- Test - lists.nanog.org

Re: NSI bulletin 097-004 | Root Server Problems
by Steve Goldstein 18 Jul '97

18 Jul '97

At 05:02 PM 7/17/97 PDT, Randy Bush wrote: >Go back to Taiwan. > > I'm still in Taipei. Returning saturday. --/sg ========================================== Steven N. Goldstein, Ph.D. Program Director, International Networking National Science Foundation 4201 Wilson Blvd. Room 1175 Arlington, Virginia 22230 USA +1-703-306-1949 +1-703-306-0621 FAX ==========================================

1 0

NSI bulletin 097-004 | Root Server Problems
by David Holtzman 18 Jul '97

18 Jul '97

On Wednesday night, July 16, during the computer-generation of the Internet top-level domain zone files, an Ingres database failure resulted in corrupt .COM and .NET zone files. Despite alarms raised by Network Solutions' quality assurance schemes, at approximately 2:30 a.m. (Eastern Time), a system administrator released the zone file without regenerating the file and verifying its integrity. Network Solutions corrected the problem and reissued the zone file by 6:30 a.m. (Eastern Time). Thank you. David H. Holtzman Sr VP Engineering, Network Solutions dholtz(a)internic.net

7 9

RE: BOOM! there goes WorldCom
by Brian Koehl 18 Jul '97

18 Jul '97

As to railroad right-of-ways, they will only let you within 30 feet of the tracks, which is still in the ror. It is safer than anywhere else because of the difficulty of getting permits to dig in railroad ror (5+ years here in Washington) even for other utilities. Some idiot with a backhoe can still be way of target though and get you. Derailments and floods are also big problems in rr-ror. The only real protection you have is ringed SONET, however, even in that arrangment some carriers do not have the ability to reroute everything, just the priority stuff. So they'll still have outages. It's all a matter of capacity and what electronics they use. <lurking mode off> Previously, the telephony industry has had a surplus of fiber. But since the Internet has taken off, much of that excess has been taken by dedicated Internet customers, and expanding POP sites. Now, many facilities-based carriers are having to lay new fiber to meet the unending demand. Perhaps the fiber cuts we are seeing are more often related to construction of this new fiber, and not freak accidents (farmer joe and his shovel). If so, then no matter where you run the fiber, if it sits next to existing fiber run, you run the risk of new construction accidents. In effect, the backhoe in the above example will be "targeting" the fiber already in the ground. -Brian brian(a)mail.usld.net @USLD Internet Services The views come from me and are mine, not USLD's. <lurking mode on>

3 2

Mailing list Spam
by Chris Fairbanks 18 Jul '97

18 Jul '97

Is this a closed list, that only subscribed users can post to, because I got some Spam mail earlier that was sent to the list. I have included the headers below. Sorry for the off topic post. Chris =-=-=-=-=-=-=-=-=-=-= Chris R. Fairbanks Network Engineer WinterLAN, Inc. 2910 Telegraph Ave. Berkeley, CA 94705 Voice +1.510.486.1812 Fax +1.510.486.1796 http://www.wli.net/ Received: from merit.edu by wl-exchange.winterlan.com with SMTP (Microsoft Exchange Internet Mail Service Version 5.0.1457.7) id N475CV8W; Thu, 17 Jul 1997 20:49:57 -0700 Received: from localhost (daemon@localhost) by merit.edu (8.8.5/8.8.5) with SMTP id XAA24686; Thu, 17 Jul 1997 23:44:23 -0400 (EDT) Received: by merit.edu (bulk_mailer v1.5); Thu, 17 Jul 1997 23:44:08 -0400 Received: (from majordom@localhost) by merit.edu (8.8.5/8.8.5) id XAA24661 for nanog-outgoing; Thu, 17 Jul 1997 23:44:07 -0400 (EDT) Received: from web.ione.net ([206.99.162.12]) by merit.edu (8.8.5/8.8.5) with SMTP id XAA24657 for <nanog(a)merit.edu>; Thu, 17 Jul 1997 23:44:04 -0400 (EDT) From: 34706115(a)zignzag.com Received: from hilo131129.pac.gte.net (hilo131129.pac.gte.net [207.115.131.129]) by web.ione.net (NTMail 3.02.13) with ESMTP id ha119789 for <nanog(a)merit.edu>; Thu, 17 Jul 1997 23:35:00 +0100 Received: from mailhost.zignzag.com(alt.zignzag.com(890.34.34343.) by zignzag.com (8.8.5/8.6.5) with SMTP id GAA01401 for <>; Thu, 17 Jul 1997 17:19:27 -0600 (EST) Date: Thu, 17 Jul 97 17:19:27 EST To: Friend(a)public.com Subject: Credit Card$$$ NO Credit Check,,,,,,Make MoNeY!! Message-ID: <3472654.0219GAA(a)zignzag.com> Reply-To: kallen(a)zignzag.com X-PMFLAGS: 34078848 0 X-UIDL: 2610442066a78aeb1b128fda426c9a5c Comments: Authenticated sender is <success(a)moni4u2.com> Sender: owner-nanog(a)merit.edu

1 0

Re: tiers? (fwd)
by Matthew James Gering 18 Jul '97

18 Jul '97

> Somehow I don't think the Tier 1 "definition" will ever be defined and agreed > upon. Marketing depts. throughout the industry have used it and all come > up with their own definition to suit their network No kidding. I once encountered a hole-in-the-wall provider in Seattle that claimed to be "Tier 1." When I disagreed he had the gall to tell me that he was an Internet guru and had originally coined the term "Tier X." So much for truth in marketing. Unless a user is trying to do cross-country VPN, a good regional multi-homed provider has the best performance, quality and service IMHO. Matt

1 0

Re: BOOM! there goes WorldCom
by Peter Lothberg 18 Jul '97

18 Jul '97

> Sprint and AT&T are implementing SONET protection switching which will > reroute circuits around fiber cuts in 50ms. I don't know if they do it It's since long completed in the Sprint network. -Peter

1 0

Re: BOOM! there goes WorldCom
by Peter Lothberg 18 Jul '97

18 Jul '97

> On Wed, 16 Jul 1997, Nathan Stratton wrote: > > > > > Ya, we are going to kill worldcom for this one, they say they have 496 DS3 > > connections down. We have 3 connections into our Palo Alto POP, and > > Worldcom said they all are over different fiber paths. As of right now > > they all are down. > > You should never trust what your sales people say. When designing your > backbone, its important to ask them for your DLRs (design layout records) > so that you can see physically, with your own eyes, the paths involved. > > -Golan > > There is someting called "Sonet" out there, if provisioned as 1+1, eg a backup path for each primary path it's capable of switching your circiuts into the protect path in less than 100ms... So, when getting capacity make sure that it is provisioned over 1+1 sonmet and thet the local loop also is redundant and thet you get a primary and protect router. --Peter

1 0

Re: NSI bulletin 097-004 | Root Server Problems
by Steve Goldstein 18 Jul '97

18 Jul '97

At 01:11 PM 7/17/97 PDT, Randy Bush wrote: >Gosh, Perry. I wish I could be as important when I grow up. > 1. You are already most important... 2. Please don't grow up; you're more fun this way... --SG ========================================== Steven N. Goldstein, Ph.D. Program Director, International Networking National Science Foundation 4201 Wilson Blvd. Room 1175 Arlington, Virginia 22230 USA +1-703-306-1949 +1-703-306-0621 FAX ==========================================

2 1

Re: BOOM! there goes WorldCom
by Eric Kozowski 17 Jul '97

17 Jul '97

> You're right. I forgot about the decomissioned pipelines. Although, >you'd be surprised how stupid some of these morons with backhoes are! I >remember a major MCI cut along their west coast route a few years ago >that took out the entire northwest for 8 hours. The guy was 600 yards >away from where his permit allowed him to dig. He stopped when he cut >the fiber and shut his hoe down. When they got out there, they found >that he was digging right next to train tracks, was inches from cutting >a high voltage power line that surely would have killed him, and was >also fairly close to a gas line. It's not always the equipment operator. I was witness to a ConTel fiber cut along Route 66 aboard MCLB Barstow. The fiber was cut by, get this, a road grader. The fiber was buried (if you could call it that) along the shoulder of the road (which sloped upward about three feet), three feet from the edge of the road, _at road level_! It was also over 6 feet out of alignment with the orange markers. In the process of widening the road, the grader caught the conduit and broke it, along with the fiber. The operator didn't even know he did it until the ConTel crew showed up. Crude ASCII art of where the fiber was buried (in profile): ----------- / _______________/ X <------- Fiber ^ | Road This cut supposedly (according to the ConTel repair crew) killed phone service, including 911, for 120 square miles of eastern California. I heard that they went to a SONET ring shortly after this incident..... -- Eric Kozowski VP Internet Services eric(a)structured.net Structured Network Systems, Inc. (800)390-5945 Support A Verio Affiliate (800)881-0962 Sales/Info http://www.structured.net/ PGP Key fingerprint = 0A E4 91 B8 BD CB 3E 95 42 12 04 4E 14 DF 86 76 'They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.' -- Benjamin Franklin 1759

1 0

root nameserver problems ?
by MCI Hostmaster 17 Jul '97

17 Jul '97

-----BEGIN PGP SIGNED MESSAGE----- - From our vantage point, it looks like most of the root nameservers have bad delegation data. Most of them return no delegation info for what should be working domains: roy@ns% foreach ns ( a b c d e f g h i j k l m ) do echo $ns.root-servers.net host -t ns digital.com $ns.root-servers.net host -t ns webcrawler.com $ns.root-servers.net echo done a.root-servers.net digital.com NS CRL.DEC.COM digital.com NS NS11.digital.com digital.com NS NS.DEC.COM webcrawler.com NS NS00.EXCITE.COM webcrawler.com NS NS01.EXCITE.COM webcrawler.com NS NSE00.EXCITE.COM webcrawler.com NS NSE01.EXCITE.COM b.root-servers.net digital.com does not exist at b.root-servers.net (Authoritative answer) webcrawler.com does not exist at b.root-servers.net (Authoritative answer) c.root-servers.net digital.com does not exist at c.root-servers.net (Authoritative answer) webcrawler.com does not exist at c.root-servers.net (Authoritative answer) d.root-servers.net digital.com does not exist at d.root-servers.net (Authoritative answer) webcrawler.com does not exist at d.root-servers.net (Authoritative answer) e.root-servers.net digital.com does not exist at e.root-servers.net (Authoritative answer) webcrawler.com does not exist at e.root-servers.net (Authoritative answer) f.root-servers.net digital.com does not exist at f.root-servers.net (Authoritative answer) webcrawler.com does not exist at f.root-servers.net (Authoritative answer) g.root-servers.net digital.com does not exist at g.root-servers.net (Authoritative answer) webcrawler.com does not exist at g.root-servers.net (Authoritative answer) h.root-servers.net digital.com NS NS.DEC.COM digital.com NS CRL.DEC.COM digital.com NS NS11.digital.com webcrawler.com NS NS00.EXCITE.COM webcrawler.com NS NS01.EXCITE.COM webcrawler.com NS NSE00.EXCITE.COM webcrawler.com NS NSE01.EXCITE.COM i.root-servers.net digital.com NS CRL.DEC.COM digital.com NS NS11.digital.com digital.com NS NS.DEC.COM webcrawler.com NS NS01.EXCITE.COM webcrawler.com NS NSE00.EXCITE.COM webcrawler.com NS NSE01.EXCITE.COM webcrawler.com NS NS00.EXCITE.COM j.root-servers.net digital.com NS record currently not present at j.root-servers.net webcrawler.com NS record currently not present at j.root-servers.net k.root-servers.net digital.com NS record currently not present at k.root-servers.net webcrawler.com NS record currently not present at k.root-servers.net l.root-servers.net digital.com NS record currently not present at l.root-servers.net webcrawler.com NS record currently not present at l.root-servers.net m.root-servers.net digital.com NS record currently not present at m.root-servers.net webcrawler.com NS record currently not present at m.root-servers.net To enable our resolvers to work properly, we've had to tell them to ignore the root nameservers which appear to have bad data. On a Bind 4.X system, one can do this with the 'bogusns' configuration directive: bogusns 128.9.0.107&255.255.255.255 192.33.4.12&255.255.255.255 128.8.10.90&255.255.255.255 192.203.230.10&255.255.255.255 192.5.5.241&255.255.255.255 192.112.36.4&255.255.255.255 198.41.0.10&255.255.255.255 193.0.14.129&255.255.255.255 198.32.64.12&255.255.255.255 198.32.65.12&255.255.255.255 For Bind 8.X servers, something like server 128.9.0.107 { bogus yes; } server 192.33.4.12 { bogus yes; } [etc...] should work, I think. - roy - -----BEGIN PGP SIGNATURE----- Version: 2.6.2 iQCVAwUBM84G8mx7n9NanyP9AQENvwP/dEYxFjxDh83OL9xvVImGrjO2202h4jts kK57u41y+DnnMehZitF9mtAhRPT0z469mmBrmWJC1EhgKlDjrm0YZwv7ZmHTgPQU 0GYcRMUPR8g7zYlnNwZxoEgUwpMzOj/SFbokL38Kojuy58CZDJZ7BrN5WFsV9/a9 Zc0s4eg+z8M= =fzlJ -----END PGP SIGNATURE-----

13 14