October 1998 - Test - lists.nanog.org

Re: Clue's for Clue-less
by Richard Irving 27 Oct '98

27 Oct '98

Sharif Torpis wrote: > > this must be an 11.2 thing? > Unknown... Have you ever met an 11.1.xx technician *not* wearing synthetics to test it ? :) > Richard Irving wrote: > > > > It is built into IOS.... As you approach the ceiling, it sends out an > > SNMP log alert. > > > > Now, how I select the number is confidential.... > > > > Ok. I will give you a hint.... I bring up the original sessions, > > and confirm integrity... > > > > Then I look at the routes, size and count. > > > > Then in the dark of the moon, wearing *no* synthetics, > > facing *magnetic* north... I devine the target number. > > > > In other words... It is a judgement call... > > > > Downstreams are easier. They must file routes with me, before I route > > them. I filter to the prefix. > > > > Consider this merely another fall back mechanism to the general IOPS > > model. > > > > Sharif Torpis wrote: > > > > > > can u elaborate on the detection mechanism for setting off such > > > alarms? do u look at % change in total routes in core of internet? > > > e.g. sunday saw a change of 55k -> 60k which is ~10% change ... or do > > > u look at specific blocks that should not be deaggregated and catch > > > them if they are deaggregated ... thanks > > > > > > Richard Irving wrote: > > > > > > > > The only neighbors I lost on this one, dumped something > > > > they shouldn't..... If someone de-aggregates a /16, > > > > it fires off alarms.... Although these may be valid advertisements, > > > > We have opted for the "safe, rather than sorry" perspective. > > > > (Besides, the alarms *assure* prompt attention) > > > > > > > > PS: 11.2.xx and higher have this command...

1 0

Re: root name servers
by Sean Donelan 27 Oct '98

27 Oct '98

noc(a)nso.ORG (Network Operations Center) writes: >anybody knows the physical location >of all root name servers, or a pointer ? The precise physical location (e.g. room number and rack) of all the root name servers is not generally available public information. It isn't "classified" information in the military sense, but some people get nervous nonetheless. Anyone who has been around the Net for a while will pick up that information eventually. A list of the general physical location (e.g. city and organization) can be found at <http://www.nic.mil/DNS/root-server.html>. It isn't precisely accurate, but I don't know if it is deliberate deception, or just an Oops, close enough for government work. It doesn't include the more recent root-servers servers at LINX and WIDE. However, as the last few attempts by NSI proved, geographic location isn't as important as network topology. Putting a GTLD name server in London doesn't do any good if its network connectivity sucks. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

1 0

Re: Collecting Stories
by snash＠lightning.net 27 Oct '98

27 Oct '98

> Collecting stories and instances where you were too tired, intoxicated, > inebriated, confused, distracted or otherwise temporarily rendered > unqualified to touch that router or switch - but went ahead and did it > anyways. Successes and failures welcome. Date: Tue, 27 Oct 1998 14:32:51 PDT Well there was that one time Reid had me bent over the 7200 and I broke it.

4 3

Re: test
by John Fraizer 27 Oct '98

27 Oct '98

At 02:59 PM 10/26/98 -0800, you wrote: >Other lessons learned, > Linux won't route between multiple NICs, using multiple aliases, > even with ip-forwarding enabled. Someone needs to tell one of our routers this. It has been routing between multiple NICs for well over a year without any problem. ------- John Fraizer | __ _ The System Administrator | / / (_)__ __ ____ __ | The choice mailto:root@EnterZone.Net | / /__/ / _ \/ // /\ \/ / | of a GNU http://www.EnterZone.Net/ | /____/_/_//_/\_,_/ /_/\_\ | Generation A 486 is a terrible thing to waste...

3 2

Re: Is Qwest leaking routes?
by Sean Donelan 27 Oct '98

27 Oct '98

spork(a)inch.COM (Charles Sprickman) writes: >> >Yes, it appears Qwest was leaking routes, but they are fixing it... > >Am I just silly to assume that it shouldn't take so long for such >well-funded companies to communicate with each other? AS286 is EUNet, >right? This all started sometime yesterday... I still see this: > >BGP routing table entry for 206.97.128.0/19, version 10233204 >Paths: (1 available, best #1) > 3847 1239 1800 209 286 3561 > 207.240.48.45 from 207.240.48.45 (207.240.48.1) > Origin IGP, localpref 100, valid, external, best Have you tried asking your upstream, Sprint (AS 1239) why they are still listening to, and re-announcing the route? Was it the old cisco bug, where routes sometimes get stuck in the route table due to dropping the withdrawl? Or some other reason? As part of the post-mortum, was everyone able to easily reach the correct people at the responsible NOC's of other providers to resolve this problem? I've been told in the past there are no communication difficulties between the billion dollar providers, and bi-lateral processes work well. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

3 2

Re: Lucent WaveLAN PC Cards?
by Brent Sweeny 27 Oct '98

27 Oct '98

and on the subject of 802.11 wireless, does anyone know of 802.11 PCCards with drivers for Linux? We haven't been able to find any. thanks/ Brent Sweeny sweeny(a)indiana.edu Indiana University }> Does anyone know it there will be support for the new 802.11 2.4GHz WaveLAN }> cards at the Atlanta NANOG or only the older, non 802.11 compliant ones? }> }> Either way, can someone recommend a clueful wholesale distributor (or worst }> case, "retail" dealer) of the Lucent WaveLAN hardware so that I can buy an }> appropriate PC card before I come? }> }> Regards, }> }> -- }> Wayne D. Correia Critical Path Inc. main: +1.415.808.8800 }> CTO <wayne(a)cp.net> San Francisco, CA 94105 fax: +1.415.826.6100

1 0

Re: Is Qwest leaking routes?
by Sean Donelan 27 Oct '98

27 Oct '98

>Yes, it appears Qwest was leaking routes, but they are fixing it... Its always so much fun to watch billion dollar companies dance around the issues. Qwest's spokesperson could not confirm nor deny any problem, and even if such a problem existed considered it a confidential matter. On the other hand, C&W received top billing on the news story, probally because they were one of the few providers to actually give any information to the public and reporters. Grrr. I'm not a big fan of C&W, what are your peering requirements this week?; but I don't like to see providers getting zinged for being open about network issues. However the root cause is a very old issue. It takes two to create a route leak. One to announce the erroronous route, and one to listen to the erronous route. The proper router filters on either side of the connection would mitigate the problem. Everyone knows the basic issues. 1) cisco routers can't hold the extremely large access-lists required to fully filter peer-to-peer connections between large providers. 2) some providers don't make available the information needed to accurately verify their routing information. 3) review the archives. -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

2 2

Re: test
by Roeland M.J. Meyer 27 Oct '98

27 Oct '98

At 09:39 PM 10/25/98 -0800, Brian Reid wrote: >> Shall I resend them? Best...\Stef > >Yes. They are lost in the aether. After a few daze at reduced service, MHSC is now fully operational once more. In a nut-shell, we lost our main switch (we have a switched 10baseT/100baseTX LAN). Lesson learned, never buy LinkSys ever again. Good product, lousy warranty RMA service (went down last Wed, didn't get replacement until this morning <sigh>) Other lessons learned, building static routes for 16 multi-tiered sub-nets, under pressure, sux big ones! Building and learning gated-3.5.9 can not be done in 24-hours. Linux won't route between multiple NICs, using multiple aliases, even with ip-forwarding enabled. ___________________________________________________ Roeland M.J. Meyer, ISOC (InterNIC RM993) e-mail: <mailto:rmeyer@mhsc.com>rmeyer(a)mhsc.com Internet phone: hawk.mhsc.com Personal web pages: <http://www.mhsc.com/~rmeyer>www.mhsc.com/~rmeyer Company web-site: <http://www.mhsc.com/>www.mhsc.com/ ___________________________________________ I bet the human brain is a kludge. -- Marvin Minsky

2 1

B8ZS DS3??? HUH??
by Martin, Christian 27 Oct '98

27 Oct '98

All, We have a DS1 line down, that is the result of a line cut along a DS3 trunk. Apparenty, the cut is either so bad, or so hard to reach, that we are getting an ETR of 2 days+. This is not my problem, however. When asked about rerouting, one of the Telco techs stated that they couldn't reroute right away, becasue the DS3 is not provisioned for B8ZS. To which I replied, "So. I didn't know any DS3 used B8ZS line coding." Anyway, my question is this: Is there a DS3 signalling standard that would call for B8ZS instead of B3ZS? And if so, what difference would it make if the coding were coming in as B8ZS, AMI, 4B/5B, NRZI, etc - it still needs to be demuxed back into the original signal, to terminate in the switch, right? Besides, I would think channelized DS1s riding a DS3 would already have been ones-density ensured, so any ones-density scheme on a channelized DS3 seems moot. If I am wrong, however, let me know so I can explain this to my techs for future reference. Thanks! Chris

1 0

Collecting Stories
by Jason Weisberger 26 Oct '98

26 Oct '98

Simple request, if you have time. Collecting stories and instances where you were too tired, intoxicated, inebriated, confused, distracted or otherwise temporarily rendered unqualified to touch that router or switch - but went ahead and did it anyways. Successes and failures welcome. Please send them directly to me, -- Jason Weisberger Chief Technology Officer SoftAware, Inc. 310/305-0275

3 4