the number of IPv6 allocation requests from organizations which are not multihomed but require a portable block.
Effect on NIRs -------------- NIRs may need to adapt their own similar policy. Otherwise, the organizations requiring a portable block may become APNIC LIRs to access to it.
Jordi,
I am not sure I understand the intention behind your proposal properly. You seem to want to broaden the definition of "end site" to include multi-site networks and allow allow this kind of end site to qualify for an allocation when they will provide connectivity to 'sites' within their own organisation.
Allowing organisations that only make internal assignments to qualify for a /32 allocation would make dramatically increase the number of organisations who qualify. I think almost any organisation with multiple departments and internal contracts could qualify for a /32 allocation. Is that your intention or have I misunderstood you?
Regards,
Hi Leo,
I don't think this will create an explosion in the demand for prefixes for any kind of organizations. I expect that only those organizations that have a real need and can pay for the membership/space will go for it.
Moreover, the existing policies only allow that for organizations which are multihomed. There is an example already in the policy proposal for a big university. Especially in developing regions, it is easy to find those entities not being multihomed, because they can't simply pay for that and at the same time, they may need to change the upstream provider every time there is a better deal. So the need is real, in my opinion.
Under those considerations, I think is fair to allow those organizations to obtain the addressing space.
So trying to answer your question, no, the intend is not to allow *any* organization, just everyone that has the need.
In think in a previous meeting the staff was asked about how many organizations tried to get a prefix in cases similar to this, and the number was very low (not sure if it was just a couple of them). May be with this policy proposal, if it becomes accepted by the community, the number can be 10 times the number of rejections. Still a low number, and clearly not creating a dramatic increase.
Regards, Jordi
De: Leo Vegoda leo.vegoda@icann.org Responder a: sig-policy-bounces@lists.apnic.net Fecha: Tue, 23 Jan 2007 15:24:03 +0100 Para: sig-policy@apnic.net Asunto: Re: [sig-policy] prop-045: Proposal to modify "end site"definition and allow end sites to receive IPv6 allocations
Jordi,
I am not sure I understand the intention behind your proposal properly. You seem to want to broaden the definition of "end site" to include multi-site networks and allow allow this kind of end site to qualify for an allocation when they will provide connectivity to 'sites' within their own organisation.
Allowing organisations that only make internal assignments to qualify for a /32 allocation would make dramatically increase the number of organisations who qualify. I think almost any organisation with multiple departments and internal contracts could qualify for a /32 allocation. Is that your intention or have I misunderstood you?
Regards,
-- Leo Vegoda IANA Numbers Liaison
sig-policy: APNIC SIG on resource management policy
sig-policy mailing list sig-policy@lists.apnic.net http://mailman.apnic.net/mailman/listinfo/sig-policy
********************************************** The IPv6 Portal: http://www.ipv6tf.org
Bye 6Bone. Hi, IPv6 ! http://www.ipv6day.org
This electronic message contains information which may be privileged or confidential. The information is intended to be for the use of the individual(s) named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, including attached files, is prohibited.
Hi Jordi,
I agree that making address space available to networks that need it is good. My concern isn't the goal but the phrasing of the policy. I think the proposal is phrased in a way that would allow almost any organisation where one department 'bills' another for a service to qualify for a /32 allocation. You implied that the membership fees are the main deterrent. As the APNIC Fees WG is reviewing the fee structure I think it might not be a good idea to rely on it when designing policy.
I think your goal is to ensure that multi-site enterprise networks can receive portable address space from APNIC. If that is the case, maybe it would be useful to ensure that 'site' is less vaguely defined and change the policy to allow multi-site networks under common ownership and operation to qualify for an allocation?
Regards,
Hi Leo,
See below, in-line.
Regards, Jordi
De: Leo Vegoda leo.vegoda@icann.org Responder a: leo.vegoda@icann.org Fecha: Wed, 24 Jan 2007 17:18:44 +0100 Para: jordi.palet@consulintel.es CC: "sig-policy@lists.apnic.net" sig-policy@lists.apnic.net Asunto: Re: [sig-policy] prop-045: Proposal to modify "end site"definition and allow end sites to receive IPv6 allocations
Hi Jordi,
I agree that making address space available to networks that need it is good. My concern isn't the goal but the phrasing of the policy. I think the proposal is phrased in a way that would allow almost any organisation where one department 'bills' another for a service to qualify for a /32 allocation. You implied that the membership fees are the main deterrent. As the APNIC Fees WG is reviewing the fee structure I think it might not be a good idea to rely on it when designing policy.
I think it should be on the other way around. I guess the fees can't be changed each time a new policy is passed, but the fees should make sure that "in general", fees are charged in a fair way in order to cover APNIC costs+budget provision for continued operation for a certain number of years, etc.
So policies should be "fee agnostic", but somehow fees should be coherent.
What I meant is that even if the cost is low, not all the companies will go for this addressing space, but just in case, fee structure should consider this. I guess the questions are should an ISP which is making business pay more for addresses than an enterprise which is using those for its own infrastructure ? Or the other way around ? Should this space considered a luxury and thus need to pay more ? Or is a need and should pay less ? Or should the cost reflect somehow the extra "slot" in the routing tables which in this case is for a single company (while the one for an ISP is for many) and consequently be more expensive ? Should the cost be used to discourage using extra routing slots/addressing blocks when it is a need ?.
And of course, we can keep moving on that direction, but it will be a totally different topic and probably scope of the fees WG ...
I think your goal is to ensure that multi-site enterprise networks can receive portable address space from APNIC. If that is the case, maybe it would be useful to ensure that 'site' is less vaguely defined and change the policy to allow multi-site networks under common ownership and operation to qualify for an allocation?
It depends on what we consider is a multi-site network. A university may be split in many buildings in a single campus, or many buildings in several campuses. I think the policy should cover both cases.
I'm not sure it is so easy to describe so many possibilities ... May be it is easier to describe what type of network should NOT receive this allocation ?
Not really sure, just voicing openly my thoughts.
Regards,
-- Leo Vegoda IANA Numbers Liaison
********************************************** The IPv6 Portal: http://www.ipv6tf.org
Bye 6Bone. Hi, IPv6 ! http://www.ipv6day.org
This electronic message contains information which may be privileged or confidential. The information is intended to be for the use of the individual(s) named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, including attached files, is prohibited.
Hi Jordi,
On Jan 24, 2007, at 6:34 PM, JORDI PALET MARTINEZ wrote:
[...]
I agree that making address space available to networks that need it is good. My concern isn't the goal but the phrasing of the policy. I think the proposal is phrased in a way that would allow almost any organisation where one department 'bills' another for a service to qualify for a /32 allocation. You implied that the membership fees are the main deterrent. As the APNIC Fees WG is reviewing the fee structure I think it might not be a good idea to rely on it when designing policy.
I think it should be on the other way around. I guess the fees can't be changed each time a new policy is passed, but the fees should make sure that "in general", fees are charged in a fair way in order to cover APNIC costs+budget provision for continued operation for a certain number of years, etc.
So policies should be "fee agnostic", but somehow fees should be coherent.
My point isn't that the fee schedule will change radically but that if the policy uses a particular aspect of the fee schedule it may find that things work quite differently if the fee schedule does change. For that reason I would suggest aiming for a policy that does not rely on fees to stimulate or limit demand.
What I meant is that even if the cost is low, not all the companies will go for this addressing space, but just in case, fee structure should consider this. I guess the questions are should an ISP which is making business pay more for addresses than an enterprise which is using those for its own infrastructure ? Or the other way around ? Should this space considered a luxury and thus need to pay more ? Or is a need and should pay less ? Or should the cost reflect somehow the extra "slot" in the routing tables which in this case is for a single company (while the one for an ISP is for many) and consequently be more expensive ? Should the cost be used to discourage using extra routing slots/addressing blocks when it is a need ?.
And of course, we can keep moving on that direction, but it will be a totally different topic and probably scope of the fees WG ...
I think discussion of fees should take place in the Fees WG. I don't want to argue for any particular model for the fees, though.
I think your goal is to ensure that multi-site enterprise networks can receive portable address space from APNIC. If that is the case, maybe it would be useful to ensure that 'site' is less vaguely defined and change the policy to allow multi-site networks under common ownership and operation to qualify for an allocation?
It depends on what we consider is a multi-site network. A university may be split in many buildings in a single campus, or many buildings in several campuses. I think the policy should cover both cases.
I think you are conflating 'enterprise' and 'site'. I agree that a single enterprise might comprise multiple sites. It might be appropriate for (some?) multi-site enterprises to receive a /32 allocation. My concern is that the current definition of 'site' is vague and this makes it difficult to amend the policy in the way you want. If 'site' is clearly defined it will make it much easier to write policy text that is really clear on who qualifies for a /32 allocation.
Regards,
So then you're suggesting that we need also a complete review of the "site" definition ? Any specific suggestions ?
Regards, Jordi
De: Leo Vegoda leo.vegoda@icann.org Responder a: leo.vegoda@icann.org Fecha: Wed, 24 Jan 2007 19:36:58 +0100 Para: jordi.palet@consulintel.es CC: "sig-policy@lists.apnic.net" sig-policy@lists.apnic.net Asunto: Re: [sig-policy] prop-045: Proposal to modify "end site"definition and allow end sites to receive IPv6 allocations
Hi Jordi,
On Jan 24, 2007, at 6:34 PM, JORDI PALET MARTINEZ wrote:
[...]
I agree that making address space available to networks that need it is good. My concern isn't the goal but the phrasing of the policy. I think the proposal is phrased in a way that would allow almost any organisation where one department 'bills' another for a service to qualify for a /32 allocation. You implied that the membership fees are the main deterrent. As the APNIC Fees WG is reviewing the fee structure I think it might not be a good idea to rely on it when designing policy.
I think it should be on the other way around. I guess the fees can't be changed each time a new policy is passed, but the fees should make sure that "in general", fees are charged in a fair way in order to cover APNIC costs+budget provision for continued operation for a certain number of years, etc.
So policies should be "fee agnostic", but somehow fees should be coherent.
My point isn't that the fee schedule will change radically but that if the policy uses a particular aspect of the fee schedule it may find that things work quite differently if the fee schedule does change. For that reason I would suggest aiming for a policy that does not rely on fees to stimulate or limit demand.
What I meant is that even if the cost is low, not all the companies will go for this addressing space, but just in case, fee structure should consider this. I guess the questions are should an ISP which is making business pay more for addresses than an enterprise which is using those for its own infrastructure ? Or the other way around ? Should this space considered a luxury and thus need to pay more ? Or is a need and should pay less ? Or should the cost reflect somehow the extra "slot" in the routing tables which in this case is for a single company (while the one for an ISP is for many) and consequently be more expensive ? Should the cost be used to discourage using extra routing slots/addressing blocks when it is a need ?.
And of course, we can keep moving on that direction, but it will be a totally different topic and probably scope of the fees WG ...
I think discussion of fees should take place in the Fees WG. I don't want to argue for any particular model for the fees, though.
I think your goal is to ensure that multi-site enterprise networks can receive portable address space from APNIC. If that is the case, maybe it would be useful to ensure that 'site' is less vaguely defined and change the policy to allow multi-site networks under common ownership and operation to qualify for an allocation?
It depends on what we consider is a multi-site network. A university may be split in many buildings in a single campus, or many buildings in several campuses. I think the policy should cover both cases.
I think you are conflating 'enterprise' and 'site'. I agree that a single enterprise might comprise multiple sites. It might be appropriate for (some?) multi-site enterprises to receive a /32 allocation. My concern is that the current definition of 'site' is vague and this makes it difficult to amend the policy in the way you want. If 'site' is clearly defined it will make it much easier to write policy text that is really clear on who qualifies for a /32 allocation.
Regards,
-- Leo Vegoda IANA Numbers Liaison
********************************************** The IPv6 Portal: http://www.ipv6tf.org
Bye 6Bone. Hi, IPv6 ! http://www.ipv6day.org
This electronic message contains information which may be privileged or confidential. The information is intended to be for the use of the individual(s) named above. If you are not the intended recipient be aware that any disclosure, copying, distribution or use of the contents of this information, including attached files, is prohibited.
On Jan 25, 2007, at 5:35 PM, JORDI PALET MARTINEZ wrote:
So then you're suggesting that we need also a complete review of the "site" definition ? Any specific suggestions ?
I think it would be less complicated to just make a proposal for certain classes of non-ISP network to qualify for an allocation. Changing the definition of 'end site' makes things more complicated and is likely to have unanticipated knock-on consequences.
I'm still not entirely certain which groups of networks you want to include. The example you gave was a university but there are a number of universities that already have IPv6 allocations without this policy. How are they different from the universities you are trying to help? Is it a major difference in the way they run their networks or is it something else?
Regards,
Activity Summary
- 6083 days inactive
- 6083 days old
- sig-policy@lists.apnic.net
- 3 participants
- 7 comments