September 2001 - Test - lists.nanog.org

Re: 220v/50hz power rig
by Steven M. Bellovin 06 Sep '01

06 Sep '01

In message <9n7a89$mkc$1(a)ncc1701.cistron.net>, Miquel van Smoorenburg writes: > >Find a no-break UPS that can handle both 115/230 in and outputs 230 > Note that Randy's requirements were for 10A at 220V, or 20A at 110V. Also note that 110V suffers greater voltage drops, so you want a really short line from the panel box, independent of any inefficiencies in the UPS. (Power -- watts -- is I^2*R, so if you double the current the loss in a given piece of wire is 4x greater. This is why power lines run at massively high voltages.) Since the discussion is ongoing, I'll post the comments I sent Randy privately, wearing my hat as the co-author of the Wiring FAQ: -------- You probably can't do precisely what you want. 10A at 220V is 20A at 110V, which you can find; however, unless the outlet is very close to the panel box, you may get enough voltage drop (at 110V it's more than at 220V) to matter. And those specs look like the device is really intended for a 15A or 20A 220V circuit -- possible, but not common, in American houses. OTOH -- as was pointed out by several folks on NANOG, every American house (and some American businesses) have 220V in the panel box, and many have 220V outlets in a few places. The gadget is designed for 50/60 hz, which means that that shouldn't matter. The real issue in the U.S. is the power switch and assumptions about one line being neutral; for safety's sake, you should use a 220-to-220 isolation transformer, I believe. Many businesses have 3-phase power instead, which provides 208V across two legs, rather than 220-240V. I know that I don't know enough about 3-phase to offer any advice. --Steve Bellovin, http://www.research.att.com/~smb

1 0

What's tool does CW use for performance monitoring?
by Yu Ning 06 Sep '01

06 Sep '01

Hi nanog, Has anyone seen the C&W SLA/performance monitoring page ? The N x N latency and PL matrix is very beautiful. I wonder if anyone know the name of the tool ? Or any tool like it to draw a beautiful table like C&W. http://sla.cw.net/sla/Matrix.jsp?report=T&region=US-US&refresh=300 http://sla.cw.net/ thanks for your input. Yu Ning ____________________________________________ (Mr.) Yu Ning, Chief Engineer ChinaNET(AS4134) Sr. Sup.& New Sev. R&D Internet Product Dep. DCB, China Telecom Beijing, P.R.China +86-10-62072357/62072354 ____________________________________________

2 2

Register.com .....routing issue?
by Los, Ralph 06 Sep '01

06 Sep '01

Hey folks, Has anyone tried to trace to Register.com's main page yet? I haven't been able to hit it, and when I try to trace down to them I end up hitting a 10.0.30.x network, then go back to public IP's...how is that possible? Anyone else having trouble connecting to them? Happy Monday! (still) Ralph M. Los Asst. Vice-President, Internet Systems and Security EnvestNetPMC rlos(a)envestnet.com (312) 827-3945 (direct) (312) 296-9003 (wireless w/voicemail) * If you haven't been hacked, you don't know where your vulnerabilities lie*

9 8

Number of cable cuts is down
by Sean Donelan 05 Sep '01

05 Sep '01

The ATIS Network Reliability Steering Committee met last week to review last quarters network outages. I haven't seen what they've come up with, but last quarter they noted a decreas in facility outages. I also noticed an interesting change in Internet-related outages. The number of cables cuts is down since January 1 2001 compared with the same time period in 2000 and 1999. Previously, one of the primary causes of Internet cable cuts was other carriers' subcontractors installing new fiber in the same right of ways. Most high-bandwidth Internet-carriers seem to use the same right of ways, which aren't identical to the LEC voice infrastructure. For example, the Baltimore tunnel fire had little impact on voice calls, i.e. no carrier filed an FCC outage report, but the fire's impact showed up on Internet traffic. With the change in economic climate, many cable projects have been stopped. My guess is less carrier work in the same right of ways means fewer cable cuts. The number of farmers burying a cow, and gophers chewing on cable is the same. There is a silver lining on every cloud.... Those carriers which have fiber into the ground, are less likely to have cable cuts now.

1 0

RE: ATM failure - No the other kind of ATM
by Rowland, Alan D 05 Sep '01

05 Sep '01

Folk, The story I read referred to a software glitch resulting from unusually high volume over the Labor Day weekend. Most ATMs use OS2 (stand around one until the maintenance person re-boots it). My WAG (wild a** guess) is a fixed length field somewhere, say X,XXX,XXX, and the network had 10,000,000 whatevers over the Labor Day holiday that was then repeated when it came back up due to the pent-up demand from the system being off line. When was the last time you took a look at any data that increments in fixed length fields? Your mileage may vary. -Al -----Original Message----- From: Sean Donelan [mailto:sean@donelan.com] Sent: Wednesday, September 05, 2001 1:11 PM To: nanog(a)merit.edu Subject: ATM failure - No the other kind of ATM As most US residents are aware, the Citibank ATM/Automatic Teller Machine network failed last night. Citibank is still having problems today. >From Reuters: "The Citibank ATM network and other systems experienced a temporary outage last evening." "While our systems were restored by early Wednesday morning, we have continued to experience difficulties this morning. We're working hard to resolve the issues as quickly as possible." Since there is very little "unique" network equipment in the world now, just about everyone buys equipment from the same vendors. So the question is, Can we learn anything from Citibank's experience. Is there anything about their continuing problems which may be used to improve general network reliability? I tried to find out some information from Citibank's web site about the issue, but I didn't see anything.

3 2

Re: ATM failure - No the other kind of ATM
by Patrick W. Gilmore 05 Sep '01

05 Sep '01

At 06:13 PM 9/5/2001 -0400, Chris Woodfield wrote: >As a small clarification, when you say IP, do you mean over the public >internet, or on a private IP network? I'm going to guess the latter, but >thought I'd ask... Ahhhhhhhhhhh, important distinction. Sorry it was not clear. These were all over private lines, modems or private FR. When I said "IP", I meant they just used IP as a transport protocol. (As opposed to raw SNA or something.) I have never set up an ATM that used the Internet. -- TTFN, patrick

1 0

RE: Register.com .....routing issue?
by Matt Zito 05 Sep '01

05 Sep '01

> -----Original Message----- > From: ISPPART:rlos@envestnet.com> > Sent: Wednesday, September 05, 2001 4:13 PM > To: 'nanog(a)merit.edu' > Subject: Register.com .....routing issue? > Sensitivity: Confidential > > > > Hey folks, > > Has anyone tried to trace to Register.com's main page yet? I > haven't been able to hit it, and when I try to trace down to > them I end up > hitting a 10.0.30.x network, then go back to public IP's...how is that > possible? > > Anyone else having trouble connecting to them? > > Happy Monday! (still) > Well, based on our monitoring software, we're up from a variety of places Out There(tm). As far as the 10.30.0 space, we do leak a bit of RFC1918 address space. If you feel you're having connectivity issues, please feel free to contact me directly with your relevant information, including source ip addresses and traceroute info. We'll take this offlist from here. Thanks, Matt -- Matthew J. Zito Systems Engineer Register.com, Inc., 11th Floor, 575 8th Avenue, New York, NY 10018 Ph: 212-798-9205 PGP Key Fingerprint: 4E AC E1 0B BE DD 7D BC D2 06 B2 B0 BF 55 68 99

1 0

RE: ATM failure - No the other kind of ATM
by Chris Boyd 05 Sep '01

05 Sep '01

> -----Original Message----- > From: Sean Donelan [SMTP:sean@donelan.com] > Sent: Wednesday, September 05, 2001 3:11 PM > To: nanog(a)merit.edu > Subject: ATM failure - No the other kind of ATM > [] > Since there is very little "unique" network equipment in > the world now, just about everyone buys equipment from the > same vendors. So the question is, Can we learn anything from > Citibank's experience. Is there anything about their continuing > problems which may be used to improve general network reliability? > > I tried to find out some information from Citibank's web site > about the issue, but I didn't see anything. [] > [] In my experience, banking ATM networks tend to use pretty odd equipment. Most ATM networks tend to be leased lines or use FRADs that speak the teller machine lingo. Some banks use boxes that encapsulate the protocol in TCP, so I guess that a failure there could be a big problem. --Chris

1 0

Re: 220v/50hz power rig
by Sean Donelan 05 Sep '01

05 Sep '01

You want to talk to California Instruments. I believe the box you want is http://www.calinst.com/data/wp.html On Wed, 05 September 2001, Randy Bush wrote: > > you're gonna love this one. > > i have a piece of equipment i will be shipping to china that i want to test > before i ship it over. it has a <gasp!> electric motor! so i need some > kind of transformer/inverter setup that plugs into red blooded american > 115v/60hz and can handle > > Power requirement: 1.5 kW > Voltage requirement: 220 V > Frequency: 50/60 Hz > Current requirement: 10 A > Current protection by user: 16 A (slow blow) > Protection type: IP 54 > Processor: C167, 20 MHz > Spindle motor power: AC 380 Watts > Brake: Electro-dynamic & mechanical > > any clues? thanks. > > randy

1 0

Looking to Buy OC48 Interfaces.
by Martin, Steve 05 Sep '01

05 Sep '01

Greetings, We are in the market to purchase a number of Cisco GSR12012 and/or Juniper M40/M160 OC48 interfaces. If anybody out there has some inventory they would like to off-load (at discount and/or depreciation), please respond off-line. Regards, Steve Steve Martin Sr. Manager, Network Engineering and Operations Conxion Corporation 4201 Burton Drive Santa Clara, CA 95054 DID: 408.566.8565 FAX: 408.980.8240 Cell: 408.988.0003

1 0