At 01:36 PM 6/4/97 -0400, you wrote:
Isn't the biggest problem with PI space having a big enough network so Sprint and everyone else doesn't filter you? Specifically a /19 or larger (~ 8000 addresses)? Can you justify to them 32 Class C's so you can be seen everywhere? I know of another startup provider who had this problem. Its a chicken and egg scenario. They needed to sell connectivity, but couldn't reach portions of the net, specifically Sprint. Kind of kills your marketing when you can't get to the entire 'Net. How do others deal with this scenario in a startup environment, or even an established environment where you can't justify 8000 addresses yet? Eric
Eric Germann enlightens us with ...
In effect Sprint is encouraging the waste of IP space. I'm putting together a proposal now for a web farm type of facility for a group of investors and a block of /19 is way more than is needed. But the plan is going to have at least 4 points of multi-homing to diverse backbone providers, so a fully announceable block is essential. Are Sprint and InterNIC butting heads over this? If not, does anyone here think they should be? I'm assuming the reason Sprint does this is to reduce the total number of routes they have to carry. What are the other backbones doing? How well is everyone else doing in terms of building up the capacity to carry more and more routes? What alternative plans are coming along, before and after IPv6 (which would certainly unleash the floodgates on routes if we are still using BGP technology at IPv6 deployment)? I've got one idea I'm mulling over that could help in the current situation. Once it is a more mature idea, I might suggest it. In the mean time I just want to read other thoughts, comments, ideas, and... uh... no... not flames. -- Phil Howard KA9WGN +-------------------------------------------------------+ Linux Consultant | Linux installation, configuration, administration, | Milepost Services | monitoring, maintenance, and diagnostic services. | phil at milepost.com +-------------------------------------------------------+
Hi, Phil Howard wrote:
Idea: How about getting provider-dependent space from each one, then make the web servers listen on different addresses each. Rig the DNS with a low TTL for the server A records, or perhaps use dynamic updates (haven't tried it yet though) to remove the IP from the A list if a link goes down. Example you get space from: ISP A 10.0.0.0/24 ISP B 10.1.0.0/24 ISP C 10.2.0.0/24 ISP D 10.3.0.0/24 So you have: www.customer1.com IN A 10.0.0.1 IN A 10.1.0.1 IN A 10.2.0.1 IN A 10.3.0.1 www.customer2.com IN A 10.0.0.2 IN A 10.1.0.2 IN A 10.2.0.2 IN A 10.3.0.2 Now if ISP C goes down, delete 10.2.0.1 and 10.2.0.2 from the list. -- miguel a.l. paraz <map@iphil.net> +63-2-893-0850 iphil communications, makati city, philippines <http://www.iphil.net>
Miguel, Interesting. I haven't contemplated any aspect of your idea other than nameserver restarts. With 30,000 domains hosted here, our nameserver (which, according to the INTERNIC, is authorative for more domains than any nameserver in the world) takes about 20 minutes to restart (it's an SGI Challenge S). Our secondaries take a little longer. That's with one A record per domain. Restarting a nameserver every time a link gets sick is not an option, at least for us. Steve At 07:02 PM 6/5/97 +0800, Miguel A.L. Paraz wrote:
On Wed, 4 Jun 1997, Phil Howard wrote:
I'd like to see them start allocating recovered space from 192/8 in /22 or maybe /21 sizes. ********************************************************* Michael Dillon voice: +1-415-482-2840 Senior Systems Architect fax: +1-415-482-2844 PRIORI NETWORKS, INC. http://www.priori.net "The People You Know. The People You Trust." *********************************************************
: I'd like to see them start allocating recovered space from 192/8 in : /22 or maybe /21 sizes. I asked for this and was treated like a complete idiot. There are a lot of small blocks in legacy space that I'd like to see some system for transferring between small guys who never used them and small guys who desperately need PI space for multihoming. Using DNS TTL fields is just ridiculous when there are much better technical solutions like, uh, BGP. A lot of people I know don't need /19s and don't want to waste the space (people have already mentioned this). Also, there is no incentive for returning IPv4 space if you know it'll just sit idle anyway. The function of the InterNIC can be as an intermediary, verifying, as usual, that the requestor needs the space. -Tung-Hui Hu hhui@arcfour.com
I am sorry that you felt treated like a "complete idiot". As there are still several registries that claim be be authoritative for 192/8, this is a tough block in which to start the piara experiment. It many ways, it still is like a toxic waste area... although with existant routing policy by ISPs its treated less like a toxic hazard and more like ambrosia, with one and all declareing this space to be special and above policy filtering constraints. Again, for more details on the history behind this discussion, pleae check the piara bof archives and old nanog notes. There are even some good tips on how to support multihoming w/o 192 space. In my humble opinion, its much better to clean up the whole range first and then re-release it as a clean /8, than haphazard redelgation. And yes, there is an incentive to return space. For some its the principle of the thing, a cooperative internet is a growing internet. For others it may be financial, with the expense of renewing lease delegations. (See the naipr/arin lists for more details). I hope this helps out some. I'd apprciate your comments. -- --bill
In conversation with some NANOG participants last week I was asked more than once for clarification or further details about 192/8, continuing from Bill's remarks Friday. It does seem that 192/8 isn't the concern it was eighteen months ago. One reason is that some carriers now refuse to route legacy 192/8 delegations for new customers, requiring instead that they renumber into provider space. Folks holding delegations but not yet trying to route them were never an enormous number of prefixes but looked for awhile like they might be a significant contributor to routing table growth; back-of-the-envelope calculations a little while ago suggest this potential doesn't seem to have materialized (it's hard to be sure because routing table announcements don't have origin dates, but there's some reason to believe new announcements are simply better aggregated than older ones, even in 192/8). An informal survey: how many NANOG participants have asked customers to renumber out of 192/8? And for how long have you been doing this? IMHO the remaining major obstacle to a concerted effort at reclamation in 192/8 is the database maintenance problem. Folks with no incentive to keep their whois entries current have not been doing so-- since the typical pre-CIDR "Class C" delegate was not an ISP and has never come back for more space, the registries have limited leverage over them, with a few notable exceptions such as DDN-NIC. Accordingly, "whois" contact information simply doesn't tell you who's using a block. I haven't risked a blind survey again since the first one, but I have no reason to believe this has changed-- any takers? I hope and expect that ARIN may be able to throw some organized effort at this problem, if the membership feels that cleaning up the database is important. Suzanne Woolf woolf@isi.edu (ISI pays me to run routers'n'DNS'n'things, my opinions are mine)
[ ... ]
An informal survey: how many NANOG participants have asked customers to renumber out of 192/8? And for how long have you been doing this?
InterNex has been refusing to route 192/8 blocks, and has been requiring customers to renumber out of them since the presentation at the Feb. NANOG in San Diego last year. We will work with customers to replace their legacy 192/8 blocks provided they return them to the InterNIC.
I have long hoped that there would be a minimal annual fee required for the continued usage of IP blocks. Even a $50/year fee would be enough that given non-payment, blocks could more easily be reclaimed. I'd expect a longer "On-Hold" time, say 180 days, where the blocks would no longer be considered a valid announcement, but they would also not be re-allocated by the local IR. Of course, this is purely personal, and not representative of my employer. :)
I would cast my vote in the "yes it's important" box. :)
Matt Petach
Bill writes:
In my humble opinion, its much better to clean up the whole range first and then re-release it as a clean /8, than haphazard redelgation.
As a contractor at a site with a /19 in 192 I would muchly prefer that we not have to renumber for no good reason. I understand garbage collecting small blocks, but there *are* some well-managed, large and clean allocations in 192. -george william herbert gherbert@crl.com
BGP isn't better. You don't need BGP for multihoming. My SF NANOG talk was on exactly this topic. If you get a little bit of PA space from each provider, and run a multihomed server (or a whole pile of them) with the "interface default" mods we made available, you don't need BGP at all. ftp://ftp.vix.com/pub/vixie/ifdefault/ is the place to look for this. I just stuck a copy of my SF NANOG slides in there (SOB, please take note!)
Sure there is. It's called "not fouling your own nest." If you're not using it, and it's too small to be routable by modern providers, then you should give it back and let someone else try to aggregate it.
: BGP isn't better. You don't need BGP for multihoming. My SF NANOG talk : was on exactly this topic. If you get a little bit of PA space from each : provider, and run a multihomed server (or a whole pile of them) with the : "interface default" mods we made available, you don't need BGP at all. You are convinced that ifdefault is production quality? Your slides don't give me the feeling of absolute confidence. Is anyone using it in a production environment?
From an operational standpoint BGP is what works, now.
Also, your solution only works on servers running BSD/OS 2.1. What about other platforms where the source code is not available (How am I going to magically convert my SGI/NT boxes into BSD/OS 2.1?) What about other clients from our end who still need redundant internet connectivity? DNS round robin is also a crude way to balance loads over network interfaces compared to path selection. I am extremely skeptical that ifdefault is ready for prime time. -Tung-Hui Hu hhui@arcfour.com
At the time of the SF NANOG, it was running in production. That customer has switched providers for various unrelated reasons and is no longer multihomed. But it did run a 100Khit/day web server for about three months without trouble.
I have received reports that this code ported trivially to NetBSD and to FreeBSD. (In fact it was originally developed on NetBSD.) I don't know how to get code patches made to SGI or NT, but one assumes that you'd start by telling your sales rep that you would buy N more boxes next year if they add this feature. But that assumes you have a strong strategic relationship with your system vendor. If you don't, then you're right that you're not a candidate for this. On the other hand you could probably put up a FreeBSD or NetBSD or BSD/OS box and try this out, and maybe switch system vendors.
DNS round robin is also a crude way to balance loads over network interfaces compared to path selection.
DNS round robin was never intended to solve the www.FOO.com problem.
I am extremely skeptical that ifdefault is ready for prime time.
You are welcome to that view, ignorant and reality-detached though it is. I think that if you actually try it out, and then report back, I will listen with a lot more interest to your comments as to its readiness for prime time.
Of course, the downsides of using the interface-default hack are: 1) it does not guarantee shortest path for the packets (unless someone has hacked together an lbnamed version that talks to gated and sees which interface has a shorter path to customer <x> based on number of AS hops before it answers the DNS query). 2) It uses a separate address for each interface (not important for a single box, but a room full of boxes, say, 50 of them, 3-way homed at a single site... hmm, that's 100 extra addresses you didn't want to use). I suspect that upstream providers will not be thrilled to hand out more address space if they discover it is being put to such inefficient use. 3) I have not looked at the code, but if it is on a per-interface basis, based on the addresses in the packets, that would seem to suggest that it might not like BSDI 3.0's virtual host scheme (adding IP addresses to the loopback port and then proxy-arping them onto the wire). If this is correct, that would mean you would have to use a different physical machine for each customer. Of course, on this point I'm purely speculating. 4) It puts the onus for fail-over on the DNS server, which means one is going to be using very short TTL. 5) Unless (#1), (#4) implies that fail-over will be manual. Is your Emacs ready to rock and roll on 50 zone files? I admire Paul's hack; it is spiffy for what it is, but I would hardly promulgate it as an advised way to multihome without using BGP. ---Rob
On Tue, 10 Jun 1997, Robert E. Seastrom wrote:
4) It puts the onus for fail-over on the DNS server, which means one is going to be using very short TTL.
Suppose you run two nameds. One with an addr on provider A, and one with an addr on provider B. Each one responds with only the address from its respective provider. Suppose A goes down. Then (most) new folks won't be able to reach your A name server and so won't even be trying your A address. (You have to assume here that your site isn't popular enough to be in most ISPs caches already... because if it were that popular you probably can get PI space anyhow.) This violates the rules of DNS however. Those nameservers can't claim to be authoritative and yet yield different answers. (And Paul is probably disappointed I'm suggesting it!) But going to this extreme is probably not even necessary any longer. MSIE 3 will skip past dead addresses and try all A records. Netscape communicator is supposed to do that (I haven't tested). And Squid will try all addresses. That leaves navigator 3.0. Give it six months. Dean
It seems to me that Ron himself can order all these things over the pay phone. I'll just have Kimba pick-up the stamps when they are ready. --david
oops Thats what you get for typing without your glasses on. ;) --david
At 10:50 AM 6/8/97 -0700, Michael Dillon wrote:
I enjoyed my first NANOG (10) and learned a great deal from several people. It was a pleasure to meet Michael, Justin, Robert of priori.net and several others, including Kim Hubbard of InterNIC and I am glad to see this issue on the NANOG list. Often, it is helpful to look at how other industries handle similar issues; as a result of a suggestion by Wayne Shirley, chairman of the New Mexico Public Utilities Commission, I looked into some federal documents for guidance on the issue and found the following at <http://www.ece.iit.edu/~power/taccess/rm95-8.000>:
The final codification may be FERC Order 888 or 18 CFR Part 388, which apparently are not online save West Law. What do NANOG people think about this issue? Should InterNIC grant small ISPs (this one serves a rural area between Dallas and Oklahoma City) fully routable and portable IP space? Or should the denial of instruments necessary for competition be used to force market consolidation? Your thoughts are appreciated. Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
Should InterNIC grant small ISPs (this one serves a rural area between Dallas and Oklahoma City) fully routable and portable IP space?
You are talking about a scarce resource. The InterNIC should allocate as much space as can be justified for real use. We discussed this to death a long time ago. randy
At 06:57 PM 6/8/97 PDT, Randy Bush wrote:
Has the situation changed since a long time ago? What scare resources, if any, should a rural ISP serving a rural area of say, 250K people, be allocated? Should the rural ISP be allowed the scarce resources to multi-home with portable IP space and thus be allowed to offer diversity and redundancy to customers? Your thoughts and those of others are appreciated. Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
And besides, if it weren't for US governement interference, ARIN would already be up and running, and the members could find solutions for this problem. But regardless, the allocation group does not controll the policies of the individual companies, and therefore, if you insist on causing grief for the rest of us, your best bet to to file Anti-trust actions against carriers, like Sprint, Digex, and others that are filtering. (hint, good luck, you will need it.) If ISPs were to push their local federal congress people, and get ARIN going, and join ARIN, perhaps this issue could be solved. Besides if 192/8 was given out in /21's etc, large companies would filter those addresses based on prefix length also. (Assuming enough were recycled.) In the world of competative access there are real costs assumed by those that want to interconnect, in the eletric world, there is a substantial cost of equipment. In the CLEC/CAP world there are filings, build out requirements, etc that are highly non-trivial. In the IP world there is the cost of renumbering until you can justify enough space. Larry, if you want to discusse this issue at more length, I would be happy to move it off-line to email or the TISPA list. Nanog is not interested in rehashing this issue again, anymore than the discussion of the sale of intergers. In message <m0wati8-0007zZC@rip.psg.com>, Randy Bush writes:
--- Jeremy Porter, Freeside Communications, Inc. jerry@fc.net PO BOX 80315 Austin, Tx 78708 | 1-800-968-8750 | 512-458-9810 http://www.fc.net
At 09:39 PM 6/8/97 -0500, Jeremy Porter wrote:
I have no intent of causing anyone grief, but rather wish to learn from the NANOG members (save any dogmas) and compete on a level playing field. Our only legal actions have been defensive in nature. Number 1 was against telcos red-lining PRI service to rural Texas. Number 2 regarded an email privacy/ECPA issue raised by the Texas Attorney General through the use of an improperly issued subpoena. We had good luck in both, thank you [8-)) The input of those from NANOG is very valuable, IMHO. I'm looking for the thoughts of NANOG members, not to convince them of anything. Please excuse my thoughts that the NANOG members may have more experience with these issues and that we and other small ISPs can benefit from NANOG input. In fact, the real question we posed at NANOG 10 is: What can be done to properly operate a network with diversity and redundancy without the use of scarce resources, presuming the downside of them not being granted? Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
Use DHCP, get your customers to use DHCP. Current Best Current Pratices for IP allocations (http://ftp.fc.net/rfc/rfc2050.txt) states: 5. Due to the requirement to increase the utilization efficiency of IPv4 address space, all assignments are made with the assumption that sites make use of variable length subnet mask (VLSM) and classless technologies within their network. Any request for address space based on the use of classfull assumptions will require a detailed justification. The use of classfull technologies for the purposes of administrative convenience is generally insupportable due to the limited availability of free IPv4 address space. ... 7. While it is understood that the use of static addressing may ease some aspects of administration, the current rate of consumption of the remaining unassigned IPv4 address space does not permit the assignment of addresses for administrative ease. Organizations considering the use of static IP address assignment are expected to investigate and implement dynamic assignment technologies whenever possible. This means that you should use renumbering technology as much as possible, so that when you can justify a /19 you can get one, until then, find an ISP that is multi-homes and can provide you the IP addresses you need, and consider two connections to them, giving you the benefit of being multi-homed without the cost of the routers needed for full BGP, or needing to justify a /19. In message <3.0.1.32.19970608221802.00a56078@texoma.net>, Larry Vaden writes:
--- Jeremy Porter, Freeside Communications, Inc. jerry@fc.net PO BOX 80315 Austin, Tx 78708 | 1-800-968-8750 | 512-458-9810 http://www.fc.net
This is all great and dandy, but then why does it appear that anybody with a cable modem this side of the sun are using static IP's. Granted that the Nic probably didn't allocate the current /8 (or the next one), but I don't see any (and didn't see any prior) 'investigation' to make dynamic allocation possible (or using RFC1918 addresses). Are they looking at DHCP or RFC1918 as a solution for their userbase ? While we're here pulling our teeth on returning addresses and not allowing static IP's for a combined user base of several million internet users, a group of users with less than 50K (correct me if I'm wrong here) and no track history of rwhois/swipped addresses gets gets /8's for allocation to a static IP userbase....... Ed On Sun, 8 Jun 1997, Jeremy Porter wrote: : :Use DHCP, get your customers to use DHCP. Current Best Current Pratices :for IP allocations (http://ftp.fc.net/rfc/rfc2050.txt) states: : 5. Due to the requirement to increase the utilization efficiency : of IPv4 address space, all assignments are made with the : assumption that sites make use of variable length subnet mask : (VLSM) and classless technologies within their network. Any : request for address space based on the use of classfull : assumptions will require a detailed justification. The use of : classfull technologies for the purposes of administrative : convenience is generally insupportable due to the limited : availability of free IPv4 address space. :... : 7. : While it is understood that the use of static addressing may : ease some aspects of administration, the current rate of : consumption of the remaining unassigned IPv4 address space does : not permit the assignment of addresses for administrative ease. : Organizations considering the use of static IP address assignment : are expected to investigate and implement dynamic assignment : technologies whenever possible. : : :This means that you should use renumbering technology as much as :possible, so that when you can justify a /19 you can get one, :until then, find an ISP that is multi-homes and can provide you :the IP addresses you need, and consider two connections to them, :giving you the benefit of being multi-homed without the cost of :the routers needed for full BGP, or needing to justify a /19. : :
On Mon, 9 Jun 1997, Edward Fang wrote:
We're doing cablemodems out of RFC1918 address space using PIXes in several communities and it hasn't been fun. Many of the latest-n-greatest network apps (games, video, voice, what have you) are broken by NAT. They seem to like to transmit the client's address at the application layer. This of course doesn't work, since the client's address is 10.x.x.x... You can dismiss this problem by saying the apps are broken (which they are), but the simple fact is our customers want to use these apps. I'd recommend DHCP. In communities where we've used it, it has worked fine and not caused any of the problems that NAT does. Regards, Joel Gallun Community Networks
At 11:27 AM 6/9/97 -0400, Joel Gallun wrote:
Thanks for your comments. What sort of DHCP lease times do you use? What kind of ratios between customers and IP addresses work in your situation? In other words, I'm looking for how much more IP space, if any, you need versus the dynamic IP assignment we're used to seeing in dial up ppp service. Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
A long time ago, a group asked the same question. The answer turned out to be that the Internic, not being an ISP, has no clue about the routability of -any- prefix that is delegated. No delegation registry can ensure the routability of any given prefix. Thngs may have changed, can you describe to me what you consider a "fully routable and portable IP space" might look like and how such conditions might be enforced? -- --bill
At 07:13 PM 6/8/97 -0700, Bill Manning wrote:
Bill, I feel you could provide a better definition, but what others tell me is that CIDR blocks with prefixes longer than /19 are fully routable if they are in the 192/8--205/8 range (a regurgitation of Sprint el al's routing policy?). What is your definition of "fully routable"? What is your definition of "portable"? What is the role of "legacy" equipment, if any, in such matters? Are the larger players concerned that the smaller players don't have "certified" BGP configurations? I appreciate your input and that of others as well. Regards, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
I appreciate your interest and consideration. Your pride in using the legal system to "enforce" your will leads me to caution. I suggest that your best bet is to review the archives of the various mailing lists regarding the discussions regarding address space delegations and the rational for the criteria now in place. I will note that in general, the only consistant thing that remains consistant is change. Attempting to set legal precedence for a system that moves with as much rapidity as the Internet may be doomed. A much better tact is to consider the various players in the field as partners and friends, not combatents and enemies. I see a clear precident for sharing and cooperation in helping entities grow to their fullest potential, while not squandering precious resource. If you really need it, it is available. Change the model and this may not be true anymore.... --bill
At 09:21 PM 6/8/97 -0700, Bill Manning wrote:
partners
Bill, thank you I think [8-)) What do you understand to be the immediate utilization percentage for an ISP's first /19? I am truly and genuinely interested in your answer. Thanks, Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
Well, Let me first say that this discussion likely belongs on PAGAN, and not, NANOG, so why don't we move it over there? Secondly, its good to have you back on the list Bill ;). As a discussion starter, I believe that times have changed since RFC 2050 was written, and the document needs revision. At 09:21 PM 6/8/97 -0700, Bill Manning wrote:
partners
Justin W. Newton Senior Network Architect Priori Networks http://www.priori.net ISP/C, Director at Large http://www.ispc.org
On Sun, 8 Jun 1997, Bill Manning wrote:
Hmmm...the InterNIC appears to have a definition for "fully routable". Maybe we should ask them? -=- BEGIN EXCERPT -=- Date: Mon, 2 Jun 1997 11:04:14 -0400 (EDT) From: Network Registration Role Account <netreg@internic.net> [...] Sorry, Sprint isn't the only one with filters. You don't justify a /19. Less than that would not be fully routable ^^^^^^^^^^^^^^ [emphasis added] and we won't issue it. [...] -=- END EXCERPT -=- pbd -- You can make it illegal, but you can't make it unpopular.
At 07:13 PM 6/8/97 -0700, Bill Manning wrote:
I think a better question to ask is whether Sprint and others who are filtering out routes, forcing the issue of getting /19 blocks or not getting portable space at all, are being Anti-competitive in the sense of the Sherman Anti-trust act. By their actions the large network providers are essentially limiting the ability of many small to mid-sized ISP businesses from being able to compete in offering high-reliability (i.e. multihomed) services to their customers. This could be seen as an effort to limit the number of ISP vendors to the big, national folks, and those who have existed long enough to have their own resources (independent address space). Folks have said the InterNIC, not being an ISP, is not in a position to understand this issue, and that makes sense. The large ISPs are concerned about the size of routing tables in their routers, and that is certainly a reason to be concerned about having smaller independent address blocks. The problem of large routing tables can be solved by making routers that support more memory and/or routers that use their memory more efficiently. Dan The thoughts expressed here are my own ramblings and do not in any way reflect any position, policy or opinion of my employer.
Indeed, a good question. Should you care to read the archives <gasp!>, you'll find that the answer is no. Folk are trying to keep core routers from falling over. And, to answer one of the next questions expected by students of history, no, they are not obliged to buy more or other (unreliable) hardware for their backbones so that you (or whomever) can be in business. The one sure thing on the net is that the newbie influx is sufficient to keep the majority of mailing list traffic repeating itself. randy
At 09:47 PM 6/8/97 PDT, Randy Bush wrote:
Folk are trying to keep core routers from falling over.
Probably like you, I've been in the Information Technology field for about 30 years this fall, but I've never seen fear nor protectionism used as a basis for policy in Information Technology before and it is certainly disconcerting to see it used this early in the life cycle of the Internet. Good press for creating a "scarcity" scenario, but bad practice IMHO. To repeat an earlier unanswered question, what and whose legacy hardware and software is causing the problem?
I certainly agree with part of your statement, but don't feel the incumbents should be able to prevent new firms from competing through unfair practices. One would normally prefer to allow market forces to set the rewards for those who won't upgrade.
The one sure thing on the net is that the newbie influx is sufficient to keep the majority of mailing list traffic repeating itself.
Last year's answers to last year's problems are not necessarily the best answers to this year's problems given the current high rate of change. Larry Vaden, founder and CEO help-desk 903-813-4500 Internet Texoma, Inc. <http://www.texoma.net> direct 903-870-0365 bringing the real Internet to rural Texomaland fax 903-868-8551 Member ISP/C, TISPA and USIPA pager 903-867-6571
The of the routing table getting larger will never be fixed by "build the bigger machine" theory. It doesn't matter if you have a router that has 1 Gig of memory and has the fastest processor in it.. the time it takes to parse the memory for a route is always going to be the limiting factor. Aggregation is a good start, that is why Sprint is filtering and people like ANS are implementing policy routing. The problem is that the IP space was never designed to be used like it is now. One of the biggest problem with people wasting address space is the fact that a lot of people out there don't know how to subnet. Eric _______________________________________________________ Eric D. Madison - Senior Network Engineer - ACSI - Advanced Data Services - ATM/IP Backbone Group 24 Hour NMC/NOC (800)291-7889 Email: noc@acsi.net On Mon, 9 Jun 1997, Daniel Senie wrote:
On Wed, 4 Jun 1997, Eric Germann wrote:
Renumber a lot. It is hard for the Internic to hand out /19 when most of the new providers will be out of business in 6 months. The only way to do it is to get space from your ISP and then renumber if you move. After you do that a few times if you do it will, you will be able to get a /19 and then /18 and so on. Nathan Stratton President, NetRail,Inc. ------------------------------------------------------------------------ Phone (888)NetRail NetRail, Inc. Fax (404)522-1939 230 Peachtree Suite 500 WWW http://www.netrail.net/ Atlanta, GA 30303 ------------------------------------------------------------------------ "No king is saved by the size of his army; no warrior escapes by his great strength. - Psalm 33:16
Hi, Nathan Stratton wrote:
The tools for keeping IP independence should also be more mature and widely deployed... like DHCP for instance, and IP aliasing, which isn't available on older Unices. -- miguel a.l. paraz <map@iphil.net> +63-2-893-0850 iphil communications, makati city, philippines <http://www.iphil.net>
participants (24)
-
bmanning＠ISI.EDU -
Bradley Dunn -
Daniel Senie -
David Holub -
Dean Gaudet -
Edward Fang -
Eric D. Madison -
Eric Germann -
George Herbert -
Jeremy Porter -
Joel Gallun -
Justin W. Newton -
Larry Vaden -
Matthew Petach -
Michael Dillon -
Miguel A.L. Paraz -
Nathan Stratton -
Paul A Vixie -
Phil Howard -
randy＠psg.com -
Robert E. Seastrom -
Stephen A Misel -
Suzanne Woolf -
Tung-Hui Hu