- Test - lists.nanog.org

Colo & b/w in Australia
by Anshuman Kanwar 30 Aug '04

30 Aug '04

My company will be setting up presence in a colo in Australia, hopefully Sydney. Business is content heavy-ish. Looking for advice on : 1. Transit back to the US ( low [1-5Mbps] initial commit ). 2. Biggest players in the local DSL/Cable/Broadband market and peering with them. 3. Choice and quality of Colo facilities. I have quotes from vendors, am looking for any personal experience / horror stories. Thanks !

2 1

FBI bust DDoS 'Mafia'
by John Obi 30 Aug '04

30 Aug '04

No comments, check the url http://www.theregister.co.uk/2004/08/27/ddos_mafia_busted/ I'm happy some of these criminals sent to jail! _______________________________ Do you Yahoo!? Express yourself with Y! Messenger! Free. Download now. http://messenger.yahoo.com

9 11

Foundry and Policy Routing
by Scott Weeks 30 Aug '04

30 Aug '04

Good Morning Everyone! I will be asking this on other lists, but I hope to hear from you guys as you're most likely going to have more experience, collectively, than the other lists. I'm trying to do policy routing on a couple of Foundry BigIron 4000 routers. I've put sho ver info below. I am seeing really strange failure modes when doing this. The main one is a random subnet or two will be slowed down to a crawl and then finally not communicate at all. There is seemingly no commonality in the failures as different subnets fail and other subnets on the same router are not affected. There is nothing in the logs. There are some other less obvious failures as well. Anyone out there in NOG-land seeing similarly strange issues on your policy routing Foundry routers? Thanks, scott SW: Version 07.6.04mT53 Copyright (c) 1996-2003 Foundry Networks, Inc. Compiled on Mar 02 2004 at 16:00:56 labeled as B2R07604m (2882638 bytes) from Primary b2r07604m.bin HW: BigIron 4000 Router, SYSIF version 21, Serial #: Non-exist ========================================================================== SL 1: B8GMR Fiber Management Module, SYSIF 2, M3, ACTIVE Serial #: CH20006393 4096 KB BRAM, SMC version 1, BM version 21 512 KB PRAM(512K+0K) and 2048*8 CAM entries for DMA 0, version 0209 512 KB PRAM(512K+0K) and shared CAM entries for DMA 1, version 0209 512 KB PRAM(512K+0K) and 2048*8 CAM entries for DMA 2, version 0209 512 KB PRAM(512K+0K) and shared CAM entries for DMA 3, version 0209 ========================================================================== SL 2: B8G Fiber Switch Module, SYSIF 2 (Mini GBIC) Serial #: ? 2048 KB BRAM, SMC version 1, BM version 21 512 KB PRAM(512K+0K) and 2048*8 CAM entries for DMA 4, version 0209 512 KB PRAM(512K+0K) and shared CAM entries for DMA 5, version 0209 512 KB PRAM(512K+0K) and 2048*8 CAM entries for DMA 6, version 0209 512 KB PRAM(512K+0K) and shared CAM entries for DMA 7, version 0209 ========================================================================== SL 3: B8G Fiber Switch Module Serial #: Non-exist 2048 KB BRAM, SMC version 1, BM version 21 256 KB PRAM(256K+0K) and 2048*8 CAM entries for DMA 8, version 0209 256 KB PRAM(256K+0K) and shared CAM entries for DMA 9, version 0209 256 KB PRAM(256K+0K) and 2048*8 CAM entries for DMA 10, version 0209 256 KB PRAM(256K+0K) and shared CAM entries for DMA 11, version 0209 ========================================================================== Active management module: 400 MHz Power PC processor 740 (version 8/8202) 66 MHz bus 512 KB boot flash memory 8192 KB code flash memory 256 KB SRAM 128 MB DRAM

1 0

RE: FBI bust DDoS 'Mafia'
by Spencer Wood 30 Aug '04

30 Aug '04

I forwarded this message to a Guy I work with in the office..Here is his response.. No crap, this guy lived in our neighborhood... Remember me telling you about the Ryder truck full of Compaq servers and all the FBI agents???? We thought maybe he was into kiddie porn or something... he had a bunch of servers in his basement, and a datacenter-class HVAC unit with a weird looking exhaust pipe coming out of the side of his house. I think it's still there... Spencer ************************************************************ Spencer Wood, Network Manager Ohio Department Of Transportation 1320 Arthur E. Adams Drive Columbus, Ohio 43221 E-Mail: Spencer.Wood(a)dot.state.oh.us Phone: 614.644.5422/Fax: 614.887.4021/Cell: 614.774.3123 *************************************************************

2 1

Re: FBI bust DDoS 'Mafia'
by John Obi 30 Aug '04

30 Aug '04

It's not not OT, or about politics. It's not about they are guilty or not since I Don't care. It's about that "the FBI finally cares, and will bust the script kiddies." This will scare them and will decrease the DDoS attacks from now and on. So ISPs and NSPs should work more with FBI and such parties to decrease the DDoS. Thanks, -J --- "J.A. Terranson" <measl(a)mfn.org> wrote: > > On Sun, 29 Aug 2004, Ricardo "Rick" Gonzalez wrote: > > > > No comments, check the url > > > > > > > http://www.theregister.co.uk/2004/08/27/ddos_mafia_busted/ > > > > > > I'm happy some of these criminals sent to jail! > > > > You know, here in America, we have this concept > called "innocent until > > proven guilty". What country are you from? > > I note for the record that this concept is only held > to be accurate when > applied to politicians or their financiers. The > rest of us are guilty > until convicted, after which our obvious guilt was > finally proven in an > "impartial" court. > > -- > Yours, > > J.A. Terranson > sysadmin(a)mfn.org > 0xBD4A95BF __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - Send 10MB messages! http://promotions.yahoo.com/new_mail

3 2

RE: optics pricing (Re: Weird GigE Media Converter Behavior)
by Michel Py 30 Aug '04

30 Aug '04

>> Michel Py wrote: >> Economics 101. Cisco (and many other vendors, BTW) are not >> charities. Their purpose is to make investors and >> shareholders (which includes me) happy. And yes, this >> includes reselling OEM hardware at astronomical >> prices when they can, because it never lasts long. > Richard A Steenbergen wrote: > Obviously. But us folks who run networks aren't charities either FTR, I made the exact same point many times myself. > Obviously it is the job of the vendor to try and squeeze as > much money out of their customers as they possibly can, but > at least smart folks have the CHOICE not to take the bait. It's not that simple, because of two things: 1. Support and 2. The FUD factor. <me is the devil's advocate one more time> 1. Support: sometimes you will need vendor support, and this is especially true of new products. Putting Kingston DRAM in a 2600 is one thing; a limited test on a few routers will quickly show if it works or not, and the odds of an IOS upgrade that would suddenly trigger the third-party memory to cause problems are close to zero, as DRAM as long passed the development stage and is now a commodity. OTOH, you won't have that many OC-192 IRs or LRs to play with. Maybe you'd try one third party PHY, then another one if the first one works, and so on. And suddenly something changes (which does happen with new products) and your vendor implements the changes on their PHYs but not on yours. You're screwed. 2. FUD: How much revenue do these 4 OC-192s bring? This combined with the multi-million cost of a CRS-1 system, is it worth the risk to save a mere $200k street price or $400k list? Would you put your job on the line for it? </me is the devil's advocate one more time> > We start to get annoyed when the vendors remove that choice > by engaging in practices like locking down GBIC/SFP modules > by vendor ID codes for no reason other than to force customers > into paying absurd markup for their optics, intentionally > designing interfaces with fixed optics so that you have to > purchase more cards than you might actually need in order > to have the necessary optics, etc. Don't get me wrong: I'm equally annoyed, and I think that some pointing out of the practice and whining about it can be healthy. After all, if nobody whined about memory prices, we would still be facing the choice of buying it at 20x cost with support or at 2x without. The point I was trying to make is that during the initial phase of a product, there is likely nothing we can do about it no matter how loud we whine (which should not stop us to whine). > Robert E. Seastrom wrote: > [..CRS-1..] and nobody, not even certain well known crazy people, > is going to pay money for one to use as a SOHO router). Never say never. 7500s are common soho routers and there are a growing number of us that are GSR wannabes, because we can't stand to see our buddies have one. I do not plan to install it in my WC though. Michel.

2 1

RE: Best Practices for Enterprise networks
by Fergie (Paul Ferguson) 30 Aug '04

30 Aug '04

Of course it can work. My point is that it is a fact of life, nothing more. Pointing out the obvious: Dependent upon who is/are your upstream provider(s), and how specific the prefix announcements are made to their peers (re: your reachability) determines just how symmetric your traffic patterns will be. - ferg -- "Michel Py" <michel(a)arneill-py.sacramento.ca.us> wrote: > Asymmetric paths are a fact of life in the Internet. Not for enterprise operators except the largest ones. Asymmetric traffic does happen in the core, where there are no firewalls or NATs; as far as the edge is concerned though I know several companies that multihome to two or more ISPs but only in one location, largely because they don't want to deal with NAT/firewall issues. Although it can work, it requires extra engineering and most of the time a fat pipe to replicate state information between the sites. -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net

1 0

Re: Best Practices for Enterprise networks
by Fergie (Paul Ferguson) 30 Aug '04

30 Aug '04

Asymmetric paths are a fact of life in the Internet. - ferg -- Iljitsch van Beijnum <iljitsch(a)muada.com> wrote: On 30-aug-04, at 0:50, Tracy Smith wrote: > Hello. I am tyring to gauge what the Best Practices are for > Enterprise network connections to the Internet. Specifically, to NAT > or not to NAT? At what point should NAT-ting be performed ... > exclusively at the Egress point or at decentralized points? What > about firewalling - centralized/decentralized? Fortunately, I've never been in the position to make such decisions, but I can tell you one thing: if you have multiple connections to the internet, you had better make sure that your NATs and firewalls are equipped to handle the case where you send a packet out through connection A and the reply comes back through connection B. -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net

2 1

Best Practices for Enterprise networks
by Tracy Smith 29 Aug '04

29 Aug '04

Hello. I am tyring to gauge what the Best Practices are for Enterprise network connections to the Internet. Specifically, to NAT or not to NAT? At what point should NAT-ting be performed ... exclusively at the Egress point or at decentralized points? What about firewalling - centralized/decentralized? Thanks in advance for any feedback! Tracy Smith tsmith(a)illinois.net

2 1

10 GE WAN PHY status?
by Stefan Mink 29 Aug '04

29 Aug '04

Hi, I'm currently looking into how to best integrate 10G lambda lines into a network. The two obvious ways are via "native" SDH/SONET or via 10 GE WAN PHY, but the latter isn't available on any router platform so far. What I like with 10GE is that its possible to run a ring segment via L2-switches on each end and connect to them via multiple L3-devices. So you can basically run a 10G line with a redundant pair of routers on each line concurrently and without any static bandwith partitioning. Am I missing something? Sine none of the long distance providers I talked with so far can provide 10 GE LAN PHY as client side interface, I'm currently looking for L2-switches that can perform this conversion between LAN and WAN PHY, but I only found the Nortel Passport and the Force10 E-Series, both AFAIK quite pricy product lines. Last informations I got about WAN PHY XENPAKs was that most switch manufacturers can only provide them sometime in Q1/2005 (but I read the same in an article on this list about Q1/2004 :] ). Are there other products I missed? Are there XENPAKs already available which work with main stream switch manufacturers although they don't sell them themselves so far? Grateful for any pointers or experiences... tschuess Stefan -- Stefan Mink, Schlund+Partner AG (AS 8560) Primary key fingerprint: 389E 5DC9 751F A6EB B974 DC3F 7A1B CF62 F0D4 D2BA

4 7