May 1994 - lir-wg - mailman.ripe.net

Re: Address space for individuals
by Per Gregers Bilse 25 Aug '94

25 Aug '94

Ignoring for a minute that I don't work with local-ir stuff in any way or capacity, I find it odd the lengths ``we'' (Internet NICs, registries, providers) go to, to help and support people and organisations who don't want to play our game, ie connect to the global Internet, anyway. This is even more so, given the address space depletion (although opinions wrt this problem varies). Individuals or micro companies who want to connect a couple of PCs on a LAN can renumber the day they connect. Larger companies who clearly state that they have no intention of connecting, and want numbers for closed networks, don't need unique numbers (this has been discussed elsewhere recently). Medium size companies who intend to connect (later) -- no problem, of course they can get official numbers. But is it completely outrageous to consider unique/official IP numbers to be the property of the global, connected Internet? And categorically rule that they will only be issued to members of this crowd, ie those that actually (intend to) connect? "Intend to connect" ... A sledgehammer approach (nothing new in this, but I think it bears repeating): Organisations get allocations as usual, with the usual justifications for size etc. If somebody don't use (ie connect and route) at least part of their allocation within, say, 12 months, they have forfeited their chance. Their numbers may be re-assigned to somebody else, who now has first call. If the original requestor wants to connect at a later stage, they'll have to renumber. Considering the "interesting" future ahead of us, when IP numbers become really scarce, I don't see any reason to dole them out right, left and center to people who don't intend to use them. Unless we want to start building the future market of IP numbers; I wonder what a /16 will sell for around the turn of the century. -- bilse <bilse(a)EU.net> +31 20 592 5109 (dir: 5110); fax +31 20 592 5163 ``We used to ! but now we @'' (jensen)

2 2

Re: Address space for individuals
by Hank Nussbacher 29 May '94

29 May '94

On Tue, 24 May 94 12:31:14 +0200 you said: >>is the decision to be made then I would like to see a table of all RIPE >>countries that provide free registry services with who provides the >>funding: >> >>a) government funded registry >>b) university funded registry >>c) country Internet consortium funded registry > >For DE it's model c above. > >>d) other >> >>Do people feel that such a table would be worthwhile to determine who >>is currently funding registry services in Europe today? > >Yes - very good idea. > >>Hank >Dave Ok. Plz send me your reply to the above and I will summerize to the list. Please indicate what region you manage. Thanks, Hank

1 0

map of assigned networks
by Willi Huber 24 May '94

24 May '94

Hallo, prior to the past RIPE meeting I sent a (rather long) message onto this list about a 'graphical' representation of the assigned networks from the blocks 193/8 - 194/8. Nandor Horvath has asked me to make available the script that generated this output. Here it is. Make with it whatever you like. Willi Huber -------------- An example: netmap 193.5/16 <ripe.db netmap: looking for networks 193.5.0 <= network < 193.6.0 193.005.000 .*#*#...**##****#*#*#*#*#*##*#*#****#*#*#*#*##**#*#*#*##**#.*#*# 193.005.064 ....**##****####*#*#*#*#********#*#*#*#*#*#*####****####*#*#*#** 193.005.128 ################******..####****#####...****##**##*#************ 193.005.192 ########################********################**********#.*#* -------------- #!/bin/perl $c = "*"; $counter = 65; $arg = shift(@ARGV); if ($arg =~ /^\d+.*\/\d+$/) { ($firstnet,$bits) = split(/\//,$arg); } else { $bits = 32; } &binnet($firstnet); $bits = 24-$bits; if ($bits < 0) { print <<"end_of_text"; exit; } Usage: netmap net/bits file net/bits: the range of network addresses to be analysed file: the input file in the format of the ripe.db. If file is omitted, STDIN is read example: netmap 193/8 ripe.db NOTE: this script does not run propperly for class B addresses, use it only with calss C's. end_of_text $mask = 1; while ($bits-- > 0) { $mask *= 2; } $firstnet = int($firstnet / $mask) * $mask; $lastnet = $firstnet+$mask; print "\nnetmap: looking for networks "; &printnet($firstnet); print " <= network < "; &printnet($lastnet); print "\n\n"; open(INET,">/tmp/netmap.$$") || die "netmap: Couldn't open >/tmp/netmap.$$\n"; while (<>) { if (/\*in/) { chop; ($key,$value) = split(/\s+/,$_,2); if ($value =~ /-/) { ($value,$value2) = $value =~ /(\S+) - (\S+)/; &binnet($value); &binnet($value2); } else { &binnet($value); $value2 = $value; } if (($value >= $firstnet) && ($value < $lastnet)) { printf INET "%09d %09d\n",$value,$value2; } } } close INET; $lastplotted = $firstnet-1; open(SORTED,"sort /tmp/netmap.$$ |") || die "netmap: Couldn't open sort /tmp/netmap.$$ |\n"; while (<SORTED>) { ($value,$value2) = split; &plot($value,$value2); } print "\n"; close SORTED; unlink("/tmp/netmap.$$"); exit; sub binnet { undef(@bytes); (@bytes) = split(/\./,$_[0]); $_[0] = ($bytes[0]*256+$bytes[1])*256+$bytes[2]; } sub printnet { local($i) = $_[0]; printf "%d.%d.%d", int($i / 0x10000), int($i / 0x100 % 0x100), $i % 0x100; } sub plot { local($v,$v2) = @_; if ($lastplotted >= $v) { return; } $lastplotted++; while ($lastplotted != $v) { &plotnet(".",$lastplotted); $lastplotted++; } &plotnet($c,$v); while ($v != $v2) { $v++; &plotnet($c,$v); } $lastplotted = $v2; if ($c eq "*") { $c = "#"; } else { $c = "*" } } sub plotnet { local($c,$i) = @_; if ($counter >= 64) { printf "\n%03d.%03d.%03d ", int($i / 0x10000), int($i / 0x100 % 0x100), $i % 0x100; $counter = 0; } $counter++; print $c; }

1 0

Re: Address space for individuals
by Dave Morton 24 May '94

24 May '94

>is the decision to be made then I would like to see a table of all RIPE >countries that provide free registry services with who provides the >funding: > >a) government funded registry >b) university funded registry >c) country Internet consortium funded registry For DE it's model c above. >d) other > >Do people feel that such a table would be worthwhile to determine who >is currently funding registry services in Europe today? Yes - very good idea. >Hank Dave

1 0

Re: Address space for individuals
by Erik-Jan Bos 24 May '94

24 May '94

Daniel, > > Geert Jan de Groot <GeertJan.deGroot(a)ripe.net> writes: > > > - If this person only has a few hosts, then it is probably a good idea to > > ask him to renumber once he connects to the Internet. I don't believe > > that renumbering 3 PC's would be that much of a problem. > > 1597 might be useful after all.. > > This sums up my personal opinion. Great, quite along my personal opinion, but we need a consistent approach among all Local IRs. > If they are not going to connect immediately, then let them use private > address space and renumber their 3 hosts later. > > If they are going to connect immediately, let the service provider > registry assign numbers. Sure. > I know of cases where they subnet part of > the SP space. Soon - when we have a classless allocation registry, > this can even be registered. The world, now classless, might have /30s and /29s all over the place. I do not want to think of /32s. This gives us a neat "tool" to make sure that everybody has address space assigned to her or him that fits the needs. Looking into my cristal ball (sorry, I sound like somebody else :-) ), I see a world in which the bakery on the corner of the street has a brand new /28 assigned to his one-man company by a ISP he selected. After a few month the guy making bread discovers there is an ISP for the bakery branch in his city and he wants to switch over to the bakery-ISP. This ISP welcomes his new customer with open arms and announces this /28 to the Internet at large. Remember, this /28 is from the first ISP in this story. Think of what this will do to the efficiency of CIDR... Bottom line of this story is that there needs to be a mechanism in place that forces the bakery to renumber to a CIDR range of his new service provider. For the bakery with a /28 this is not too complex, but what about this large company with a /15? __ Erik-Jan.

10 16

Re: Address space for individuals
by Dave Morton 24 May '94

24 May '94

>On Fri, 20 May 94 17:32:13 BST you said: >>One point I'd suggest though -- I don't believe it would be realistic >>to expect a uniform charge across Europe. As with everything else in life, >>costs will vary from country to country. One discussion point will be Yes - agreed Bob, we've already this problem in many pan-European projects. Trying to explain why overheads are different from PT to DE is not easy. >>whether one wants the idea of free competition across Europe, with >>customers "shopping around" (the Maastricht ideal, I suppose), or a >>situation where customers aren't "allowed" to apply in other than >>their own country (ie regulation, or a cartel, depending on your view.) >>If the latter then there's an interesting consequence of what to do >>if a one country's Last Resort Registry starts charging unreasonably >>high prices ... this is a problem that hasn't been solved yet in the >>well known area of line provision! In the case of IP numbers and domains one can sort of argue that a certain monopolistic type structure is required for technical reasons (cidr, tlds etc.) - how long this holds if we have competing pan-European or global service providers is crystal ball stuff. >Imagine the havoc it would play on CIDR if everyone shopped around >to save a few ECUs on an IP address. RIPE could come out with >a recommendation range, i.e. $10-$30 per class C. Not possible Hank - I don't believe RIPE should try to regulate markets, assuming such a market were to emerge. But you're right - it'll play havoc with cidr if such a market does emerge before IPng gets widely implemented, available and installed. How is NSAP allocation done in other European countries ? I know that here in DE one obtains NSAP for around 32ecu per NSAP per year. Does this compare and is there some sort of international (european agreement on this) - I doubt it ? >>Could be an interesting discussion.... Indeed. >> >>Bob Day > >Hank Nussbacher Dave

1 0

Re: Address space for individuals
by Dave Morton 24 May '94

24 May '94

> > poole(a)eunet.ch (Simon Poole) writes: > > > > A note on the side: not allowing purchase of IP addresses in other > > countries would very likely be illegal at least inside the EU including > > any other restrictions on usage of not nationally obtained IP addresses. > >Stop it right there. There is no purchase involved. Once we talk about >purchase and ownership we are doomed. No aggrgation, no reclamation, >no leasing to refuce consumption. Erase those words from the discussion >please. Yes - agreed. But then one is now faced with the problem of explaining. "Ahem - well remember that Class Bs we assigned you back in 1989 - well we'd now like to rent it to you for X ecu per annum - please transfer X or esle". >Now to Simon's remark: Where can I get a Swiss phone number in the >Netherlands? I think we should not be overly concerned with this >problem at the present time. Not so sure - currently one can always get a gsm phone in country a assuming one lives in country b and you don't like the tariffs in country b. yes - hypothetical perhaps and unlikely but... > > The main problem is that if it's a small enough ammount to be painless, > > it's too expensive to bill for (a one time bill for $10 is clearly not > > going to make any sense at all). > >I support leasing of address space eventually. Me too but how are we going to explain this to people ? >With all the problems involved. >One idea that is feasible then is to lease small amounts of address >space for longer times with advance payment. This gets around the >small bills problem. Yes - that was my suggestion to DE-NIC (also for domain registration ala Hank's problem) but this was rejected as unworkable. If SPs can send out bills for 35KECU per month then why not also for renting address space and domain registration. Have we looked at the possibility of a routing charge ? >Daniel Dave

1 0

Re: Address space for individuals
by Hank Nussbacher 24 May '94

24 May '94

On Fri, 20 May 94 17:32:13 BST you said: >One point I'd suggest though -- I don't believe it would be realistic >to expect a uniform charge across Europe. As with everything else in life, >costs will vary from country to country. One discussion point will be >whether one wants the idea of free competition across Europe, with >customers "shopping around" (the Maastricht ideal, I suppose), or a >situation where customers aren't "allowed" to apply in other than >their own country (ie regulation, or a cartel, depending on your view.) >If the latter then there's an interesting consequence of what to do >if a one country's Last Resort Registry starts charging unreasonably >high prices ... this is a problem that hasn't been solved yet in the >well known area of line provision! Imagine the havoc it would play on CIDR if everyone shopped around to save a few ECUs on an IP address. RIPE could come out with a recommendation range, i.e. $10-$30 per class C. > >Could be an interesting discussion.... > >Bob Day Hank Nussbacher

4 4

Re: Address space for individuals
by Dave Morton 24 May '94

24 May '94

> > Arnold Nipper <nipper(a)xlink.net> writes: > > > > We had lot of discussion about that within the DE-NIC steering-financing > > committee. We finally decided to give every one a free shot. If the applica > > nt > > sent in a detailed technical plan justifying the requested address space, i > > t's > > fine. Otherwise the templates are sent back with addresses where he/she may > > buy consultancy. News to me... >Who gets on that list? Presumably a select few :-) Dave

1 0

Re: Address space for individuals
by bob＠informatics.rutherford.ac.uk 24 May '94

24 May '94

> It would help the debate if, as early as possible, other > last resort registries could indicate their views, their > current practice and their funding methods. If recommendations > are to be made, they should be based on as wide a poll as > possible. For the record, the UK situation. Within JANET, we undertake two registry roles: 1. We assign numbers for those connecting to JANET. This is part of the general service we provide to customers connecting to JANET, and is effectively funded through the JANET connection charge and annual rental. Therefore we wouldn't unbundle a charge for network number assignment in these cases. 2. We're the NIC of last resort for the UK. This service is effectively unfunded -- we do it as part of the JANET registry activities. Our funding authorities in general think that doing this sort of thing is desirable (ie in the interests of networking in the UK), but (if we were to ask) wouldn't pay for it directly -- they'd expect us to fund it by charging those wanting numbers. Obviously, we don't bother asking, but the operation of this service is a constant drain, with the real possibility that we'll have to give it up (ie hand it back to the NCC) if it gets more onerous that at present. For information, I'd reckon that processing a "simple" request, billing etc would cost around 250 ECU. A more complex one, where there's a need to discuss the situation in detail with the applicant might cost 500 ECU, maybe more. With these sort of figures it'd probably be worth insisting on having the registration fee come with the application, rather than billing afterwards -- this should cut costs a bit. You'd still have to generate a VAT receipt, though. Bob

4 3