September 1998 - Test - lists.nanog.org

Re: And we thought the text part of the Starr Report would be bad
by Eric Sobocinski 23 Sep '98

23 Sep '98

On Tuesday, 22 Sep 1998 at 22:43 EDT, Robert Szarka <szarka(a)downcity.net> wrote: >At 11:22 AM 9/22/98 , you wrote: >> For modem to modem communications, start and stop bits were used >> up through 2400 baud when one bit = one tone. The method of one >> bit = one tone was dropped for other forms of encoding when modems >> started supporting connection speeds in excess of 2400 baud. > >I assume you mean 2400 bps modems, i.e., v.22bis? Admittedly, it's been a >while since I used one, but my recollection is that 2400 bps modems operate >at 600 baud, with 4 bits/baud using QAM. Your memory about v.22bis at 2400 bps is entirely correct. v.22 at 1200 bps/baud was the last of the 1 bps = 1 baud dialup modem standards. However, v.32/v.32bis/v.32ter/v.34/v.34bis all use trellis coding and additional tricks to get high bit rates out of 2400 baud. It is technically possible to get 2400 bps at 2400 baud from any of these latter standards, though realistically the only place you'll find conditions twisted enough to do that is in the lab. --eric (speaking only on my own authority, as usual)

1 0

Re: the GOV top-level domain
by Sean Donelan 23 Sep '98

23 Sep '98

>GOV. was missing from A.ROOT-SERVERS.NET sometime yesterday, but the >example with a timestampe has scrolled past my scroll buffer.... Apparently it is a "known" bug in bind that crops up from time to time. I haven't been closely monitoring the root name servers so I hadn't noticed it before. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

4 4

RE: Frame Relay encap vis-a-vis point-to-point at UUNET
by Chris Fairbanks 23 Sep '98

23 Sep '98

Well, UUnet does terminate T3s into the Cascades. One of our T3s happens to go to UUnet and it is into one of their Cascades at PAIX. It is currently only a fraction circuit. From talking with their installer, they only plug you directly into one of their routers if the circuit is 15mb or higher. We have had this circuit for 9 months now and everything seems to be working great with it. Later, Chris -----Original Message----- From: Deepak Jain [mailto:deepak@ai.net] Sent: Monday, September 21, 1998 3:01 PM To: Jerry Scharf Cc: bjames(a)mis.net; nanog(a)merit.edu Subject: Re: Frame Relay encap vis-a-vis point-to-point at UUNET I have never heard complaints about the Cascades for T-1 termination except where they are terminating too many sessions in the Cascades and the router couldn't handle that many sessions. This was a bigger problem before the 7500 series routers came out and more importantly the RSP4s. Fortunately UUNet doesn't terminate T3s into Cascades or ATM switches (yet...) -Deepak.

3 4

Traffic exceeds link capacity.
by jhyang 23 Sep '98

23 Sep '98

We have router which connected 45m hssi(A) with FR switch. That router contains two other interfaces, one is 45m hssi(B) and the other is 100m fddi (C). Because the most of the inbound traffic from (B) and (C) goes to (A) interface, the toal inbound traffic of (B) and (C) cannot exceed 45m. (Is it right ?) But the output rate of interface (A) exceeds 45m, -I can verify them in "sh int <int-name>"-, it increases upto 60~70 mbps. (Not really, but the statistic from CISCO commands.) When the phenomenen occurs, the total inbound traffic of (B) and (C) shows near 45 mbps (about 43.x mbps). My question is: Is it true that it indicates 60~70 mbps ? (Of cause, it has many output drops.) or Is it bug? The cpu load increases upto 99%, and it freezes router for short time -about 1~2 minute-. How to resolve this errors ? Our Router is: CISCO 7507 RSP2 IOS: 11.1.19 (CC) Enable CEF on (A): It is better than not using CEF. But the result doen't matter. (High CPU load, ..) Any Idea ? Thanks in advance --Jaeho Yang.

4 3

Re: Be Afraid -> InterNIC Email Addresses CD: $249
by Andrea Di Lecce 23 Sep '98

23 Sep '98

All, Just got off the phone with one of the admins at DN.NET, who hosts domainsondisc.com and domainsondisk.com. He informed me that they have received about 40 complaints, and have sent out a communication to the offender. If he does not respond in 24 hours, and cease all violations of their AUP, DN.NET will remove the domain. Hopefully it won't get spread around too much in the next 24 hours. I've also notified the upstream netaxs.net, and sent an email to UU.NET to remove the (probably throw-away) dialup account the spam was sent from. FYI. Andrea. ************************************************** Andrea Di Lecce (416) 935-5700 Rogers@Home Network Operations 1 Mount Pleasant Road, Toronto, M4Y 2Y5 **************************************************

1 0

RE: Get out your textbooks for this one!
by Martin, Christian 23 Sep '98

23 Sep '98

Problem 3, I haven't a clue, but problem 4 is the classical Dijkstra/Bellman-Ford algorithm problem. To build the routing table for each node step-by-sptep would consume much paper! -Chris > -----Original Message----- > From: Lutz, Kyle [mailto:Kyle@birchtel.com] > Sent: Wednesday, September 23, 1998 11:15 AM > To: 'nanog(a)merit.edu' > Subject: Get out your textbooks for this one! > > > I have implemented hundreds of networks and designed them > over the years > but I have not a clue on these questions. > > I realize that this is much more towards hardware design, but I know > that many of us have background there. > > Networking design 101 > > > problem 3: clearly illustrate the design of a knockout > > crossbar switch > > that enables an output port to choose packets from 9 input ports > > simultaneously in a 12x12 switch. your switch design must use > > the FEWEST > > number of 2x2 switching elements. > > > > problem 4: consider the network topology given in figure 2. > > the label for > > each of the links in the figure denotes the delay experience > > by any packet > > traversing the link (in either direction). each node is assumed to > > initially know the delays of its outgoing links and all links > > are assumed > > to be bi-directional. clearly show the routing table of > node A step by > > step, from the initial state to the final state (when the > > routing table > > stabilizes), for the following two cases: > > > > o when the routing algorithm is the Distance Vector algorithm > > o when the routing algorithm is the Link State algorithm > > > > __ 3 __B__ 4 __ > > / \ > > A------- 5 -------C\ H > > |\ | \ / > > | 2 | 1 2 > > 1 \ | \ / > > | E 6 / D > > | | 3 > > F------- 9 -------G/ > > > > > >

1 0

Be Afraid -> InterNIC Email Addresses CD: $249
by Andrea Di Lecce 23 Sep '98

23 Sep '98

All, Looks like someone has catalogued all admin/billing/technical contacts and put them on a CD..... and is selling them. Could someone from the domains mentioned below take action against this fool ? The spam originated from a UUNet dialup account. 1Cust100.tnt2.sdg1.da.uu.net [208.250.122.100] on Tue, 22 Sep 1998 21:16:21 -0700 (PDT). The web site is hosted by netaxs.net: Tracing the route to 209.207.155.207 1 172.16.2.70 24 msec 32 msec 24 msec 2 pennsauken.netaxs.com (192.157.69.47) 152 msec 28 msec * 3 pennsauken-gw.mfs-phl.h5-0-45M.netaxs.net (207.106.2.205) [AS 4969] 72 msec * 68 msec 4 brain.phl-core.fe0-0-100M.netaxs.net (207.8.186.81) [AS 4969] 68 msec 40 msec 40 msec 5 phl-blt.balt-core.h0-0-45M.netaxs.net (207.106.2.18) [AS 4969] 32 msec 44 msec 36 msec 6 blt-dc.dc-core.h5-0-45M.netaxs.net (207.106.2.2) [AS 4969] 36 msec 36 msec 40 msec 7 dn-netaxs-gw.dc-core.h5-0-45M.netaxs.net (207.106.127.94) [AS 4969] 48 msec 48 msec 40 msec 8 209.207.190.5 [AS 7019] 48 msec 48 msec 44 msec 9 * * 209.207.141.41 [AS 7019] 80 msec -- Here's his domain information: Registrant: Techno Trade Online Solutions (DOMAINSONDISC-DOM) 4735 Clairemont Dr. #370 San Diego, CA 92117 US Domain Name: DOMAINSONDISC.COM Administrative Contact: Afifi, Sammy (SA5136) sales(a)DOMAINSONDISC.COM 619-274-2936 Technical Contact, Zone Contact: dnsadmin (DNS37-ORG) dnsadmin(a)DN.NET tel.: 703-642-2800 Fax- .: 703-642-3747 Billing Contact: Afifi, Sammy (SA5136) sales(a)DOMAINSONDISC.COM 619-274-2936 Record last updated on 04-Aug-98. Record created on 04-Aug-98. Database last updated on 23-Sep-98 04:50:10 EDT. Domain servers in listed order: NS1.DN.NET 206.43.192.76 NS2.DN.NET 206.197.81.10 NS3.DN.NET 206.197.81.11 NS5.DN.NET 207.153.203.5 -- And here's his lovely spam. It was sent to my admin account, which rarely hits the public 'net. He must be marketing this CD to the contacts themselves ! -- >Received: from bbcgate.bbc.co.uk by sunf0.rd.bbc.co.uk; Wed, 23 Sep 98 05:17:27 BST >Received: from goose.prod.itd.earthlink.net (goose.prod.itd.earthlink.net [207.217.120.18]) by bbcgate.bbc.co.uk (8.8.5/8.7.2) with ESMTP id FAA29405 for <brandon(a)rd.bbc.co.uk>; Wed, 23 Sep 1998 05:17:25 +0100 (BST) >From: sammy(a)technotrade.com >Received: from mail.earthlink.net (1Cust100.tnt2.sdg1.da.uu.net [208.250.122.100]) > by goose.prod.itd.earthlink.net (8.8.7/8.8.5) with SMTP id VAA12849; > Tue, 22 Sep 1998 21:16:21 -0700 (PDT) >Message-Id: <199809230416.VAA12849(a)goose.prod.itd.earthlink.net> >To: massmail_to_admins(a)technotrade.com >Date: Wed, 23 Sep 98 21:13:51 EST >Subject: InterNIC Email Addresses CD: $249 > > NOW AVAILABLE: > > Every email address of every domain > admin/tech/billing contact on the internet! > > Only $249: > > http://domainsondisk.com or > http://domainsondisc.com > > Domains On Disc is the most up-to-date, complete > and searchable list of contact information for every > registered .com, .net, .edu and .org domain name on > the Internet. > > You may now email offers for your products/services > to 100,000's of prospects for pennies... > > GUERILLA MARKETING AT ITS FINEST!!!!!!!!!! > > The internet will never be the same again! > >----- End Included Message ----- ************************************************** Andrea Di Lecce (416) 935-5700 Rogers@Home Network Operations 1 Mount Pleasant Road, Toronto, M4Y 2Y5 **************************************************

2 1

Get out your textbooks for this one!
by Lutz, Kyle 23 Sep '98

23 Sep '98

I have implemented hundreds of networks and designed them over the years but I have not a clue on these questions. I realize that this is much more towards hardware design, but I know that many of us have background there. Networking design 101 > problem 3: clearly illustrate the design of a knockout > crossbar switch > that enables an output port to choose packets from 9 input ports > simultaneously in a 12x12 switch. your switch design must use > the FEWEST > number of 2x2 switching elements. > > problem 4: consider the network topology given in figure 2. > the label for > each of the links in the figure denotes the delay experience > by any packet > traversing the link (in either direction). each node is assumed to > initially know the delays of its outgoing links and all links > are assumed > to be bi-directional. clearly show the routing table of node A step by > step, from the initial state to the final state (when the > routing table > stabilizes), for the following two cases: > > o when the routing algorithm is the Distance Vector algorithm > o when the routing algorithm is the Link State algorithm > > __ 3 __B__ 4 __ > / \ > A------- 5 -------C\ H > |\ | \ / > | 2 | 1 2 > 1 \ | \ / > | E 6 / D > | | 3 > F------- 9 -------G/ > >

1 0

Re: Sat phones
by shields＠crosslink.net 23 Sep '98

23 Sep '98

In article <980922233256.fe71(a)SDG.DRA.COM>, Sean Donelan <SEAN(a)SDG.DRA.COM> wrote: > In practice, I would expect excavators careful enough to have all > the underground utilities marked ahead of time, are also careful > during their digging. Well, as you know there is no hope for us. When the contractor was digging a trench to install fiber to our building, they managed to cut the 20kV power line. It was marked. All you can do is put in enough redundancy to bring downtime to a level you(r customers) think is acceptable. I think MTTR is underappreciated and lowering it is probably the most cost-effective way to keep your percentage availability high. You periodic NOC contact checking is a good step there. -- Shields, CrossLink.

1 0

RE: How to loadshare over many E1 links
by Martin, Christian 23 Sep '98

23 Sep '98

Jared, I meant the latter. I knew that there were switch to switch FDX capabilities - I'd never seen a router interface or NIC though. Cool. Know any vendors off the top of your head??? -Chris > -----Original Message----- > From: Jared Mauch [mailto:jared@puck.nether.net] > Sent: Friday, September 18, 1998 12:34 PM > To: Martin, Christian; 'Scott Whyte'; 'Jesper Skriver'; > 'nanog(a)merit.edu' > Cc: 'tdk-backbone(a)t.dk' > Subject: Re: How to loadshare over many E1 links > > > On Fri, Sep 18, 1998 at 12:17:47PM -0400, Martin, Christian wrote: > > I didn't know there was any full-duplex 10BaseT spec... > > Yes there is. You need to have cards that support it, > not everything does. > > Most modern switches support 10M fdx links, but not all > router interfaces support them. Some pc cards support it, but > not all. > > It does exist, but is not wideley used at all. Most people > with that much traffic opt to use FE or Fddi > > - jared > > -- > Jared Mauch | pgp key available via finger from jared(a)puck.nether.net > | http://puck.nether.net/~jared/ >

8 8