October 1998 - Test - lists.nanog.org

Re: *sigh* NT/netbios over IP over frame
by Roeland M.J. Meyer 13 Oct '98

13 Oct '98

At 12:23 AM 10/13/98 -0700, Raymond Wong wrote: > > >I swore I'd never do anything like this, but I've been tasked with >establishing NT network connectivity between 2 private networks >over frame relay. I got a pair of 2501s running 11.3, and I've got IP >routing right (all the sparcs are happy). I've got NT/etc machines, running >with NetBIOS bound over the TCP stack. None of the PC people knows >IP well enough to say what they need. Does anyone who has done this >care to provide any hints? a URL to a faq would be fine, though a search >seemed to turn up lots of assumptions that wins is required, which >sounds odd to me. > >I've enabled IP forward udp, and tried setting ip helper to the broadcast >address on the other side... should this work? am I missing an obvious >reason it won't, or better, an obvious way to make it work? I'm not about >to go around and change 400 PCs for this, so a router config would really >be preferable. First off, the guys that *really* know this stuff are on the Samba list. Mickeysoft is clueless when it comes to this stuff. The first question is, if there is a Primary Domain Controller involved anywhere? The next question is, what are you plannning for your gateway machines? WINS *may* be required, but a Samba Server can fulfill that function quite well. You'll also need to set up LMHOSTS files. Your main problem is that SMB protocols do not route easily. In fact, any version less than WinNT will not route out of its own sub-net. If any PCs, are running Win95 then you'll need a pair of hosts to route the SMB protocols for them. This could be a WinNTserver, but I prefer a Linux host running Samba suite (you won't have to re-boot on each config change and you can run it headless-remote). Datacom Warehouse may have a box thatll help you, but I'd look into a "blue cube". ___________________________________________________ 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

IGPs in use
by Andre' Zehl 13 Oct '98

13 Oct '98

Is there an overview on what the dominantly used interior routing protocols are percentagewise in large AS backbones (IBGP, OSPF, IGRP, EIGRP, IS-IS, any?). I don't want to start religous wars on what's the best protocol, I'm rather interested if there is an overview available on the facts of protocols used. Is there an "objective" (vendor-independent) feature-based overview on the compared advantages/disadvantages (especially regarding redistribution) of the various protocols that goes with those numbers? Andre' zehl(a)berkom.de

2 1

International Murphy's Day [was: Discussion point: Telco quality?]
by cache22 12 Oct '98

12 Oct '98

> Subject: Re: Discussion point: Telco quality? > Date: Sat, 10 Oct 1998 23:02:02 -0700 > From: "Roeland M.J. Meyer" <rmeyer(a)mhsc.com> > To: David Lesher <wb8foz(a)nrk.com> > CC: nanog(a)merit.edu (nanog list) > References: <199810110142.SAA17274(a)condor.lvrmr.mhsc.com> > > At 10:24 PM 10/10/98 -0400, David Lesher wrote: > >Unnamed Administration sources reported that Roeland M.J. Meyer said: > >> >Are there things that can happen to both which will really ruin your day? > >> Yes. > >> <snip> > >But such will never happen, so we can all rest easy....right? > > Cascaded failures will do it. aka Murphy having a party. > ----------------------------------------------- Big Party! International Murphy's Observance Day! January 1, 2000 Everyone's Invited -- Location to be Announced! ----------------------------------------------- or will it be International Murphy's Observance Year? recent y2k news: worldcom is far behind schedule: http://june.wcom.net/about_worldcom/year_2000_compliance/timeline.shtml congressional testimony from bellatlantic on what should (and isn't) being done: http://home.swbell.net/adheath/castellano.htm little progress from bellsouth: http://www.garynorth.com/y2k/detail_.cfm/2560 amazing performance from sbc! (sw bell & pacbell) http://www.garynorth.com/y2k/detail_.cfm/2506 the itu is sleepwalking: http://www.garynorth.com/y2k/detail_.cfm/2832 international communications (and banking) status: (hopeless) http://y2ktimebomb.com/Industry/Telecom/ihugo9840.htm

2 1

Re: IGPs in use
by Danny McPherson 12 Oct '98

12 Oct '98

Fortunately, they scale well enough to pay most of our salaries, in one way or another :-) I suppose though that network design, then vendor implementation issues introduce the bulk of the "preceived" scalability issues, not that they don't exist. -danny Randy Bush wrote: >> The use of ISIS shouldn't lead folks to believe that OSPF doesn't scale well >> though. > >neither scale well. > >randy

1 0

Re: IGPs in use
by Danny McPherson 12 Oct '98

12 Oct '98

Just as with ISIS .. and most anything else. In order to continue operating when it's "ceiling" is reached, network design and vendor implementation must be tailored (optimized) to accommodate shortcomings in the specific application. I've seen one of these "theoretical" ceilings w/ISIS at a previous employer (one of the "big 5"). I guess you could say "allowing the glass to break wasn't an option". -danny > Perhaps I'm forgetting something... Doesn't OSPF have a "glass > ceiling".. > > Kind of like the current mboned issue's ? > > Routes, * states, * interfaces | macs, * changes...... != scalable... > > Perhaps I am wrong..... Anyone care to clarify ? > > Or, should we limit our scope of the term "scalable"... > In which case, OSPF is dynamite.

1 0

Re: NANOG Home Page Broken?
by Henry Linneweh 12 Oct '98

12 Oct '98

Yeah it seems to be giving the forbidden to enter message, maybe they are upgrading the site or doing maintenance. Henry R. Linneweh Mark Tripod wrote: > Anyone tried to get to www.nanog.org > > or > > www.merit.edu > > or > > www.rsng.net > > -- > > *************************************************************************** > Mark Tripod - Senior Network Architect - Exodus Communications > http://www.exodus.net - (888) 2-EXODUS - support(a)exodus.net > ASN 3967 - NASDAQ (EXDS) - Direct: (408) 346-2389 -- ¢4i1å

2 1

Re: NANOG Home Page Broken?
by Susan R. Harris 12 Oct '98

12 Oct '98

We're having server problems - please stay tuned. On Mon, 12 Oct 1998, Mark Tripod wrote: > Anyone tried to get to www.nanog.org > > or > > www.merit.edu > > or > > www.rsng.net > > -- > > *************************************************************************** > Mark Tripod - Senior Network Architect - Exodus Communications > http://www.exodus.net - (888) 2-EXODUS - support(a)exodus.net > ASN 3967 - NASDAQ (EXDS) - Direct: (408) 346-2389 >

1 0

Discussion point: Telco quality?
by Paul Ferguson 12 Oct '98

12 Oct '98

This latest fiber cut just kind of jarred my brain into the realization of what "telco quality" services really boils down to. In the end, all services, both traditional telephony services and data services, are more alike than people give them credit for. :-) - paul

8 14

Co-locate weather conditions (Re: MAE NY?)
by Sean Donelan 12 Oct '98

12 Oct '98

davediaz(a)netrail.NET (David Diaz) writes: >This was all good with the exception of temp for condensation. >Condensation can occur at just about any temp (within reason) that is what >Dew Point is. Humidity is Dew point/(room) Temperature. When they match >you have humidity of 100% and start to see condensation/fog. Anyway this >is from flight school. I have heard different engineers recommend 40% >humidity and maybe 67.5 degrees temp. Something about enough humidity to >avoid static dischargies etc...?? Normally people pay consultants a lot of money to go look this stuff up in libraries :-) You are correct, it is actually a XY curve plotted with both relative humidity and temperature. Condensation/fog is not good for a co-locate space. In addition to electrostatic discharge other issues include corrosion, fungial growth, affect on the other material stored in the area, and of course comfort for the human occupants. Go to far in any direction, and you'll have problems. You are looking for the sweet spot. The problem for a co-locate operator is there isn't one, but several depending how the occupants decide to use the space. Further, what is 'acceptable' may vary depending whether you are the one paying the utility bill, or if someone else is paying it. cigars 70f/70% R.H. wine 56f/>20% R.H. mag tapes 63f/25% R.H. (> 10yr storage) mag tapes 70f/40% R.H. (< 5yr operating) museum objects 70f/50% R.H. photographs 68f/35% R.H. color photos 40f/35% R.H. books/paper 70f/35% R.H. "clean rooms" 72f/50% R.H. You'll notice I didn't list 'co-locate space'. If you do some research you will find the 'optimum' settings are always one study away from changing. As researchers discover new and inventive ways for Murphy to attack, they also discover the previous settings sometimes created the perfect petri dish for problems to grow. People get very religious about the 'correct' environment. And we haven't gotten into things such as filtration for dust, smoke, gases, pollutants. Or other issues like electromagnetic interference, vibration, lighting, acoustic noise, sesimic stability, floor loading, fire prevention and resistance, etc. Government and technical sites of interest Magnetic media storage http://www.nml.org/ Books, paper, photographs http://www.loc.gov/ General storage http://www.nara.gov/ Electrostatic discharge http://www.eosesd.org/ -- Sean Donelan, Data Research Associates, Inc, St. Louis, MO Affiliation given for identification not representation

4 3

WCG
by jamie＠dilbert.ais.net 12 Oct '98

12 Oct '98

Anyone know any contacts with a pulse at WCG? E-mailed noc and contacts for aads ; our peering w/them has been down for weeks.. nobody seems to be alive over there. TIA. -jamie -- jamie rishaw (efnet:gavroche) American Information Systems, Inc. Tel:312.425.7140, FAX:312.425.7240 Help stop spam! router(config)#no ip routing thirty thousand feet above the earth..youre a beautiful thing..

1 0