Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assignment Global Policy
Brian, I realize that there is what we consider to be a huge amount of v6 address space. What you have to understand though is that you have a bunch of folks who are living day to day with the last time the powers that be decided that we had more address space than we could possible use. They gave out A's and B's and they too thought that there was more than we could possibly ever use. Now the registries (ARIN in particular) spends a lot of time trying to figure out how to reclaim that space. There are a lot of folks who feel that we're repeating old mistakes. Note also that if there are no justifications required to get a /32 except becoming what RIPE calls an LIR (this term varies from region to region), this means that it won't be long before folks there are many many small multihomed sites that have these blocks. Mirjam said herself in the meeting that the reason that RIPE has justification stuff required for V4 blocks now is because folks had figured this out and started getting LIR space just because they're multihimed. It isn't like we're starting from scratch here, folks are pretty sophistocated these days. Further since the actual methods of handing out the space vary from region to region it means that there may be more hurdles than just becoming an LIR. For example, in the RIPE region, when you get a block you can announce the whole block, but you have to ask permission to actually use each subnet (no matter how small). Of course their v6 policy may be different, who knows. This is not the same way that it works in all regions. Unless you get space from all three (soon to be 4) it is difficult to really understand the subtle differences. It wasn't my intent to argue this here, it was just my intent to point out that I doubted that this policy would reach consensus in all regions because of the past history. I misinterpreted the original note as a submission of changes for the document. I thought it was premature to make changes at this point. I am glad that wasn't the intent of the note. We'll see what happens. As Anne said (thanks Anne!) the APNIC meeting is coming up. The ARIN policy meeting is in April. Thanks! ---CJ From: Brian E Carpenter <brian@hursley.ibm.com> Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assignment Global Policy Cathy, I'm still quite surprised by the way in which the fully justified conservatism of the registries for IPv4 space is being over-extrapolated to IPv6 space. If you look objectively at the argument that Gert gives, and consider how the size of the IPv6 prefix space compares to the total IPv4 space, /32 just isn't risky, and it gets rid of yet another judgement call. (I wasn't at the RIPE meeting either, but I did hear Mirjam talk on this topic yesterday.) Brian CJ Wittbrodt wrote: > > Since the events of the recent RIPE working group meeting have not > been discussed at either of the other regional policy forums, > proposing something here may be somewhat premature. I do not believe, > although things never cease to amaze me, that this will reach any sort > of consensus within the ARIN region. I am not sure about the APNIC > region. Based on some meetings with the European Government > Advisory Council (just after the RIPE meeting) it is clear that it is > important that we have a global policy. Is there maybe some way > that we could come up with a compromise that would reach consensus > in all three policy forums? Something other than requiring no > justification for a /32? > > Thanks > ---CJ Wittbrodt > (ARIN Advisory Council and ASO Address Council member) > > From: Gert Doering <gert@Space.Net> > Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assig
nment > >>Global Policy > Hi, > > from your comments, I gather you have not been to the RIPE IPv6/LIR > policy meeting. So let's add a few comments (while waiting for James > Aldridge to publish the "official" word on it): > > On Fri, Feb 01, 2002 at 02:50:38PM -0500, Thomas Narten wrote: > > > 5.2.1. Initial allocation criteria > [..] > > The goal was to give a site a /32 if it can justify it will use > > it. The word "immediately" (as in demonstrate an immediate need) is > [..] > > So the real issue here (and this comes up again in later parts of the > > document) is what is a reasonable way to objectively evaluate a > > request for address space that requires some guessing as to whether a > > proposed plan will actually be carried out. If the time frame is too > > long, it becomes easy to make optimistic plans that won't pan out, and > > then the RIRs get into a different problem. > > Consensus on the IPv6/LIR policy meeting was "drop the criteria". > > To be precise, I proposed the following: > > - any LIR that is established (has done all the paperwork, paid their > fees, and whatnot) and can document the need for one IPv6 address > can get a /32. No further justification required. > > - to avoid a horrible mistake, every region is permitted to allow only > assigment of 2000 /32s per region. So the maximum wastage is 6000 > /32s (out of 500 million /32s in the 1/8th of the space we're talking > about), and 6000 additional routes. > After that, we're going to reconsider policy. > > There was concern from the other regional registries (ARIN and APNIC), > but broad consensus from the people from the RIPE region. > > Reasoning (shortened): > > - why are we putting criteria there? To keep out "some that we do not > want". Conservation is not an issue. Routing table growth might > be influenced by this, or might be not, we don't know. > > - do we want major national research networks connecting something like > "50 universities"? YES > > - will this research network meet any criteria based on "you must use up > a big number of /48s, otherwise you can't get a /32"? NO, if you > assign a /48 per university (which would be plenty!), because that > means "you can only demonstrate a need for 50 /48s" > > On the other hand, if you say "I connect lots of private customers > over DSL lines, using fixed IP addresses, giving each user a /48 (which > is OK according to IETF guidelines)", reaching over 50 /48s is very > easy. > > Does this mean the second example is "more worthy" to get a /32? Does > it mean they will make "better use" of it? > > So all technical criteria based on /48 usage must fail, and criteria > based on single IP usage will fail as well (due to the /48 rule). If we > can't propose criteria that work, drop them - BUT limit the amount of > damage that can be done. > > [..] > > Or is it the 776 end site figure (i.e., too high)? > > Think of the research networks. One /48 per university would be > "according to the /48 rule: each SITE gets a /48". > > Gert Doering > -- NetMaster > -- > Total number of prefixes smaller than registry allocations: 71770 (72395) > > SpaceNet AG Mail: netmaster@Space.Net > Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 > 80807 Muenchen Fax : +49-89-32356-299 >
Cathy, Yes, I understand why people are nervous about repeating historical mistakes. What I was clumsily trying to express is that objectively, I don't think we need to be particularly nervous about whether we give out /32 or /35 at the large end, or /48 or /51 (say) at the small end - yes I know that's a factor 8 in each case, but we do have that much slack in terms of available space. What is very nervous-making is the risk of people believing that they need, and can get, PI prefixes, and a lot of the last day's discussion is right on the mark about the risks there. Brian CJ Wittbrodt wrote:
Brian,
I realize that there is what we consider to be a huge amount of v6 address space. What you have to understand though is that you have a bunch of folks who are living day to day with the last time the powers that be decided that we had more address space than we could possible use. They gave out A's and B's and they too thought that there was more than we could possibly ever use. Now the registries (ARIN in particular) spends a lot of time trying to figure out how to reclaim that space. There are a lot of folks who feel that we're repeating old mistakes.
Note also that if there are no justifications required to get a /32 except becoming what RIPE calls an LIR (this term varies from region to region), this means that it won't be long before folks there are many many small multihomed sites that have these blocks. Mirjam said herself in the meeting that the reason that RIPE has justification stuff required for V4 blocks now is because folks had figured this out and started getting LIR space just because they're multihimed. It isn't like we're starting from scratch here, folks are pretty sophistocated these days. Further since the actual methods of handing out the space vary from region to region it means that there may be more hurdles than just becoming an LIR. For example, in the RIPE region, when you get a block you can announce the whole block, but you have to ask permission to actually use each subnet (no matter how small). Of course their v6 policy may be different, who knows. This is not the same way that it works in all regions. Unless you get space from all three (soon to be 4) it is difficult to really understand the subtle differences.
It wasn't my intent to argue this here, it was just my intent to point out that I doubted that this policy would reach consensus in all regions because of the past history. I misinterpreted the original note as a submission of changes for the document. I thought it was premature to make changes at this point. I am glad that wasn't the intent of the note.
We'll see what happens. As Anne said (thanks Anne!) the APNIC meeting is coming up. The ARIN policy meeting is in April.
Thanks! ---CJ
From: Brian E Carpenter <brian@hursley.ibm.com> Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assignment Global Policy Cathy,
I'm still quite surprised by the way in which the fully justified conservatism of the registries for IPv4 space is being over-extrapolated to IPv6 space. If you look objectively at the argument that Gert gives, and consider how the size of the IPv6 prefix space compares to the total IPv4 space, /32 just isn't risky, and it gets rid of yet another judgement call.
(I wasn't at the RIPE meeting either, but I did hear Mirjam talk on this topic yesterday.)
Brian
CJ Wittbrodt wrote: > > Since the events of the recent RIPE working group meeting have not > been discussed at either of the other regional policy forums, > proposing something here may be somewhat premature. I do not believe, > although things never cease to amaze me, that this will reach any sort > of consensus within the ARIN region. I am not sure about the APNIC > region. Based on some meetings with the European Government > Advisory Council (just after the RIPE meeting) it is clear that it is > important that we have a global policy. Is there maybe some way > that we could come up with a compromise that would reach consensus > in all three policy forums? Something other than requiring no > justification for a /32? > > Thanks > ---CJ Wittbrodt > (ARIN Advisory Council and ASO Address Council member) > > From: Gert Doering <gert@Space.Net> > Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assig
nment > >>Global Policy > Hi, > > from your comments, I gather you have not been to the RIPE IPv6/LIR > policy meeting. So let's add a few comments (while waiting for James > Aldridge to publish the "official" word on it): > > On Fri, Feb 01, 2002 at 02:50:38PM -0500, Thomas Narten wrote: > > > 5.2.1. Initial allocation criteria > [..] > > The goal was to give a site a /32 if it can justify it will use > > it. The word "immediately" (as in demonstrate an immediate need) is > [..] > > So the real issue here (and this comes up again in later parts of the > > document) is what is a reasonable way to objectively evaluate a > > request for address space that requires some guessing as to whether a > > proposed plan will actually be carried out. If the time frame is too > > long, it becomes easy to make optimistic plans that won't pan out, and > > then the RIRs get into a different problem. > > Consensus on the IPv6/LIR policy meeting was "drop the criteria". > > To be precise, I proposed the following: > > - any LIR that is established (has done all the paperwork, paid their > fees, and whatnot) and can document the need for one IPv6 address > can get a /32. No further justification required. > > - to avoid a horrible mistake, every region is permitted to allow only > assigment of 2000 /32s per region. So the maximum wastage is 6000 > /32s (out of 500 million /32s in the 1/8th of the space we're talking > about), and 6000 additional routes. > After that, we're going to reconsider policy. > > There was concern from the other regional registries (ARIN and APNIC), > but broad consensus from the people from the RIPE region. > > Reasoning (shortened): > > - why are we putting criteria there? To keep out "some that we do not > want". Conservation is not an issue. Routing table growth might > be influenced by this, or might be not, we don't know. > > - do we want major national research networks connecting something like > "50 universities"? YES > > - will this research network meet any criteria based on "you must use up > a big number of /48s, otherwise you can't get a /32"? NO, if you > assign a /48 per university (which would be plenty!), because that > means "you can only demonstrate a need for 50 /48s" > > On the other hand, if you say "I connect lots of private customers > over DSL lines, using fixed IP addresses, giving each user a /48 (which > is OK according to IETF guidelines)", reaching over 50 /48s is very > easy. > > Does this mean the second example is "more worthy" to get a /32? Does > it mean they will make "better use" of it? > > So all technical criteria based on /48 usage must fail, and criteria > based on single IP usage will fail as well (due to the /48 rule). If we > can't propose criteria that work, drop them - BUT limit the amount of > damage that can be done. > > [..] > > Or is it the 776 end site figure (i.e., too high)? > > Think of the research networks. One /48 per university would be > "according to the /48 rule: each SITE gets a /48". > > Gert Doering > -- NetMaster > -- > Total number of prefixes smaller than registry allocations: 71770 (72395) > > SpaceNet AG Mail: netmaster@Space.Net > Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 > 80807 Muenchen Fax : +49-89-32356-299 >
--On Thursday, February 7, 2002 17:29 +0100 Brian E Carpenter <brian@hursley.ibm.com> wrote:
Cathy,
Yes, I understand why people are nervous about repeating historical mistakes. What I was clumsily trying to express is that objectively, I don't think we need to be particularly nervous about whether we give out /32 or /35 at the large end, or /48 or /51 (say) at the small end - yes I know that's a factor 8 in each case, but we do have that much slack in terms of available space. What is very nervous-making is the risk of people believing that they need, and can get, PI prefixes, and a lot of the last day's discussion is right on the mark about the risks there.
So perhaps we should focus on the question of who will get PI allocations before we worry about how large those allocations should be? Along those lines, perhaps we should figure out how sites are going to multi-home, since that will (well, should) have a huge bearing on who gets PI alloocations. Alec -- Alec H. Peterson -- ahp@hilander.com Chief Technology Officer Catbird Networks, http://www.catbird.com
"Alec H. Peterson" wrote:
--On Thursday, February 7, 2002 17:29 +0100 Brian E Carpenter <brian@hursley.ibm.com> wrote:
Cathy,
Yes, I understand why people are nervous about repeating historical mistakes. What I was clumsily trying to express is that objectively, I don't think we need to be particularly nervous about whether we give out /32 or /35 at the large end, or /48 or /51 (say) at the small end - yes I know that's a factor 8 in each case, but we do have that much slack in terms of available space. What is very nervous-making is the risk of people believing that they need, and can get, PI prefixes, and a lot of the last day's discussion is right on the mark about the risks there.
So perhaps we should focus on the question of who will get PI allocations before we worry about how large those allocations should be?
Along those lines, perhaps we should figure out how sites are going to multi-home, since that will (well, should) have a huge bearing on who gets PI alloocations.
Indeed. That is exactly why the IETF has a MULTI6 working group. Brian
--On Friday, February 8, 2002 9:37 +0100 Brian E Carpenter <brian@hursley.ibm.com> wrote:
Indeed. That is exactly why the IETF has a MULTI6 working group.
Yes, I am aware of the multi6 working group. However, that doesn't change the fact that we are making a rather important decision with nowhere near all of the data, because multi6's work is nowhere near done. So perhaps we have to resign ourselves to making a decision without all of the facts, but I think this warrants putting some definite limits around the scope of our decision so that we don't get hit too hard by it down the road. Alec -- Alec H. Peterson -- ahp@hilander.com Chief Technology Officer Catbird Networks, http://www.catbird.com
So perhaps we should focus on the question of who will get PI allocations before we worry about how large those allocations should be?
Along those lines, perhaps we should figure out how sites are going to multi-home, since that will (well, should) have a huge bearing on who gets PI alloocations.
Indeed. That is exactly why the IETF has a MULTI6 working group.
and the lack of progress and and solutions in this space, and v6 routing in general, may be what is causing many folk to be conservative. randy
The DNS will be quickly evolving to include both 32-bit A and 128-bit AAAA records. Using AAAA records does not imply that IPv6 will be used. In fact, 99.9% of the world is just discovering TOS routing in IPv4, and some of the 128 bits can be used to control the TOS settings. Some of the other 128-bits can be used to specify the 16-bit Port #, which did not fit in 32-bit A records and which people had to kludge with :Port domain name extensions. As for other bits in the 128-bit field, it seems obvious to use at least 32-bits for legacy IPv4 addresses when a Routing Header or Source Routing is desired. Another 32-bit field is also desirable to make it easy for legacy networks to be merged or mapped to the new address spaces. At this point the sub-total is: 32 - Routing 4 - TOS 32 - Legacy 16 - Port ------- 84 - Sub-Total with 48 to define If you turn to computer architecture and instruction set design and object-oriented engine experience, the 44 bits that remain are nicely divided into a 16 bit Op-Code, a 16-bit Security/Key/Random Field, and a 12-bit prefix for Aggregation. If you select the Op-Code to be 2002 for this year's experiments, and ignore the other 16-bit field for the moment, you then turn to the 12-bits to decide who are the 4,096 most likely managers of the space who will be fair and will not gravitate to graft and corruption and kick-backs and back-room deals and other mafia-like activities to manage the space. To be safe, one could divide that population into two 2,048 groups with 11 bits to denote the manager. For people used to playing in the graft and corruption game, that seems like a large number, they want to "keep it in the family" so to speak. They want their recurring fees to help pay for their Miami and Las Vegas vacations and trips to Africa. Unfortunately, they are finding that next year, the Op-Code could become 2003 and they lose their grip and lose their interest in the whole process. That coupled with the fact that the 128-bits do not have to be used with IPv6 but mostly can be retrofit into IPv4 headers, and people quickly realize that they are not going to be able to live like kings on the gold-coast down-under, doing almost nothing, as they collect taxes for allocations. Of course, all of this is outside of the IETF, which does not involve itself in any business discussions or other market manipulations which could bring it and the ISOC under anti-trust scrutiny. It is interesting that the same people from the IETF are involved, and appear to have no plan. Meanwhile, Linux and BSD are rapidly gaining all of the features needed to move to 128-bit DNS. Non-ICANN registrars and registries will be the leaders in this arena. The 128-bits will fall where the market-place dictates and the people will be free, from the corruption of the IPv4 address management industry. Jim Fleming http://www.IPv8.info ----- Original Message ----- From: "Brian E Carpenter" <brian@hursley.ibm.com> To: "CJ Wittbrodt" <cjw@groovy.com> Cc: "Gert Doering" <gert@Space.Net>; "Thomas Narten" <narten@us.ibm.com>; "Stuart Prevost" <stuart.prevost@btinternet.com>; <global-v6@lists.apnic.net>; <ipv6-wg@ripe.net> Sent: Thursday, February 07, 2002 10:29 AM Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address Allocation and Assignment Global Policy
Cathy,
Yes, I understand why people are nervous about repeating historical mistakes. What I was clumsily trying to express is that objectively, I don't think we need to be particularly nervous about whether we give out /32 or /35 at the large end, or /48 or /51 (say) at the small end - yes I know that's a factor 8 in each case, but we do have that much slack in terms of available space. What is very nervous-making is the risk of people believing that they need, and can get, PI prefixes, and a lot of the last day's discussion is right on the mark about the risks there.
Brian
CJ Wittbrodt wrote:
Brian,
I realize that there is what we consider to be a huge amount of v6 address space. What you have to understand though is that you have a bunch of folks who are living day to day with the last time the powers that be decided that we had more address space than we could possible use. They gave out A's and B's and they too thought that there was more than we could possibly ever use. Now the registries (ARIN in particular) spends a lot of time trying to figure out how to reclaim that space. There are a lot of folks who feel that we're repeating old mistakes.
Note also that if there are no justifications required to get a /32 except becoming what RIPE calls an LIR (this term varies from region to region), this means that it won't be long before folks there are many many small multihomed sites that have these blocks. Mirjam said herself in the meeting that the reason that RIPE has justification stuff required for V4 blocks now is because folks had figured this out and started getting LIR space just because they're multihimed. It isn't like we're starting from scratch here, folks are pretty sophistocated these days. Further since the actual methods of handing out the space vary from region to region it means that there may be more hurdles than just becoming an LIR. For example, in the RIPE region, when you get a block you can announce the whole block, but you have to ask permission to actually use each subnet (no matter how small). Of course their v6 policy may be different, who knows. This is not the same way that it works in all regions. Unless you get space from all three (soon to be 4) it is difficult to really understand the subtle differences.
It wasn't my intent to argue this here, it was just my intent to point out that I doubted that this policy would reach consensus in all regions because of the past history. I misinterpreted the original note as a submission of changes for the document. I thought it was premature to make changes at this point. I am glad that wasn't the intent of the note.
We'll see what happens. As Anne said (thanks Anne!) the APNIC meeting is coming up. The ARIN policy meeting is in April.
Thanks! ---CJ
From: Brian E Carpenter <brian@hursley.ibm.com> Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address
Allocation and
Assignment Global Policy Cathy,
I'm still quite surprised by the way in which the fully justified conservatism of the registries for IPv4 space is being
over-extrapolated
to IPv6 space. If you look objectively at the argument that Gert gives, and consider how the size of the IPv6 prefix space compares to the total IPv4 space, /32 just isn't risky, and it gets rid of yet another judgement call.
(I wasn't at the RIPE meeting either, but I did hear Mirjam talk on this topic yesterday.)
Brian
CJ Wittbrodt wrote: > > Since the events of the recent RIPE working group meeting have not > been discussed at either of the other regional policy forums, > proposing something here may be somewhat premature. I do not
believe,
> although things never cease to amaze me, that this will reach any
sort
> of consensus within the ARIN region. I am not sure about the
APNIC
> region. Based on some meetings with the European Government > Advisory Council (just after the RIPE meeting) it is clear that it
is
> important that we have a global policy. Is there maybe some way > that we could come up with a compromise that would reach consensus > in all three policy forums? Something other than requiring no > justification for a /32? > > Thanks > ---CJ Wittbrodt > (ARIN Advisory Council and ASO Address Council member) > > From: Gert Doering <gert@Space.Net> > Subject: Re: [GLOBAL-V6] New draft available: IPv6 Address
Allocation and Assig
nment > >>Global Policy > Hi, > > from your comments, I gather you have not been to the RIPE IPv6/LIR > policy meeting. So let's add a few comments (while waiting for James > Aldridge to publish the "official" word on it): > > On Fri, Feb 01, 2002 at 02:50:38PM -0500, Thomas Narten wrote: > > > 5.2.1. Initial allocation criteria > [..] > > The goal was to give a site a /32 if it can justify it will use > > it. The word "immediately" (as in demonstrate an immediate need) is > [..] > > So the real issue here (and this comes up again in later parts of the > > document) is what is a reasonable way to objectively evaluate a > > request for address space that requires some guessing as to whether a > > proposed plan will actually be carried out. If the time frame is too > > long, it becomes easy to make optimistic plans that won't pan out, and > > then the RIRs get into a different problem. > > Consensus on the IPv6/LIR policy meeting was "drop the criteria". > > To be precise, I proposed the following: > > - any LIR that is established (has done all the paperwork, paid their > fees, and whatnot) and can document the need for one IPv6 address > can get a /32. No further justification required. > > - to avoid a horrible mistake, every region is permitted to allow only > assigment of 2000 /32s per region. So the maximum wastage is 6000 > /32s (out of 500 million /32s in the 1/8th of the space we're talking > about), and 6000 additional routes. > After that, we're going to reconsider policy. > > There was concern from the other regional registries (ARIN and APNIC), > but broad consensus from the people from the RIPE region. > > Reasoning (shortened): > > - why are we putting criteria there? To keep out "some that we do not > want". Conservation is not an issue. Routing table growth might > be influenced by this, or might be not, we don't know. > > - do we want major national research networks connecting something like > "50 universities"? YES > > - will this research network meet any criteria based on "you must use up > a big number of /48s, otherwise you can't get a /32"? NO, if you > assign a /48 per university (which would be plenty!), because that > means "you can only demonstrate a need for 50 /48s" > > On the other hand, if you say "I connect lots of private customers > over DSL lines, using fixed IP addresses, giving each user a /48 (which > is OK according to IETF guidelines)", reaching over 50 /48s is very > easy. > > Does this mean the second example is "more worthy" to get a /32? Does > it mean they will make "better use" of it? > > So all technical criteria based on /48 usage must fail, and criteria > based on single IP usage will fail as well (due to the /48 rule). If we > can't propose criteria that work, drop them - BUT limit the amount of > damage that can be done. > > [..] > > Or is it the 776 end site figure (i.e., too high)? > > Think of the research networks. One /48 per university would be > "according to the /48 rule: each SITE gets a /48". > > Gert Doering > -- NetMaster > -- > Total number of prefixes smaller than registry allocations: 71770 (72395) > > SpaceNet AG Mail: netmaster@Space.Net > Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0 > 80807 Muenchen Fax : +49-89-32356-299 >
participants (5)
-
Alec H. Peterson -
Brian E Carpenter -
CJ Wittbrodt -
Jim Fleming -
Randy Bush