January 2001 - Test - lists.nanog.org

Sprint / C&W peering issues?
by jlewis＠lewis.org 19 Jan '01

19 Jan '01

The past few days, I've been noticing packet loss, apparently at the points where Sprint and C&W exchange traffic. Today, I emailed a note about this including the output from a few mtr's in each direction to noc(a)sprint.net, and though I got no reply, an hour or so later, I noticed packets were taking a slightly different route (apparently going through a different peering connection in a different city if you believe the hostnames) and the packet loss was gone and round trip times much better. I left the office for a bit, and now that I'm back, I see the packets are back to using the old peering connection that I can only assume must be overloaded. Anyone know what the deal is? Matt's traceroute [v0.42] orldfl-ns-1.atlantic.net Fri Jan 19 18:58:51 2001 Keys: D - Display mode R - Restart statistics Q - Quit Packets Pings Hostname %Loss Rcv Snt Last Best Avg Worst 1. orldflwcom-br-1-fe0-0.atlantic.net 0% 30 30 0 0 0 1 2. sl-gw8-orl-3-0-TS11.sprintlink.net 0% 30 30 1 0 1 1 3. sl-bb11-orl-5-2.sprintlink.net 0% 30 30 1 1 1 2 4. sl-bb21-atl-9-1.sprintlink.net 0% 30 30 11 11 11 12 5. sl-bb2-atl-0-0-0.sprintlink.net 0% 30 30 12 11 30 190 6. core3-serial2-0-0.Atlanta.cw.net 20% 24 30 37 33 37 41 7. corerouter1.Atlanta.cw.net 27% 22 30 34 33 36 50 8. acr1-loopback.Atlantaald.cw.net 20% 24 30 36 32 37 39 9. bar7-loopback.Atlantaald.cw.net 27% 22 30 48 33 37 48 10. ??? Hop 10 is a router with some packet filtering...no response is expected there. -- ---------------------------------------------------------------------- Jon Lewis *jlewis(a)lewis.org*| I route System Administrator | therefore you are Atlantic Net | _________ http://www.lewis.org/~jlewis/pgp for PGP public key_________

2 1

Too big to fail?
by Sean Donelan 19 Jan '01

19 Jan '01

Remember during the last deregulation cycle. When the Savings & Loan and Bank industries were "deregulated" one open question was: are there banks considered too big to fail. The problem with that doctrine is it warps management's risk analysis. Instead of appropriate investments, management makes excessively risky decisions in an attempt to achieve short-term returns and maximize shareholder value. Is PG&E too big to fail? On Thu, 18 January 2001, "Steven M. Bellovin" wrote: > If PG&E files for bankruptcy, control of the company passes to a > federal judge. If you subscribe to the NY Times site, see > http://www.nytimes.com/2001/01/18/national/18ENER.html > Here's the first paragraph: > > LOS ANGELES, Jan. 17 - Politicians and > power company executives have bickered > for months about how best to solve > California's energy problems, but ultimately it > may come down to this: a bankruptcy court > judge may be the only person with the authority > to ask for the rate increases and cost cuts that a > growing chorus of analysts say are necessary, > but that nobody in the state has been able to > agree upon. > > > --Steve Bellovin, http://www.research.att.com/~smb

2 1

Troubles inside ALTER.NET?
by Vladimir Litovka 19 Jan '01

19 Jan '01

Hello, are there any looking glasses inside of ALTER.NET? We are currently experiencing problems accessing to icq.com, cnn.com and many others: traceroute to 205.188.5.208 (205.188.5.208), 30 hops max, 38 byte packets 1 cisco1 (212.109.32.131) 6.797 ms 39.415 ms 19.684 ms 2 cisco0.kiev.sovam.net.ua (212.109.32.138) 4.373 ms 1.936 ms 5.050 ms 3 ccr-1.Moscow.ST.NET (194.186.157.25) 147.366 ms ccr-1.Moscow.ST.NET (194.186.157.121) 230.353 ms ccr-1.Moscow.ST.NET (194.186.157.189) 146.084 ms [ ... ] 21 0.so-4-0-0.TR2.NYC9.ALTER.NET (152.63.15.182) 201.163 ms 240.414 ms 213.036 ms 22 125.at-5-0-0.TR2.DCA6.ALTER.NET (152.63.2.201) 211.353 ms 240.997 ms 247.147 ms 23 0.so-3-0-0.XR2.DCA6.ALTER.NET (152.63.11.90) 279.065 ms 245.794 ms 200.880 ms 24 184.ATM5-0.GW2.IAD2.ALTER.NET (152.63.36.221) 221.368 ms 282.156 ms 217.497 ms 25 * * * 26 * * * but traceroute from route-views.oregon-ix.net or route-server.exodus.net is ok. Looks like internal problem in ALTER.NET? traceroute above - from 212.109.32.15 Thank you. -- Vladimir Litovka <doka(a)kiev.sovam.com> | "I've seen the future and it's . Phone/Fax: +380 44 4900111 | Cisco switches!" . ICQ/none, talk/none ;) | Cat Alyst

1 0

Re: Announcement: RIPE Whois Database Version 3.0beta2 released
by Nipper, Arnold 19 Jan '01

19 Jan '01

rpsl.ripe.net doesn't seem to exist ... whois -h whois rpsl.ripe.net Daniele Arena whois: whois: Unknown host: Datei oder Verzeichnis nicht gefunden Let's have a look ... ;; ANSWER SECTION: rpsl.ripe.net. 1d23h59m55s IN CNAME reimp.ripe.net. reimp.ripe.net. 1d23h59m55s IN CNAME orange.ripe.net. orange.ripe.net. 1d23h59m55s IN A 193.0.1.239 Yeah, don't do that. Don't use CNAME to CNAME. Pls fix, Arnold ----- Original Message ----- From: "RIPE Database Announcements" <db-news(a)ripe.net> To: <lir-wg(a)ripe.net>; <db-wg(a)ripe.net>; <routing-wg(a)ripe.net>; <eng-coord(a)ripe.net>; <db-beta(a)ripe.net>; <reimp-mig-tf(a)ripe.net> Sent: Friday, January 19, 2001 2:33 PM Subject: Announcement: RIPE Whois Database Version 3.0beta2 released > > Dear All, > > This mail is to inform you that version 3.0beta2 of the RIPE Whois > Database software has been released. This is mainly a bugfix release in > respect to the "beta1" version. > > We remind you that the RIPE Database is about to change its data format, > from RIPE181 to RPSL (the Routing Policy Specification Language, see > RFC2622). This will affect all object types, including "person", "domain" > and "inetnum". > > We suggest you to start to get used to RPSL and version 3.0 . > Please check our RIPE-181 to RPSL Migration page at: > > http://www.ripe.net/rpsl/ > > You can test your skills with RPSL and the functionality of the new > beta release in the following ways: > > - Query rpsl.ripe.net at port 43. You will find there > exactly the same objects that are in whois.ripe.net, > mirrored live and in RPSL format. > > - If you have any script that runs by querying whois.ripe.net, > test it by pointing it to rpsl.ripe.net and make it > v3.0-compliant. > > - Try our CGI interface to the RPSL server: you can find it at > http://www.ripe.net/ripencc/pub-services/db/reimp/. > > - Get to know the new update path: use our "sandbox" database for > RPSL updates. Send your updates to <auto-rip(a)ripe.net> and > query the corresponding database at rpsl.ripe.net, port 4343 . > > Remember that in a few months, that's the way the RIPE Database will be > running; so the earlier you start working with it, the less likely you are > to be unpleasantly surprised later. > > For those running Near-Real-Time Mirrors of the RIPE Database, please be > advised that as soon as the RIPE Database changes its data format, you > will not be able to mirror the RIPE Database using the version 2 of the > Software, since it is not RPSL-compliant. You will have to upgrade to > version 3 to mirror the RIPE Database. Please contact <ripe-dbm(a)ripe.net> > for detailed information. > > Please note that the software is still in a "beta" phase. So we need your > feedback! If you happen to find something that looks like a bug when you > are testing the RPSL database, don't hesitate... > Send us a bug report at <dbrip(a)ripe.net> ! > > If you want to actively participate to the migration effort, you could > join the "RIP migration Task force" and let yourself be heard. Just write > to <ripe-dbm(a)ripe.net> and ask to join the Task Force. > > You can download the latest release of the source code at the following URL: > > ftp://ftp.ripe.net/ripe/dbase/reimp/ripe-dbase-3.0beta2.tar.gz > > You can also join a discussion list related to version 3.0 beta of the > RIPE Database Software. To subscribe to this list, send a message to > <majordomo(a)ripe.net> with the line: > > subscribe db-beta > > in the body of your letter. It is a moderated mailing list so some delays > in subscription are possible. > > > Regards, > > Daniele Arena > ___________________ > RIPE Database Group > RIPE NCC > >

1 0

RE: Second day of rolling blackouts starts
by Derek J. Balling 19 Jan '01

19 Jan '01

At 10:45 AM -0800 1/18/01, Roeland Meyer wrote: >What's wrong with this picture? I see the generators, holding a shotgun at >PG&E's neck, and telling the state that they'll pull the triggers if the >state doesn't come up with the dough. They're not even wearing a mask! Yet, >no one is seeing anything wrong with this and they're acting like it's >PG&E's fault. Welcome to the PRC, ... Peoples' Republic of China^WCalifornia. I pointed out to a co-worker that the "state-mandated low-rate,low-supply,high-demand" power problem was tried in most parts of the former Soviet Union... and those citizens prepare for winters by stocking up on heating supplies as they "Expect" power to go out. This was a known-failed experiment before CA tried it. But, California essentially tried the same thing... high-demand, low-rate, and (through AQMD and other fun things) low-supply. They (through doing this) convinced people that they have a "right" to expect power at below-market-value. Who can blame PG&E's suppliers for being on the "winning side" of a supply-side-economics issue? Who can blame PG&E for being handcuffed by the state's rules? The people of California can, for some reason that boggles this CA resident. But maybe that's cuz I'm not a native. It seems that the natives are all upset, and the "non-native" folks I work with tend to blame California residents themselves for voting/etc. that put the price-fixes in place. "You wanted it, you got it, see how stupid it was? Now pay the price and move on instead of whining". Seem to be in the minority here though. ;-) D (This is operational, insofar as its good advice that "if you want to have consistent power for your colo, you should consider states that don't do silly things to power companies". ;-) ) -- +---------------------+-----------------------------------------+ | dredd(a)megacity.org | "Conan! What is best in life?" | | Derek J. Balling | "To crush your enemies, see them | | | driven before you, and to hear the | | | lamentation of their women!" | +---------------------+-----------------------------------------+

7 7

Re: Too big to fail?
by Steven M. Bellovin 18 Jan '01

18 Jan '01

In message <20010118205105.A1059(a)localhost.co-nectschools.net>, Scott Brim writ es: >On Thu, Jan 18, 2001 at 04:31:58PM -0800, Sean Donelan wrote: >> >> Remember during the last deregulation cycle. When the Savings & Loan >> and Bank industries were "deregulated" one open question was: are >> there banks considered too big to fail. The problem with that doctrine >> is it warps management's risk analysis. Instead of appropriate investments, >> management makes excessively risky decisions in an attempt to achieve >> short-term returns and maximize shareholder value. >> >> Is PG&E too big to fail? > >But what would that mean? All the infrastructure is there, the only >issues are cash flow and regulation of prices. If PGE was diddling with >its infrastructure, lending out towers to shady businesses and such, >there might be a parallel :-). During the 19th century, there were a number of railroad bankruptcies. The courts ruled that since all of their assets were good for only one thing -- a railroad -- the only choice was to keep the corporation going. The real change, per the article I cited, is who makes the hard decisions. --Steve Bellovin, http://www.research.att.com/~smb

1 0

Re: Second day of rolling blackouts starts
by Jim Duncan 18 Jan '01

18 Jan '01

Shawn McMahon writes: > X amount of power takes Y amount of money to produce. > > You can either pay it to a state-sponsored monopoly, or you can pay it to a > private company competing freely, or you can force artificially low prices > through pseudo-private companies operating under cap that generates Y minus > N amount of income, and watch X minus N amount of power be generated. > > No amount of legislation will make Y minus N amount of money produce X amount > of power. No company that is truly competing will charge Y minus N, because > they make more money if they charge Y. Are we gonna be graded on this later? They didn't cover this in Engineering Economy last semester. Jim -- Jim Duncan, Product Security Incident Manager, Cisco Systems, Inc. <http://www.cisco.com/warp/public/707/sec_incident_response.shtml> E-mail: <jnduncan(a)cisco.com> Phone(Direct/FAX): +1 919 392 6209

1 0

Re: Favorites (Re: UUNET peering policy)
by Hal Murray 18 Jan '01

18 Jan '01

> Peering is about "equal value". Consider the pure content provider > co-located witht he pure eyeball provider, where neither have a long haul > network (eg local providers, similar bit/mile cost on each side). The > costs are equally distributed, both get similar "value" from the > interconnection (and indeed, can't exist without each other in the pure > case), but the ratio may well be 10:1. Nobody has mentioned a key step. How do you decide what the value is to each player? I assume everybody here expects the free market to make that decision. I see no reason to assume that each side of a connection is getting an equal value just because traffic is ballanced or the conditions meet somebody's peering checklist. Letting the market make that decision gets complicated if one of the players is a 900 pound gorilla. Note for example, that the price of old newspaper changes from positive to negative as market conditions fluctuate. If something is scarce, it's valuable scrap and people will pay you for it. If there is too much, it's trash and you will have to pay somebody to haul it away. The guy who runs the trucks may be a broker too, trying to make a buck any way he can. I could easily imagine that the price of eyeballs and content would fluctuate in a similar way - or that a backbone would could play broker.

1 0

RE: Second day of rolling blackouts starts
by Mathew Butler 18 Jan '01

18 Jan '01

Los Angeles rates have stayed the same, and they're not having any problems. (It's a public utility, and thus wasn't forced into competition at the same rate as the private companies.) -Mat Butler -----Original Message----- From: Miles Fidelman [mailto:mfidelman@civicnet.org] Sent: Thursday, January 18, 2001 11:12 AM To: Roeland Meyer Cc: 'Sean Donelan'; nanog(a)merit.edu Subject: RE: Second day of rolling blackouts starts On Thu, 18 Jan 2001, Roeland Meyer wrote: > What's wrong with this picture? I see the generators, holding a shotgun at > PG&E's neck, and telling the state that they'll pull the triggers if the > state doesn't come up with the dough. They're not even wearing a mask! Yet, > no one is seeing anything wrong with this and they're acting like it's > PG&E's fault. Is the situation any different in the LA area, where Los Angelas Water and Power is city-owned, and they own a big chunk of their own hydro capacity (or were they forced to divest that)> Miles Fidelman ************************************************************************** The Center for Civic Networking PO Box 600618 Miles R. Fidelman, President & Newtonville, MA 02460-0006 Director, Municipal Telecommunications Strategies Program 617-558-3698 fax: 617-630-8946 mfidelman(a)civicnet.org http://civic.net/ccn.html Information Infrastructure: Public Spaces for the 21st Century Let's Start With: Internet Wall-Plugs Everywhere Say It Often, Say It Loud: "I Want My Internet!" **************************************************************************

2 1

RE: Second day of rolling blackouts starts
by Roeland Meyer 18 Jan '01

18 Jan '01

Actually, my understanding was that it was a forced sell-off. PG&E had no choice. They weren't allowed to keep the generators. > From: mdevney(a)teamsphere.com [mailto:mdevney@teamsphere.com] > Sent: Thursday, January 18, 2001 3:11 PM > > On Thu, 18 Jan 2001, Roeland Meyer wrote: > > > What's wrong with this picture? I see the generators, > holding a shotgun at > > PG&E's neck, and telling the state that they'll pull the > triggers if the > > state doesn't come up with the dough. They're not even > wearing a mask! Yet, > > no one is seeing anything wrong with this and they're > acting like it's > > PG&E's fault. > > > Also note that: > > Half of these generators used to be owned and operated by > PG&E; they were > spun off quickly after deregulation in an effort to boost > profits. Had I > been out of diapers at the time, I would have seen this day > coming back > then. > > Root cause, then: PG&E can't get us electricity because they decided > profit was more important than assured service.

1 0