Yo John! My local cable company, Bend Cable, gives out static IP with RoadRunner service. I also know several folks on @Home in Fremont, CA, that have static IP addresses. RGDS GARY --------------------------------------------------------------------------- Gary E. Miller Rellim 20340 Empire Ave, Suite E-3, Bend, OR 97701 gem@rellim.com Tel:+1(541)383-2435 On 7 Nov 1998 johnl@iecc.com wrote:

...

...

Cable modem subscribers normally are not dynamically assigned IPs like dialups. Besides, they are advertised as a constant connection, and a lot of people use it as such.

All of the cable modem subscribers I know have dynamic IPs assigned by DHCP. A suitably aggressive DHCP client can usually keep the lease renewed so the IP doesn't change, but they are most certainly not static IP addresses.

RGDS GARY --------------------------------------------------------------------------- Gary E. Miller Rellim 2680 Bayshore Pkwy, #202 Mountain View, CA 94043-1009 gem@rellim.com Tel:+1(650)964-1186 Fax:+1(650)964-1176

At 11:36 AM 11/7/98 -0600, Phil Howard wrote:

Gary E. Miller wrote:

...

My local cable company, Bend Cable, gives out static IP with RoadRunner service.

I also know several folks on @Home in Fremont, CA, that have static IP addresses.

Our dedicated accounts get static, as I am sure is the case with most ISPs. Non-dedicated can get static only for a fee. The rate structure starts at $8 for a /32 and goes up to about 25% above the ARIN fee for address space. The customer must also assert why they need it.

I just wish I could swip the /30's, /31's, and /32's we give out, to ARIN. We're putting in more NAT and proxy boxes for businesses and are doing more networks smaller than /29 these days. Yet ARIN still encourages us to use a /29 when a /32 would do. I'll probably swip the /29 containing them as just a commentary saying something like "contains /32 assignments" when I get my network database going.

What this means is that ARIN is restricting the DNS management. If you have less than a /29 then you are not allowed to manage your own domain-space without a handler. That handler is the ISP. To be honest, most of our /29's are too clueless to handle DNS, in fact most of them are Microsoft-only LANs whom I don't trust DNS to anyway. I have yet to see a BIND port to any version of Windows, with resolver library, that worked. Gotta have a Unix box in the LAN somewhere in order to do that ( and a bunch else besides). Most of our /29's cann't even spell Unix. I take that back, I was just informed that we have some Apple-talk and Novell LANs as well. ___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer@mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

At 11:23 PM 11/8/98 -0500, Jon Lewis wrote:

On Sat, 7 Nov 1998, Roeland M.J. Meyer wrote:

...

What this means is that ARIN is restricting the DNS management. If you have less than a /29 then you are not allowed to manage your own domain-space without a handler. That handler is the ISP. To be honest, most of our /29's are too clueless to handle DNS, in fact most of them are Microsoft-only

Just the /29 users? DNS is irrelevant. ARIN only watches over IPv4 address space utilization. OK...they also handle in-addr.arpa delegations for the space they look after, but the fact that they don't want swips of /29 or longer prefixes has nothing to do with who can manage DNS.

If you are not handling your own in-addr.arpa then you are not fully managing your own DNS name space. In *many* cases, if the reverse does not match the forward reference then access will be denied. ___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer@mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

At 01:43 PM 11/9/98 -0500, you wrote:

There should also be no objection for listing all assigned /30 networks. I'm not talking about dial-up users -- but fully routed dedicated customers using only a /30 or more. I *WANT* to see these networks in whois lookups! These listings should be *required*.

I second this. One thing I would dearly love to see is support in client rwhois implementations for referrals. Today, if I query the root rwhois server at rwhois.net, I don't end up with a clear answer as to who the network block has been delegated. I have to manually try and find the rwhois server [if any] for the smallest block returned, and see if there are further delegations being made. At Vitts, we're attempting to document in our rwhois server, every network assignment we make. This includes the /30's. We have a handful of 1 IP customers that are connected to the Internet using a router running NAT or firewall software. It makes it easier for us to have a standards based way of identifying the owner for every IP address we manage. Dan -- Dan Watts Vitts Networks dwatts@vitts.com

Gary E. Miller wrote:

...

My local cable company, Bend Cable, gives out static IP with RoadRunner service.

I also know several folks on @Home in Fremont, CA, that have static IP addresses.

Our dedicated accounts get static, as I am sure is the case with most ISPs. Non-dedicated can get static only for a fee. The rate structure starts at $8 for a /32 and goes up to about 25% above the ARIN fee for address space. The customer must also assert why they need it.

I just wish I could swip the /30's, /31's, and /32's we give out, to ARIN. We're putting in more NAT and proxy boxes for businesses and are doing more networks smaller than /29 these days. Yet ARIN still encourages us to use a /29 when a /32 would do. I'll probably swip the /29 containing them as just a commentary saying something like "contains /32 assignments" when I get my network database going.

By asking you not to SWIP assignments longer than /29 ARIN is not encouraging you to not issue them, we just don't think it's necessary to have every single dialup user listed in WHOIS. Kim

-- -- *-----------------------------* Phil Howard KA9WGN * -- -- | Inturnet, Inc. | Director of Internet Services | -- -- | Business Internet Solutions | eng at intur.net | -- -- *-----------------------------* philh at intur.net * --

At 11:37 PM 11/8/98 -0500, Steven J. Sobol wrote:

On Sun, Nov 08, 1998 at 07:01:13PM -0500, Kim Hubbard wrote:

...

By asking you not to SWIP assignments longer than /29 ARIN is not encouraging you to not issue them, we just don't think it's necessary to have every single dialup user listed in WHOIS.

I have to agree with Kim. That's why I shuddered when I thought of every single static IP address at every ISP being SWIP'd. It would increase ARIN's database exponentially and is not necessary.

So it goes from a few hundred MB's to a few GBs? What's the big deal. Given a trivial CustomerDB, in an RDBMS, it's still under 100 GB for a few million IPs. 150 GB of RAID5 is still less than $30KUS and dropping daily, even on HP (High Priced <grin>) equipment. ___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer@mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

At 08:05 PM 11/9/98 -0500, Steven J. Sobol wrote:

On Mon, Nov 09, 1998 at 10:27:35AM -0800, Roeland M.J. Meyer wrote:

...

So it goes from a few hundred MB's to a few GBs? What's the big deal. Given a trivial CustomerDB, in an RDBMS, it's still under 100 GB for a few million IPs. 150 GB of RAID5 is still less than $30KUS and dropping daily, even on HP (High Priced <grin>) equipment.

It seemed to be a lot, to me. Of course, if you had every single IPv4 address SWIP'd, that'd be 4,294,967,296 records multiplied by the size of the record, but not every IP is going to be SWIP'd.

Please bear with me if this seems elementary to y'all. Lets look at this further. If we assume that only domain-name holders will manage their DNS entries and make them manage their individual users and hosts, the there'll be at least two hosts per domain (one server and one workstation). For a clean sub-net (/31), this burns four IP addresses (assuming a 1:1 domain/sub-net relation) for two hosts. That's also one busy server because there's no room for a router or anything else. It's not exactly efficient either. Let's skip the intermediate example and go straight to a /29, because a /30 isn't much better (6 ip addresses and four hosts, not much room there either). Now we have room for a router, file-server, print-server, web-server, and four workstations. This is a small office. I actually see the sense in ARIN not SWIPing smaller than a /29. This is the smallest unit that makes sense. Arranged differently, using NAT, one could have up to 7 workstations and one production server in the Internet and any number of servers/routers in the intra-net (I actually don't see need for more than 30 of these. That would be a very data-intensive company). This should cover most small businesses. The trick is in finding out how many servers and workstations have to have internet visibility and how many are only there for infrastructure (intra-net). At MHSC, we assume that all workstations need internet access (browsing and internet services), but only a small percentage of servers actually have to have it (serving internet services and corporate connectivity). What this means for DNS is that only such an office will need to manage their own DNS entries and that also happens to be the smallest unit that ARIN will SWIP (we also begin to understand why this is so, although ARIN should include some of this reasoning in thier web-site, or on the InterNIC, so that Joe User will understand this too). Now we get to virtual hosts. The ISP carries these on their DNS servers and on their web-sites. All of these resolve to the same apache server cluster. No addreses need be SWIPed unless the domain-name holder also has machines at their physical location (in which case, see above). These folks don't "manage" DNS at all because they have no real hosts, they are a virtual domain in every sense of th term. Their ISP does all the work for them. Their workstations are probably dynamicly assigned connections from their ISP as well. The point here is that, since they don't get SWIPed, they don't need a database entry. We've now cut down our worst-case scenario, by a factor of eight, to 536,870,912 sub-nets. Even if we assume 2048B per RDBMS entry that comes out to 1,099,511,627,776 (1.1TB). Hmmm, that's a bit out-there, but do-able for Oracle. The disk system (15 cascaded RAID5 sub-systems at 100GB each) should cost about $150KUS at today's prices. However, notice that this is a worst-case maximum use scenario. We are nowhere near there yet. Kim should have a better idea, but the real numbers should be one quarter of that, present utilization, allowing for pre-assigned ip-blocks, and the swamp. Notice something else, I just talked my-self out of SWIPing anything less than a /29 and backed into the probable reason ARIN doesn't SWIP anything less that a /29. Kim; this is NOT an obvious line of reasoning, but is probably what your analysts went through. You'd catch a LOT less flack if y'all published these things.

I don't know what it would require in terms of human processing time, if any.

No man-power required, it's all automatable. ___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer@mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

...

Lets look at this further. If we assume that only domain-name holders will manage their DNS entries and make them manage their individual users and hosts, the there'll be at least two hosts per domain (one server and one workstation). For a clean sub-net (/31), this burns four IP addresses (assuming a 1:1 domain/sub-net relation) for two hosts. That's also one busy server because there's no room for a router or anything else. It's not exactly efficient either.

Some minor technicalities: /32 Single host /31 Not feasible. No host addresses, network and broadcast only. /30 Network, two hosts, broadcast. /29 Network, 6 hosts, broadcast.

If the number of /30 and /32 SWIPs to ARIN would overwhelm their database, then I would refer to that as the very justification, based on the numbers, to do so. If there are that many /30 and /32 assignments, then they do need to be considered when ARIN is evaluating new allocations of network space. They need to be considered not just for the volume of usage, but also for the effort to keep the space usage at minimal and efficient levels. Those who do more /30's and /32's, IMHO, do justify more space when they do reach then end of what they have (which hopefully won't be as soon).

I think this misses the point. ARIN doesn't require or want you to SWIP your /30 and /32 allocations. A network that small just doesn't require that level of public contact visibility. As you've pointed out, you'll be doing most of the things that matter (from a contact perspective) for those customers. As such, it makes sense to use your larger block contact information instead of SWIPing such small networks. In fact, I'd rather see ARIN move the SWIP requirement back to /26 or so.

What if an ISP dealt exclusively in just customers that wanted these server based connections? Would an ISP with 2000 /30's be able to come to ARIN and ask for more space because their /19 was nearly full?

Yes. You simply have to document the number of /30's you've handed out in your application. They've been quite willing to accept us allocating multiple clasee C's to /30's just for point-to-point links.

Perhaps a better solution is a distributed SWIP database. Perhaps a SWIP record can be added to DNS to indicate the name of the server to query for detail information by network. It would then co-exist along side the PTR records. Appropriate SWIP lookup clients would resolve for SWIP data in the in-addr.arpa space and then query that server (or servers) for the real data about the network in question.

Although DBMS scaleability is an issue, it's not the driving one in my desire to not SWIP small blocks. In my opinion, the value of the information contained in the SWIPs is minimal at best.

IPv6 will probably force the issue, anyway, since it opens the door to what will at least today be a nearly infinite address space. Try putting that on few terabytes of Oracle.

We probably are pretty unlikely to fill the IPv6 prefix space. Remember, the prefix space is 8 octets. Owen

We are seeing a lot of the same. /32 certainly does not mean small network. Dirk On Tue, Nov 10, 1998 at 03:35:13PM -0500, Jeff Mcadams wrote:

Thus spake Owen DeLong

...

I think this misses the point. ARIN doesn't require or want you to SWIP your /30 and /32 allocations. A network that small just doesn't require that level of public contact visibility.

I think you missed his point though....with NAT/PAT technology.../30 and /32's from ISP's can indeed provide a whole corporate network with access (small corporate...not exactly Fortune 500 here, but you get the idea)...I second his point on this. We've got quite a few customers that are feeding whole networks with /32's...even providing web servers and mail servers via these NAT/PAT boxes that are available now. Just stating that the network only has one or two Internet available IP addresses and therefore its too small to be of significance is short-sighted at best. Many of these /32's for us have their own web administration, mail administration, and other local administration of many of their services. They use a single IP as almost an inherent firewall. Indeed, I have one customer that uses one of the NAT/PAT boxes to actually not have IP on their internal network at *ALL*. The box converts the TCP/IP to IPX/SPX...bizarre, but it works well for them. Anyway, they run their own mail server on this setup, and we do very little administrative functioning for them...DNS is it in this case.

...

As you've pointed out, you'll be doing most of the things that matter (from a contact perspective) for those customers. As such, it makes sense to use your larger block contact information instead of SWIPing such small networks. In fact, I'd rather see ARIN move the SWIP requirement back to /26 or so.

Put my vote in for allowing up to /32's. -- Jeff McAdams Email: jeffm@iglou.com Head Network Administrator Voice: (502) 966-3848 IgLou Internet Services (800) 436-4456

On Thu, 12 Nov 1998, Mike Pistone wrote:

Although I am almost NEVER one to recommend a Microsoft product BUT MS Proxy server is actually a very nice product.

Too bad it doesnt do socks5. And it doesnt do transparent proxying either. If you have anything other than standard windows (eg any flavor of unix), its a very very bad solution. -Dan

I think this misses the point. ARIN doesn't require or want you to SWIP your /30 and /32 allocations. A network that small just doesn't require that level of public contact visibility. As you've pointed out, you'll be doing most of the things that matter (from a contact perspective) for those customers. As such, it makes sense to use your larger block contact information instead of SWIPing such small networks. In fact, I'd rather see ARIN move the SWIP requirement back to /26 or so.

There are two purposes for SWIP. One is to provide contact information. And for that purpose, SWIPping /30 and /32, and many of our larger assignments, does not make much sense. The other purpose, however, is used to justify address space allocation. The majority of my larger networks are really no different than the /30's in terms of wanting to contact someone who really knows what is going on. The persons listed in the SWIPs for most of our networks have no idea what SWIP is, what ARIN is, or why you'd even bother to email or call them. If the only purpose of SWIP is to just provide a point of contact, then I really don't have very many SWIP templates to send in, if any at all.

...

What if an ISP dealt exclusively in just customers that wanted these server based connections? Would an ISP with 2000 /30's be able to come to ARIN and ask for more space because their /19 was nearly full?

Yes. You simply have to document the number of /30's you've handed out in your application. They've been quite willing to accept us allocating multiple clasee C's to /30's just for point-to-point links.

Of course some may argue that this is bad, and I acknowledge some issues, but in balance I find the advantages outway the disadvantages, so I have set up all my links that don't specifically need to reach the Internet in 172.30.0.0/16. When I do hand out real /30's, what I have done so far is send in an "aggregate SWIP" simply stating that the space is for /30's. At least something is on record for it. And this may be totally satisfactory.

Although DBMS scaleability is an issue, it's not the driving one in my desire to not SWIP small blocks. In my opinion, the value of the information contained in the SWIPs is minimal at best.

But I'm not trying to address the information availability issue. Instead, I am trying to address the network space justification issue. The only reason I do not my my own name as contact for the SWIPs for all my customers is so that they look like valid SWIPs that ARIN or my upstream can validate and verify the usage with. They can contact the customer listed and see if they really do exist and really are a customer of mine to be sure I am not just padding my network (and I am sure that practice goes on, so I would encourage such verification, at least on a spot-random basis). So how would you feel if the requirement to send in SWIP data was lifted so that only those networks allocated for resale, or otherwise have a specific and different point of contact, need to be SWIPped? Keep in mind this would mean that I would probably have about 4 or 5 of them, total.

We probably are pretty unlikely to fill the IPv6 prefix space. Remember, the prefix space is 8 octets.

And one of the needs for doing SWIP goes away at that point. ARIN will still have work to do as any space, no matter how large, still has to be managed. It will be a lot easier. It surely won't be any big deal to allocate a /96 to me in IPv6 space. Routing may become very interesting, though. -- -- *-----------------------------* Phil Howard KA9WGN * -- -- | Inturnet, Inc. | Director of Internet Services | -- -- | Business Internet Solutions | eng at intur.net | -- -- *-----------------------------* philh at intur.net * --

I ended up finding a good way of traking internal use rather unintentinally. Our internal IP database shoots off a SWIP everytime you make an allocation or an assignment; filling in all the proper fields. To differentiate types of internal networks I make separate customers, say "ELI - Dialup", "ELI - Backbone", etc. These customers have a flag set that means they don't do resale, so they are end assignments. A swip gets generated, but I can also use our database to generate the appropriate sections for the http://www.arin.net/templates/isptemplate.txt form. Generating these by hand would be quite a lot of work, but verifying an already created list is much easier. --Ben Kirkpatrick On Tue, 10 Nov 1998, Steven J. Sobol wrote: )On Tue, Nov 10, 1998 at 03:36:53PM -0600, Phil Howard wrote: ) )NACS is going to need a new /19 pretty soon. But we have 13 Class C's )used for things like dialups, servers, web virtual hosts, etc. So how can )we explain our need for more addresses when much of our current /19 may appear )to be unused, since it isn't SWIP'd? ) )> When I do hand out real /30's, what I have done so far is send in an "aggregate )> SWIP" simply stating that the space is for /30's. At least something is on )> record for it. And this may be totally satisfactory. ) )How do we record this for /24's which we use for internal purposes? ) )> So how would you feel if the requirement to send in SWIP data was lifted so )> that only those networks allocated for resale, or otherwise have a specific )> and different point of contact, need to be SWIPped? Keep in mind this would )> mean that I would probably have about 4 or 5 of them, total. ) )This is one possible solution.

Phil et al. Perhaps I'm off base here, but I thought this was the concept of using shared whois. "The first step in this process is to verify the utilization of at least 80% of the prior allocations. The methods available to show this utilization are SWIP and RWHOIS for ISPs who have assigned addresses to their clients." If you have a FreeBSD system you can install rwhois from the ports collection. See below for example output. Phil Howard wrote:

Roeland M.J. Meyer wrote:

.. trimmed to save bits.

Perhaps a better solution is a distributed SWIP database. Perhaps a SWIP record can be added to DNS to indicate the name of the server to query for detail information by network. It would then co-exist along side the PTR records. Appropriate SWIP lookup clients would resolve for SWIP data in the in-addr.arpa space and then query that server (or servers) for the real data about the network in question.

IPv6 will probably force the issue, anyway, since it opens the door to what will at least today be a nearly infinite address space. Try putting that on few terabytes of Oracle.

Example random query: monet# rwhois 207.10.12.0 Referral Whois (RWhois) spec version 1.0 (InterNIC V-1.0B5) Connecting to [root.rwhois.net] port [4321] -> client default server network:Class-Name:network network:Auth-Area:0.0.0.0/0 network:ID:NETBLK-SPRINT-CF0A0F.0.0.0.0/0 network:Handle:NETBLK-SPRINT-CF0A0F network:Network-Name:SPRINT-CF0A0F network:IP-Network:207.10.8.0/21 network:IP-Network-Block:207.10.8.0 - 207.10.15.0 network:Org-Name:Nysernet/Monroe County Boces #1 network:Street-Address:41 O'Connor Rd. network:City:Fairport network:State:NY network:Postal-Code:14450-1327 network:Country-Code:US network:Tech-Contact;I:JP29.edu network:Created:19960116050000000 network:Updated:19960116153537000 network:Class-Name:network network:Auth-Area:0.0.0.0/0 network:ID:NET-SPRINT-CF0A00.0.0.0.0/0 network:Handle:NET-SPRINT-CF0A00 network:Network-Name:SPRINT-CF0A00 network:IP-Network:207.10.0.0/16 network:IP-Network-Block:207.10.0.0 network:Org-Name:Stellar Communications,Inc network:Street-Address:460 State St network:City:Rochester network:State:NY network:Postal-Code:14514 network:Country-Code:US network:Tech-Contact;I:TS905.COM network:Created:19960122050000000 network:Updated:19960123012455000 network:Class-Name:network network:Auth-Area:0.0.0.0/0 network:ID:NETBLK-NETBLK-SPLK-NYNET3.0.0.0.0/0 network:Handle:NETBLK-NETBLK-SPLK-NYNET3 network:Network-Name:NETBLK-SPLK-NYNET3 network:IP-Network:207.10.0.0/16 network:In-Addr-Server;I:NS310-HST.NET network:In-Addr-Server;I:NS311-HST.NET network:In-Addr-Server;I:NS312-HST.NET network:IP-Network-Block:207.10.0.0 - 207.10.255.0 network:Org-Name:NYSERNet network:Street-Address:200 Elwood Drive network:City:Liverpool network:State:NY network:Postal-Code:13088-6147 network:Country-Code:US network:Tech-Contact;I:BG735.NET network:Created:19951212050000000 network:Updated:19960408160628000 network:Class-Name:network network:Auth-Area:0.0.0.0/0 network:ID:RESERVED-1.0.0.0.0/0 network:Handle:RESERVED-1 network:Network-Name:RESERVED network:IP-Network:0.0.0.0/0 network:IP-Network-Block:0.0.0.0 network:Org-Name:IANA network:Tech-Contact;I:JKR1.EDU network:Updated:19910115210052000 monet# -- From: Joseph T. Klein, Titania Corporation mailto:jtk@titania.net http://www.titania.net voice: +1 414 372 4565 FAX: +1 414 264 6038 "Some of the greatest companies couldn't get funded when they first started, so don't let the bozos get you down: Keep on plugging." -- Guy Kawasaki of Garage.com

Notice something else, I just talked my-self out of SWIPing anything less than a /29 and backed into the probable reason ARIN doesn't SWIP anything less that a /29. Kim; this is NOT an obvious line of reasoning, but is probably what your analysts went through. You'd catch a LOT less flack if y'all published these things.

Thanks Roeland, I'll keep that in mind. Kim

...

I don't know what it would require in terms of human processing time, if any.

No man-power required, it's all automatable.

___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer@mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

At 10:49 AM 11/9/98 -0800, you wrote:

On Sun, 8 Nov 1998, Steven J. Sobol wrote:

...

I have to agree with Kim. That's why I shuddered when I thought of every single static IP address at every ISP being SWIP'd. It would increase ARIN's database exponentially and is not necessary.

And don't forget that you have to come up with a unique network name for the SWIP'd IP, every time you change the assignment.

The answer, of course, is rwhois since that lets you go to whatever level of detail you want, and use your own resources to do so. However, using rwhois is easy to talk about but...

Agreed. I think one thing that slows acceptance of rwhois is the lack of client tools used to query the databases. Most [all?] of the rwhois clients don't support referral information, so getting complete information on a query is difficult.

When you have to justify all your allocations, it is convenient if you can have all the allocations SWIPed and not have to worry about different recording and procedures for different sized blocks. You can get over that hurdle though, and once you are it isn't that difficult to deal with.

I found using rwhois was a lot easier to use than SWIP. I should admit though that my experience predates the online forms. I also found that keeping the information up to date with rwhois is very easy [we just change our database, re-export to the rwhois server, and we're done] and we don't have to wait for the SWIP submittal turnaround time. Dan -- Dan Watts Vitts Networks dwatts@vitts.com

On Sat, 7 Nov 1998, Phil Howard wrote:

I just wish I could swip the /30's, /31's, and /32's we give out, to ARIN.

Why? Assuming you assign those blocks out of a larger block (i.e. you give out /30's out of a /25), when it comes time to apply for more space, you just tell ARIN blah/25 is split up into /30's for dedicated customers using NAT. You probably won't get any argument beyond them possibly asking for a list of who the customers are. ---dont't waste your cpu, crack rc5...www.distributed.net team enzo--- Jon Lewis <jlewis@fdt.net> | Spammers will be winnuked or Network Administrator | nestea'd...whatever it takes Florida Digital Turnpike | to get the job done. ______http://inorganic5.fdt.net/~jlewis/pgp for PGP public key________

At 05:42 11/7/98 -0000, johnl@iecc.com wrote:

All of the cable modem subscribers I know have dynamic IPs assigned by DHCP. A suitably aggressive DHCP client can usually keep the lease renewed so the IP doesn't change, but they are most certainly not static IP addresses.

Yes, DHCP is used, but the IP addresses remain the same unless a network renumbering is necessary. ************************************************** Andrea Di Lecce (416) 935-5700 Rogers@Home Network Operations 1 Mount Pleasant Road, Toronto, M4Y 2Y5 **************************************************

On Sat, 7 Nov 1998, Kevin Oberman wrote:

What you don't seem to understand is that cable modem customers are connected 24x7. @Home does use DHCP, but the address almost never changes since DHCP is designed to keep assigning the same address to the same host as long as possible. Think of them as pseudo-static.

If this is the case, then all of the mid to large size ISPs should be able to justify a whole bunch of new space based on DSL customers... They are always connected so they can have a full time address too, whether through static or DHCP... Tim ---------------------------------------------------- Timothy M. Wolfe | Why surf when you can Sail? tim@clipper.net | Join Oregon's Premier Sr. Network Engineer | Wireless Internet Provider! ClipperNet Corporation | http://www.clipper.net/ ----------------------------------------------------