- NANOG - lists.nanog.org

Re: Phone networks struggle in Hurricane Katrina's wake
by Fergie (Paul Ferguson) 31 Aug '05

31 Aug '05

Wrong Paul. - ferg p.s. I'm doing a blog already. I also run a few networks. It's all relative. Feel free to experiment at will! ;-) - ferg -- Randy Bush <randy(a)psg.com> wrote: >> I'll file that comment where it belongs -- in file 13. manners, paul >> If a major catastrophe, albeit more human than network-related >> (although lots of network-related issues here, too), isn't on-topic, >> than I fail to see what is. > operational material maybe? nah, i'm just a confused lurker, haven't seen > any of it here for a while. the steering committee has been discussing the idea of a nanog blog. of course it would be directed to operational content and not your daily pointer to some cartoon etc. but, in the spirit of an open group, we are very interested to hear what the community thinks of this. but please let's discuss it over on nanog-futures(a)nanog.org. HINT! randy -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net ferg's tech blog: http://fergdawg.blogspot.com/

1 0

MPLS or Site2Site VPN
by Todd Reed 31 Aug '05

31 Aug '05

I'm looking at connecting 15+ multi-state locations together to start forming a private corporate network. The sites are small with 25-30 devices. I want to avoid direct-T1's due to cost, therefore I'm looking for alternatives. I know I can do site-to-site VPN, but I've also heard a lot about MPLS and from what I've read, it may be a good option. Over the next year, we will be adding 5-10 more sites, so expansion is important. I'm not planning to do voice, but it may be an option in 2-3 years. If anyone has any suggestions on their experiences, I would greatly appreciate it. Thanks, Todd

10 9

A useful oversimplification for network surveillance?
by Howard C. Berkowitz 30 Aug '05

30 Aug '05

I'm developing some guidance for ISP surveillance for infrastructure attacks, and my increasing impression is that for other than the expert level, there may be some useful simplifications of the applicability of tools. Remember that I am speaking of surveillance here, not the detailed analysis in a sinkhole. Perhaps this could be the basis of some security architecture presentations/tutorials at NANOG. Let me put up the following strawmen and invite people with flaming torches to go for them, with the caveat that these simplifications are for an introduction to the topic. NetFlow is the key to analyzing traffic patterns outside the router, looking for DDoS signatures when known, and for traffic anomalies that may become DDoS. SNMP is the key to analyzing the effect of exploits on network elements. For example, NetFlow might tell you there is a flood directed at TCP port 179, but your router may implement rate-limiting/policing such that the control processor doesn't see this flood and processor utilization stays within reasonable ranges. Syslog and SNMP traps focus on physical events by people (e.g., reconfiguration), physical problems ranging from temperature alarms to router and interface shutdown, and exploits against security mechanisms. Some of this asynchronous information has undergo root cause analysis: the interface you see go down may be perfectly fine; the problem is in the medium or distant interface.

2 1

Donate [Was Re: Phone networks struggle in Hurricane Katrina's wake]
by Fergie (Paul Ferguson) 30 Aug '05

30 Aug '05

Apologies for the multiple posts, but I think this is important enough to warrant a follow-up. I send out a public challenge to each and every one of you reading this to make a donation to the American Red Cross, if for nothing else, think of it as a small effort to assist the Red Cross in their efforts to establish emergency communications in the region. Of course, the donation will go towards all assistance efforts in the wake of Katrina, so rationalize it however you will. :-) http://www.redcross.org/ Click and donate. - ferg -- "Fergie (Paul Ferguson)" <fergdawg(a)netzero.net> wrote: And via Slashdot: [snip] "In this age of cheap commoditized consumer electronics and advanced mobile technology, why can't all the people of a city make contact during an emergency? Cell phone circuits filled up during 9/11 attacks and in the wake of hurricane Katrina very few victims can make contact with their families, despite the fact that they have all those mobile phones. The Red Cross is looking to deploy satellite equipment From the article: to restore communications in affected areas." "Katrina made landfall in Louisiana early this morning with sustained winds of 145 mph, but veered just enough to the east to spare New Orleans a direct blow. Even so, flooding, power outages and heavy damage to structures were reported throughout the region. The Red Cross tomorrow expects to begin deploying a host of systems it will need, including satellite telephones, portable satellite dishes, specially equipped communications trucks, high- and low-band radio systems, and generator-powered wireless computer networks, said Jason Wiltrout, a Red Cross network engineer." [snip] http://www.computerworld.com/securitytopics/security/recovery/story/0,10801… - ferg -- "Fergie (Paul Ferguson)" <fergdawg(a)netzero.net> wrote: Via Reuters. [snip] Telephone companies struggled to restore service and measure the damage to their networks in Louisiana and Mississippi on Tuesday after Hurricane Katrina cut power and triggered severe flooding. A spokesman for BellSouth Corp., the largest local telephone company in the region, said while the company estimated about 53,000 lines were out in the two states, the actual numbers were likely to be higher. Cingular Wireless and Sprint Nextel Corp. said cellular service in the area had been affected as well. All three companies said power losses were the main threat to further service failures, but that flooding was hampering their efforts to reach network equipment. Entergy Corp. reported more than a million customers without power in Louisiana and Mississippi, and warned customers to expect a long and difficult restoration that could take weeks. New Orleans Mayor Ray Nagin told television station WWL that 80 percent of the city was under water, and authorities declared martial law in some areas. [snip] http://go.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=9512696 - ferg -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net ferg's tech blog: http://fergdawg.blogspot.com/

1 0

RE: Phone networks struggle in Hurricane Katrina's wake
by Peering 30 Aug '05

30 Aug '05

We just received word they were evacuating the folks out of the Superdome too, as that's now in 3 feet of water...and rising. Diane Turley Sr. Network Engineer Xspedius Communications Co. 636-625-7178 -----Original Message----- From: owner-nanog(a)merit.edu [mailto:owner-nanog@merit.edu] On Behalf Of Fergie (Paul Ferguson) Sent: Tuesday, August 30, 2005 2:46 PM To: nanog(a)merit.edu Subject: Phone networks struggle in Hurricane Katrina's wake Via Reuters. [snip] Telephone companies struggled to restore service and measure the damage to their networks in Louisiana and Mississippi on Tuesday after Hurricane Katrina cut power and triggered severe flooding. A spokesman for BellSouth Corp., the largest local telephone company in the region, said while the company estimated about 53,000 lines were out in the two states, the actual numbers were likely to be higher. Cingular Wireless and Sprint Nextel Corp. said cellular service in the area had been affected as well. All three companies said power losses were the main threat to further service failures, but that flooding was hampering their efforts to reach network equipment. Entergy Corp. reported more than a million customers without power in Louisiana and Mississippi, and warned customers to expect a long and difficult restoration that could take weeks. New Orleans Mayor Ray Nagin told television station WWL that 80 percent of the city was under water, and authorities declared martial law in some areas. [snip] http://go.reuters.com/newsArticle.jhtml?type=technologyNews&storyID=9512 696 - ferg -- "Fergie", a.k.a. Paul Ferguson Engineering Architecture for the Internet fergdawg(a)netzero.net or fergdawg(a)sbcglobal.net ferg's tech blog: http://fergdawg.blogspot.com/

1 0

Re: NANOG as the Internet government?
by J. Oquendo 30 Aug '05

30 Aug '05

> > http://www.networkworld.com/columnists/2005/082205johnson.html > /* ARTICLE Did the Internet/IETF governance model work? In many respects, yes. Early on, the IETF produced key protocols at a much faster clip than other network standards bodies (such as the IEEE and the ITU). Although many IETF veterans strenuously object to calling the IETF a "standards body," whatever you call it, the IETF did an outstanding job midwifing protocols and accelerating the 'Net's adoption. */ Agree and disagree. There are still many things that need to be ironed out within the IETF and so called standards. Let's take a look at some of the protocols that have been broken, remodified, rebroken (slight Bushism), patched, entirely rewritten into a new "concept", RFC, etc. I think that a body similar to the IETF would do justice, but selection of who's on first, would have to be done on a voting basis or sorts. /* ARTICLE Does the model still work? I'm not sure. In my view, the biggest concerns facing the Internet today are regulatory and operational, rather than technical. For example, how do we encourage providers to respect each other's QoS tags? Is it acceptable for providers to censor traffic for competitive advantage? Should providers be required to devote some of their revenues toward services "for the common good," such as universal Internet access? */ Model of what... Putting in an RFC, getting comments from everyone what's the saying? "Too many indians not enough chiefs" or "Too many hands in the pot spoil the stew". Anyhow, I say this looking at broken protocols that are vulnerable to all sorts of mayhem and have been broken for years because it wouldn't be in the best interest to make things right and have everyone reconfigure their networks. Granted rebuilding a backbone NAP would be a horror story, everyone points to IPv6 as a solution and how grandiose it will be, yet IPv6 ("The Secure IP!) has been broken too. Not only that how many large providers are willing to take a hit in the pockets getting everything running the way it should be run. Why should they when they could do some shoddy patchwork until the next big hit. I know I'm rambling on, but come on now NANOG'ers as the Internet government. Some of the people here are great teachers in their own right and over the years I've probably learned more from NANOG than I have from any book, RFC, professor, etc., but I also know there are plenty of crybabies, plenty of morons, and even some on the IETF who have snubbed the notion of fixing broken protocols. I say this on the basis of me contacting quite a few on the issues of BGP/SBGP, ICMP and how I could break it out of boredom. Response "Shoo fly... You don't have any certs..." or "Hush... By you releasing horribly written papers with information you're going to cause mayhem." And other things along those lines. /* ARTICLE So what should we do? One answer is to call in the federal government. I'm not a huge fan of government regulation; it can be better than the alternatives, but regulation tends to slow down an industry's rate of innovation. Moreover, the Internet is international, so whose federal government would we turn to as the referee? Yet waiting for the free market to answer these questions doesn't seem to be working, either. */ Problems with selecting people from any company or government are "agendas". Who is to say that someone could be trusted from say taking a nice little payout to hush up on a problem. Not making an accusation lest someone at Cisco want to bore me with the threat of a lawsuit, but who is to be certain that even if some body was selected, you wouldn't have to worry about the big boys in industry paying to tweak "the Internet" to their liking. What if say Cisco (who has this huge issue their trying their best to cover up), greased the pocket of those in this body to quash the notion of Cisco having broken routers. Aside from that, what standards would this body set? Ten Commandments of the Interweb i. Thou shall route thy competitors packets fairly ii. Thou shall not install network analyzers without international warrants iii. Thou shall not allow evil traffic to pass through ones routes iv. Thou shall give access to any authority figure with or without warrants v. Thou shall maintain route tables vi. Honor thy NEIGHBOR_AS vii. Honor thy Backbone viii. Thou shall not null route thy neighbor ix. Thou shall play fairly with VoIP carries even whenst thine own's ILEC/CLEC loss revenue x. Thou shall remember all routes and AS's /* ARTICLE Call it the International Association of Networking Service Providers (IANSP). */ What about "Yet Another Acronym to Add in Some Dictionary That No One Will Respect in the Morning" (YAAASDTNOWRM) =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ J. Oquendo GPG Key ID 0x97B43D89 http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x97B43D89 It is much easier to suggest solutions when you know nothing about the problem. -- Niklaus Wirth

4 3

NANOG as the Internet government?
by Steven M. Bellovin 30 Aug '05

30 Aug '05

http://www.networkworld.com/columnists/2005/082205johnson.html --Steven M. Bellovin, http://www.cs.columbia.edu/~smb

1 0

Martial Law declared in New Orleans Was: RE: Katrina could inundate New Orleans
by Hannigan, Martin 30 Aug '05

30 Aug '05

Breaking news..Apparently a 200 foot section of levee broke last night and is gradually burying the city. Martial Law has been declared in the area as well. Overnight Levee Break: http://www.theadvertiser.com/apps/pbcs.dll/article?AID=/20050830/NEWS05/508… Martial Law: http://jurist.law.pitt.edu/paperchase/2005/08/breaking-news-martial-law-dec… -- Martin Hannigan (c) 617-388-2663 VeriSign, Inc. (w) 703-948-7018 Network Engineer IV Operations & Infrastructure hannigan(a)verisign.com > -----Original Message----- > From: owner-nanog(a)merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of > Matthew Kaufman > Sent: Monday, August 29, 2005 11:47 AM > To: nanog(a)merit.edu > Subject: RE: Katrina could inundate New Orleans > > > > Dave Stewart: > > Y'know... I do have to wonder whether Internet access is > > nearly as important as power and communications (traditional > > comms, such as the PTSN). > > > > Granted, it'll be interesting to see how things shake out - > > but I just can't buy that getting the Internet working > > should/will be a really high priority. > > Back when I was running ISPs, we had several county and city Emergency > Operations Centers as customers... Either on T1 or frame > relay for their > primary service, or as their "backup" dial-on-demand ISDN > provider. These > connections were how the EOC got river gauge data for planning flood > evacuations (at the time, no other source other than having > the numbers read > off from the state-level agency office over the phone if they > weren't too > busy), USGS earthquake epicenter (also available over EDIS) > and shake map > (Internet only) data, weather service radar and satellite > images (backup was > TV broadcasts, if still on the air), and in some counties, > the only access > to the hospital emergency room status tracking system used for > multi-casualty incidents... While there's more private data > networks online > now, there's also more Internet-available data that the EOCs > would like to > have access to, I'm sure (I know that some cities are using > Internet-connected webcams to do security monitoring, look at > shorelines, > etc.) > > In many incident scenarios (and a few actual incidents), the > priority was > that the radio system stayed up, then Internet access, *then* > PSTN (and > having cellphone access to people in the field to supplement the radio > system was more important than landline calls to anywhere > else). And power, > of course, is easily generated locally, so not a big priority at all. > > Interestingly, almost none of the agencies told sales what > the connection > was going to be used for... Only when engineering made a > followup inquiry > would we learn that, yes, in an emergency, they'd like theirs > fixed first > please, and yes, they'd need first dibs on the backup power > if we didn't > have enough to run everything. > > Matthew Kaufman > matthew(a)eeph.com > > >

2 1

Order of ASes in the BGP Path
by Abhishek Verma 30 Aug '05

30 Aug '05

Hi, Is the order of AS numbers (except for perhaps the first one which denotes the AS the route was originated from) in the AS_PATH in BGP important? In fact, does anybody even care for the first AS number that appears in the Path? AFAIK, AS numbers in the BGP serves two purposes. It helps in loop detection and it helps us count the AS Path length. If this is the case then the order should not really matter much. My question is that whether the operators care if the order, for some reason changes? Eg. Legend: {} denotes the sequence, while [] denotes the set Path {1 2} [3 4] {5} Would somebody mind if this was represented as {1 2 5} [3 4] ? Thanks, Abhishek

11 16

Re: Order of ASes in the BGP Path
by Robert Bonomi 30 Aug '05

30 Aug '05

> Date: Mon, 29 Aug 2005 22:15:26 +0530 > From: Abhishek Verma <abhishekv.verma(a)gmail.com> > Subject: Order of ASes in the BGP Path > > > Hi, > > Is the order of AS numbers (except for perhaps the first one which > denotes the AS the route was originated from) in the AS_PATH in BGP > important? In fact, does anybody even care for the first AS number > that appears in the Path? > > AFAIK, AS numbers in the BGP serves two purposes. It helps in loop > detection and it helps us count the AS Path length. AS_PATH is used for various administrative 'policy' matters, as well. It is not used in the actual routing of data. that functionality is accomplished using only the destination netblock and 'next hop' info. > > If this is the case then the order should not really matter much. Except for standards compliance. <evil grin> Per the RFCs on the subject, if you _receive_ an unordered set from a downstream, you can propogate that unordered set, but you must prepend your AS in the 'ordered' fashion. And you must use the ordered path tagging for any new stuff you generate. > My question is that whether the operators care if the order, for some > reason changes? > > Eg. > > Legend: {} denotes the sequence, while [] denotes the set > > Path {1 2} [3 4] {5} As I understand the specs, that is -not- allowed. an unordered set can appear only as the _last_ element of the AS path list. > > Would somebody mind if this was represented as {1 2 5} [3 4] ? Did the topology change? Or did "somebody" just decide to re-write the AS_PATH elements ? your first illustration would 'apparently' describe a topology on the order of; +-3-+ / | \ 1--2 | 5 \ | / +-4-+ where the latter describes something on the order of: 3 /| 1--2--5-+ | \| 4 Anybody that is 'mapping' interconnections is going to be "unhappy" if the 'representation', and the 'reality' are not consistent. :)

3 2