- Test - lists.nanog.org

RE: Important IPv6 Policy Issue -- Your Input Requested
by Hannigan, Martin 09 Nov '04

09 Nov '04

> -----Original Message----- > From: owner-nanog(a)merit.edu [mailto:owner-nanog@merit.edu] > Sent: Tuesday, November 09, 2004 12:14 AM > To: Daniel Senie > Cc: Randy Bush; kent crispin; nanog(a)merit.edu > Subject: Re: Important IPv6 Policy Issue -- Your Input Requested > > > > > On 8 Nov 2004, at 22:53, Daniel Senie wrote: > > > Is it SO hard for people to understand that it's possible > today to use > > private address space and public address space in a network WITHOUT > > using NAT? > > I think the hard thing to understand is why you would bother > using 1918 > space if you didn't have to. Yes. > > > In today's networks, printers do NOT need global addresses. > > If they did have globally-unique addresses, I bet they would > still work > just fine, though. > They would of course. I'm not sure why the proposal wouldn't block off some space to cover "unforseen" circumstances and leave it at that. IIRC, 1918 came into existence because of clashes with stock SunOS and Solaris and growth of the internet - for the most part. Much like the need for example.com to be maintained by IANA so publications can have their domain.tld examples. I agree with Leo about fees and RIR responsibility though. There's no reason *not to maintain the space in a reasonable fashion and that costs money. When money is spent, justifications are required. I have no problem justifying use of endless space regardless of the end. ;) I understand why people want to make it free. The reality is that it won't work that way again. Let's not even go down the "free the network" path. I fear another RMS song. YMMV. -M<

2 1

Status of FCAPS model? Useful? Obsolete?
by jm 09 Nov '04

09 Nov '04

Someone at Forrester research wrote an article in 2003 that said FCAPS was an obsolete model because it was conceived during a time when mainframes were in use. I haven't read the article but the premise of it seemed a bit overboard to me. Does the FCAPS model still hold currency among network managers/engineers today? Do the functional categories it uses make sense of the management activities of modern networks? Is there a set of activities that in and of itself warrants further categorization? For example, should project management be added to the model since most well run networks emanate from adherence to generally accepted project management processes? We can all think of examples where a piece of network technology does not map neatly into the OSI model. But because enough of what goes on in networking does map neatly to that model, it is still useful to refer to it. I believe the same is true with FCAPS. I see too that companies like Cisco still refer to it in some of their Networkers presentations. As I understand it, the purpose of the model is to outline the overarching categories of activities necessary for the successful management of a sizeable network. The ultimate impact of failing to manage these areas properly is loss of productivity, revenue or opportunity. Outside the realm of profitability, the FCAPS model is a teaching device whose purpose, like all models, is to explain phenomena which otherwise would be a confusing, jumbled mess. Does the study of FCAPS profit a person if they intend to manage well a network? Is there some better model worthy of study? If so, why? Sean had some comments on the shortcomings of FCAPS for carrier networks awhile back. Any fresh comments are welcome. Thanks, JM

1 0

Content Delivery Networks/GSLB
by M. Huda 09 Nov '04

09 Nov '04

Hello, I am wondering if somebody can point me to the links where I can found information about Content Delivery Network Solutions used in the market today. I need to know about the technology and how the solution/company (such as Akamai) caters its customers. Do they mirror the content across their server's network? If this is the case then how a request is directed to the closest and lightly loaded server on Internet? There are other hardware (GSLB on F5, BigIP and Cisco CSS) and software (ultraDNS) solutions in the market as well but its difficult to relate those with a CDN solution such as of Akamai's or others. Does anybody have experience with F5 or BigIP GSLB solutions? I would appreciate if somebody can help me out here as well. Thanks, Huda

3 2

DNS Problems on Saturday Night?
by John Neiberger 08 Nov '04

08 Nov '04

Forgive me for not having more technical information about this issue. Beginning sometime around 4:00 PM MST on Saturday, I started seeing horrible slowness on my home Internet connection through Comcast, and I also noticed that I was seeing numerous timeouts on DNS lookups. At the same time, a friend of mine who also uses Comcast was seeing the same thing. Approximately a third of my DNS lookups were timing out. That alone would be fairly easily explainable. However, our DNS servers at work also were having a really bad time trying to resolve external addresses. One server uses a Sprint circuit and the other server uses a Time Warner Telecom circuit, but they both point to UltraDNS. This strange behavior continued until roughly 9:00 PM MST and then the DNS problems cleared up both at home and at work. I haven't seen anyone else mention it yet but was there some sort of fairly large DNS problem for a few hours on Saturday? I don't know enough about DNS to know what form that problem might take. At this point I'm assuming it was just one of those things but I'm curious to know if the problem was more widespread. Regards, John --

2 1

RE: Important IPv6 Policy Issue -- Your Input Requested
by Jason Frisvold 08 Nov '04

08 Nov '04

> -----Original Message----- > From: Eric Gauthier [mailto:eric@roxanne.org] > Subject: Re: Important IPv6 Policy Issue -- Your Input Requested > > > > Hello, > > I must admint, I'm really not up on the more subtle aspects > of v6 addressing > nor have I read the drafts you posted, but I've never Nor have I ... I'm just starting to look at IPv6 now.... This seems like a good discussion to jump in on though. :) > understood why we needed > a new set of RFC1918-like IPv6 space. Wouldn't 0::10.0.0.0/104, > 0::192.168.0.0/112, and 0::172.16.0.0/116 (or whatever the > appropriate masks > would be) all function as v6 addresses with exactly the same > properties at the > current RFC1918 space? If the existing RFC1918 space will exist in IPv6 as described above, that can, presumably, be used in the same way existing 1918 space is. For instance, as non-routable loopback addresses for routers, switches, etc. Correct? Or is IPv6 NAT batter suited for this in the future? > Eric :) -- Jason Frisvold Penteledata

1 0

Update to Strange DNS Issue
by John Neiberger 08 Nov '04

08 Nov '04

I have some additional information that's making me think what I saw on Saturday was just a coincidence. We have two DNS servers that were unable to resolve external addresses beginning around 4:00 PM MST on Saturday, but I found out that they both had just been rebooted. We use some version of BIND on Solaris 9 and it was verified that named was running, yet external names would not resolve for a few hours. External names began to resolve in a few hours with no intervention on our part. BIND gurus, any idea why this might have occurred? If named is running and internal names resolve *and* there are no problems with Internet connectivity, what might cause external lookups to timeout for a few hours only to resolve spontaneously? It seems a little strange and I'm sure there's some important information that I'm missing that would make it all makes sense. Thanks, John --

1 0

k.gtld-servers.net 0wned?
by Iljitsch van Beijnum 08 Nov '04

08 Nov '04

12 out of 13 gtld servers agree: [alumange:~] iljitsch% host -v www.imdb.com m.gtld-servers.net Trying "www.imdb.com" Using domain server: Name: m.gtld-servers.net Address: 192.55.83.30#53 Aliases: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 60746 ;; flags: qr rd; QUERY: 1, ANSWER: 0, AUTHORITY: 2, ADDITIONAL: 2 ;; QUESTION SECTION: ;www.imdb.com. IN A ;; AUTHORITY SECTION: imdb.com. 172800 IN NS udns1.ultradns.net. imdb.com. 172800 IN NS udns2.ultradns.net. ;; ADDITIONAL SECTION: udns1.ultradns.net. 172800 IN A 204.69.234.1 udns2.ultradns.net. 172800 IN A 204.74.101.1 But k says something different: [alumange:~] iljitsch% host -v www.imdb.com k.gtld-servers.net Trying "www.imdb.com" Using domain server: Name: k.gtld-servers.net Address: 218.234.22.221#53 Aliases: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 248 ;; flags: qr aa rd; QUERY: 1, ANSWER: 2, AUTHORITY: 6, ADDITIONAL: 12 ;; QUESTION SECTION: ;www.imdb.com. IN A ;; ANSWER SECTION: www.imdb.com. 1800 IN A 200.139.105.69 www.imdb.com. 1800 IN A 218.234.22.221 ;; AUTHORITY SECTION: com. 1800 IN NS ns1.thirdfloordoors02.com. com. 1800 IN NS ns1.thirdfloordoors03.com. com. 1800 IN NS ns1.thirdfloordoors04.com. com. 1800 IN NS ns1.thirdfloordoors05.com. com. 1800 IN NS ns1.thirdfloordoors06.com. com. 1800 IN NS ns1.thirdfloordoors01.com. ;; ADDITIONAL SECTION: ns1.thirdfloordoors01.com. 1800 IN A 200.139.105.69 ns1.thirdfloordoors01.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors02.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors02.com. 1800 IN A 200.139.105.69 ns1.thirdfloordoors03.com. 1800 IN A 200.139.105.69 ns1.thirdfloordoors03.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors04.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors04.com. 1800 IN A 200.139.105.69 ns1.thirdfloordoors05.com. 1800 IN A 200.139.105.69 ns1.thirdfloordoors05.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors06.com. 1800 IN A 218.234.22.221 ns1.thirdfloordoors06.com. 1800 IN A 200.139.105.69 Received 470 bytes from 218.234.22.221#53 in 1184 ms

3 4

Re: Important IPv6 Policy Issue -- Your Input Requested
by Fergie (Paul Ferguson) 08 Nov '04

08 Nov '04

Would NAT be considered an application? :-) - ferg -- Joe Abley <jabley(a)isc.org> wrote: I don't know of any applications that require RFC1918 addresses to be deployed. (Clearly, this is not to say there are none.) -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net

1 0

RE: British Telecom to buy Infonet
by Scott Weeks 08 Nov '04

08 Nov '04

: From: Fergie (Paul Ferguson) : Date: Mon Nov 08 14:05:49 2004 : : British Telecom will acquire Infonet Services in a deal worth $965 : million, BT said Monday. : : http://news.com.com/British+Telecom+to+buy+Infonet/2100-1037_3-5442816.html… I wonder if they'll do any better than C&W did with Digital Island and Exodus. If not, my complete condolences go out to the Infonet employees. Looks like the investors think they will: http://finance.yahoo.com/q/bc?s=IN&t=5d&l=on&z=m&q=l&c= scott

1 0

British Telecom to buy Infonet
by Fergie (Paul Ferguson) 08 Nov '04

08 Nov '04

British Telecom will acquire Infonet Services in a deal worth $965 million, BT said Monday. http://news.com.com/British+Telecom+to+buy+Infonet/2100-1037_3-5442816.html… - ferg -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net

1 0