- NANOG - lists.nanog.org

Re: The MSN Messenger
by Moe Allen 09 Jul '01

09 Jul '01

Well, we are one of those small ISP's. Our company, over the past 6 years, had an association with a larger ISP. This larger ISP had customers in a Metroplex area as well as customers in South America. If you believe in the old adage, all things are equal, when everyone is running the same two programs (MSIE & Netscape) and same type of equipment, then the percentage of trouble tickets would be the same in a small ISP as it would be in a large one. On one occasion the President of the larger ISP call and ask if I could come down for a meeting. So I did. The main topic of this meeting was, why were we having less customer complaints, than the big boys. Both companies were handling everything the same. My answer was, we started a Help Desk. Everytime something went wrong, I sent out an e-mail. The e-mail detailed the problems and how long I thought it was going to take to fix. If during the repair, we found it was going to be longer, I sent another e-mail to our customers. Upon repair, I sent another e-mail and explained the problems and include, if the repair was a band-aid or a solid repair. I also e-mailed our group about planned outages, C&W problems, etc. Well it have been 6 years and I am still writing the e-mails. But now, we call the "Vidnet -- Alerts" In our last poll, I asked if our customers wanted the "Alerts" stopped. To my surprise, a 100% of the users that replied stated, "No, keep up the good work." So, I don't believe you can go wrong, if you keep your customers informed. It doesn't matter if you are an ISP or a Software Company. The more information you can supply to the customer, the happier the customer. Even if they don't understand everything in the e-mail... Respectfully, Morris Allen President VidcomNet, Inc.

2 1

IN-ADDR.ARPA Progress Report
by ARIN IN-ADDR Role 08 Jul '01

08 Jul '01

Goal: Reduce the size of the IN-ADDR.ARPA zone file. This file will be reduced in size by only containing name servers for /8 networks. Completed Steps: Apr 2001 - Established relationships with 2 companies to provide independent server networks for all ARIN IN-ADDR zone files. Jun 2001 - Released a new IN-ADDR.ARPA zone file generator to correct existing erroneous data. Next Step: July 9, 2001 - ARIN will begin revising the IN-ADDR.ARPA zone for those direct /8 delegations to ARIN. These /8 zone files will be serviced by the two ARIN nameserver networks. These changes will not affect the performance of reverse lookup in any of the IPv4 Address space. Please direct any questions to inaddr(a)arin.net. Ginny Listman Director of Engineering American Registry for Internet Numbers

2 1

The MSN Messenger
by Sean Donelan 07 Jul '01

07 Jul '01

Since it is always difficult to predict how long it will take to fix a problem, the question for the PR people always is: Do you say as little as possible, hoping it will be over soon? The problem with this strategy is if your problem continues for a long time, you look incenitive or even incompetant. Do you confirm you have a problem, and provide customer updates? The problem with this strategy is you draw attention to minor issues, which few people would have noticed otherwise. After four days, Microsoft finally added a statement to their MSN Messenger web site. I guess Microsoft's PR people decided they couldn't make things worse at this point. > July 6, 2001/8:00 pm PST > > MSN Messenger is currently experiencing a service outage for many > customers on a worldwide basis. We wholeheartedly apologize for the > inconvenience you may be experiencing. Our operations team is working > diligently to restore full service (including restoration of all > personal contact lists) as soon as possible. We are working around > the clock to have MSN Messenger up ASAP. We expect to have the service > restored by the end of the day today. Please continue to check back > for updated status reports. Thank you for your patience. > > The MSN Messenger

3 2

The Cidr Report
by Tony Bates 07 Jul '01

07 Jul '01

This is an auto-generated mail on Fri Jul 6 23:00:00 PDT 2001 It is not checked before it leaves my workstation. However, hopefully you will find this report interesting and will take the time to look through this to see if you can improve the amount of aggregation you perform. The report is split into sections: 0) General Status List the route table history for the last week, list any possibly bogus routes seen and give some status on ASes. 1) Gains by aggregating at the origin AS level This lists the "Top 30" players who if they decided to aggregate their announced classful prefixes at the origin AS level could make a significant difference in the reduction of the current size of the Internet routing table. This calculation does not take into account the inclusion of holes when forming an aggregate so it is possible even larger reduction should be possible. 2) Weekly Delta A summary of the last weeks changes in terms of withdrawn and added routes. Please note that this is only a snapshot but does give some indication of ASes participating in CIDR. Clearly, it is generally a good thing to see a large amont of withdrawls. 3) Interesting aggregates Interesting here means not an aggregate made as a set of classful routes. Thanks to GX Networks for giving me access to their routing tables once a day. Please send any comments about this report directly to me. Check http://www.employees.org/~tbates/cidr-report.html for a daily update of this report. ------------------------------------------------------------------------------ CIDR REPORT for 06Jul01 0) General Status Table History ------------- Date Prefixes 290601 100654 300601 101008 010701 100929 020701 100843 030701 101022 040701 101049 050701 100922 060701 101036 Check http://www.employees.org/~tbates/cidr.plot.html for a plot of the table history. Possible Bogus Routes --------------------- AS Summary ---------- Number of ASes in routing system: 11154 Number of ASes announcing only one prefix: 6601 (3771 cidr, 2830 classful) Largest number of cidr routes: 1007 announced by AS701 Largest number of classful routes: 1423 announced by AS701 1) Gains by aggregating at the origin AS level --- 06Jul01 --- ASnum NetsNow NetsCIDR NetGain % Gain Description AS1221 1342 1000 342 25.5% UNKNOWN AS701 1423 1209 214 15.0% UNKNOWN AS4293 394 282 112 28.4% UNKNOWN AS6595 158 57 101 63.9% UNKNOWN AS7018 743 651 92 12.4% UNKNOWN AS6429 201 109 92 45.8% UNKNOWN AS705 225 134 91 40.4% UNKNOWN AS4755 207 122 85 41.1% UNKNOWN AS4151 159 77 82 51.6% UNKNOWN AS8013 298 218 80 26.8% UNKNOWN AS13999 94 15 79 84.0% UNKNOWN AS724 246 172 74 30.1% UNKNOWN AS577 247 174 73 29.6% UNKNOWN AS6471 118 55 63 53.4% UNKNOWN AS3464 152 90 62 40.8% UNKNOWN AS7046 306 245 61 19.9% UNKNOWN AS1 584 524 60 10.3% UNKNOWN AS16758 63 6 57 90.5% UNKNOWN AS1239 446 391 55 12.3% UNKNOWN AS226 144 91 53 36.8% UNKNOWN AS6413 65 13 52 80.0% UNKNOWN AS3908 205 154 51 24.9% UNKNOWN AS17561 112 61 51 45.5% UNKNOWN AS852 211 161 50 23.7% UNKNOWN AS2149 323 273 50 15.5% UNKNOWN AS9842 54 5 49 90.7% UNKNOWN AS12302 79 30 49 62.0% MobiFon S.A. AS5106 81 33 48 59.3% UNKNOWN AS15290 237 191 46 19.4% UNKNOWN AS3602 221 176 45 20.4% UNKNOWN For the rest of the previous weeks gain information please see http://www.employees.org:80/~tbates/cidr-report.html 2) Weekly Delta Please see http://www.employees.org:80/~tbates/cidr-report.html for this part of the report 3) Interesting aggregates Please see http://www.employees.org:80/~tbates/cidr-report.html for this part of the report

1 0

Building NOCs
by Timothy Brown 06 Jul '01

06 Jul '01

Hi NANOG, hope everyone had a great July 4th. Our operation is at a stage where it's time to really invest some money in building a classy network operations facility. Our data center functions are handled seperately, so at this time, this is merely a place for housing operations staff. I've searched the cctec archives and found a few really old threads regarding how to tackle the NOC building process. We're a systems house, but I imagine the principles exist along the same lines. Obviously, some of the principles from building data centers apply here - i'm not totally out of my depth. I've thought about things like lighting, ergonomics, information sources, phones, power, environmental, etc. What i'm really looking for from the community are a) links to manufacturers of good NOC gear, b) recommendations on the above, and c) financial worksheets detailing cost proposals, if any. In addition, any contractor suggestions would be greatly appreciated. When I say NOC gear, what I really mean is NOC furniture, large information displays, and guidance/standards on how to build facilities. It's been a long time since i've done this sort of thing on any kind of a serious level. Since I know that some of the information i'm requesting is proprietary, feel free to eliminate the names of the accused and reply privately off-list. I'll summarize if it looks necessary. Thanks, Tim

2 1

Re: Common causes of performance problems and outages
by Avi Freedman 06 Jul '01

06 Jul '01

In article <10162.15117.22199(a)avi.netaxs.com> Sean wrote: : In roughly the order : 1. Network Engineers (What's this command do?) : 2. Power failures (What's this switch do?) : 3. Cable cuts (Backhoes, enough said) : 4. Hardware failures (What's that smell?) : 5. Congestion (More Bandwidth! Captain, I'm giving you all she's got!) : 6. Attacks (malicious, you know who you are) : 7. Software bugs (Your call is very important to us....) Very much disagree. Or at least, separate performance and outages. Outage/Instability: (Tie for 1st): 1) Shitty software from Major Router Vendors (and everyone else) 1) Misconfiguration a) Often caused by shitty/easy to fumble CLIs from Major Router Vendors 2) Attacks 3) Power failures 4) Cable cuts 5) (distant)Hardware failure Performance has been covered; the only real broadband is Universities, and relatively few transit networks can get to network large and small well. Avi

2 1

Re: SWIP update intervals (was: Re: Getting an AS and /18)
by Bohdan Tashchuk 06 Jul '01

06 Jul '01

> ...and while we're on the subject of SWIPs, can I get a pet peeve off my > chest - the fact that RADB entries are updated in minutes, yet it takes > typically 24 hours or more for a SWIP database change to show up in the > servers (and the next day to even determine that the change was accepted > by the servers)? Is there any real reason for ARIN to not be able to > update the database in near-real-time? The problem is that none of these organizations have any real incentive to care. They're all "monopolies". For example, recently NetSol updated whois database at 2AM on 6/28/01 and didn't bother to update again until 6PM on 7/2/01. Approx. 112 hours later. There are a LOT of things that depend on whois. But, despite language at icann.org web site that says NSI Registry Agreement (form approved 4 November 1999) 9. Publication by NSI of Registry Data. (A) NSI shall provide an interactive web page and a port 43 Whois service providing free public query-based access to up-to-date (i.e. updated at least daily) registry database data NetSol obviously doesn't give a damn about whether or not whois is up to date. They don't have to. Their lucrative monopoly was recently extended for many more years. Unfortunately I don't know of a simple way that databases like these can be maintained without relying on a monopoly organization doing the actual work. It's just something you need to learn to live with. Nothing described similar monopolies better than Lily Tomlin's memorable character Ernestine the telephone operator: We're the phone company. We don't care, we don't have to.

7 7

High speed access
by Roeland Meyer 06 Jul '01

06 Jul '01

If you'd get off the broadband schtick and onto the high-speed schtick (which is what you're really asking about) then we might get somewhere. At one time, PacBell almost had 10Mbps to the home (HFC). Yes, it was asymetric. ATT/TCI brings such speed. 802.11b carriers (Sprint, et al) are trying to bring in symetric 11 Mbps. The we have the local telcos where affordability seems to top out at T1 speeds (1.54Mbps). SDSL seems to top out there as well. Yes, I am not interested in asymetric services. Every study I have done indicates that most users would do more symetric usage if they can. Email is only an asymetric service if you allow spam. Otherwise, it is almost 1:1. Mailing lists are different, but an amazing number of people are not on any mailing list whatsoever. There are a finite number of programs that one can download. Most folks don't download and install software everyday. Also, now that the "new" is wearing off the web, you don't see as much random surf traffic there either. Instant messaging is usually person:person and IRC is falling out due to cracker problems. VOIP is also very symetric, even for teleconfrences. Usage is getting more purposful. Those who do not do "computing" for a living are limiting their Internet time in order to get real work done. As that occurs, usage should get more symetric. This becomes moreso if the user is running personal web pages (which most asymetric service providers do not allow) for a small business or cottage industry site. Another factor is that the rise in home LANs should actually reduce the need for high-speed access (the local server then plays "store and forward"). Personally, I'd like to see switched FDX 10baseT to the home. That's why I went to work for PacBell/ACN, in 1995. There was a chance of actually making that happen. Unfortunately, that window was closed by SBC in 1997 (We had 10K users online and 3 million homes passed). -- From: Larry Diffey [mailto:ldiffey@technologyforward.com] Sent: Thursday, July 05, 2001 2:35 PM The question then remains: What (in your opinion) constitutes broadband according to the services that have been promised to consumers but not yet delivered? Yes, I understand that it's not just speed, but take everything else into account when you consider the minimum speed.

5 4

Drum roll, please! (Re: peering requirements (Re: DDOS anecdotes))
by Paul Vixie 06 Jul '01

06 Jul '01

> (randy, i'm betting that the number of respondants will be a single digit.) And the number is... 5. I'll continue this privately for a while.

1 0

RE: Cogent Communications Info
by Marshall Eubanks 06 Jul '01

06 Jul '01

> IMHO multicast is a killer app, maybe THE killer app, for BLECs, as it conserves sparse bandwidth, fits well with the ethernet LANs typically used with BLECs for the "last mile," and as multicast peering is fairly straightfoward in the BLEC scenario. BLEC multicast also offers interesting opportunities for advertisers, as ads can target specific clusters of buidlings. We have convinced at least one BLEC of these advantages; others on NANOG should feel free to contact me off list for assistance in setting up multicast. Regards Marshall Eubanks Multicast Technologies, Inc. 10301 Democracy Lane, Suite 410 Fairfax, Virginia 22030 Phone : 703-293-9624 Fax : 703-293-9609 e-mail : tme(a)on-the-i.com http://www.on-the-i.com Test your network for multicast : http://www.multicasttech.com/mt/ Check the status of multicast in real time : http://www.multicasttech.com/status/index.html >The wierd thing here is that Williams doesn't have an overabundance of >peering either. They had been SBC's primary IP transit provider for their >DSL product, but rumor has it that SBC has selected other providers to >replace them. Perhaps Williams has massive transit contracts of their own, >and they are trying to cut their losses? Reselling transit for $50/meg is >better than not reselling it at all. > >I doubt Cogent's long term plans are dependent on Williams' firesale transit >pricing. > >- Daniel Golding > >> -----Original Message----- >> From: owner-nanog(a)merit.edu [mailto:owner-nanog@merit.edu]On Behalf Of >> Vincent J. Bono >> Sent: Thursday, July 05, 2001 10:03 PM >> To: David U. >> Cc: Nanog List >> Subject: Re: Cogent Communications Info >> >> >> >> Just to test them out we ordered their "private line" service (rather than >> the transit product) where it is $10/meg point to point, i.e. >> 100Mbps cross >> country is $1,000 per end per month. Its a tunneled IP product >> with an ATM >> infrastructure. >> >> They are already 90 days overdue and show no signs of delivery >> anytime soon. >> >> They are heavily involved with Williams for the ATM backbone and >> I know that >> Williams is selling IP transit to big telecom accounts for as little as >> $50/meg. Since they don't seem to have a lot of peering possibly they are >> filling in the gaps with Williams transit and coming up with a skewed cost >> model? >> >> -Vincent >> >> >> >> >> >> ----- Original Message ----- >> From: "David U." <davidu(a)everydns.net> >> To: "Nanog List" <nanog(a)merit.edu> >> Sent: Thursday, July 05, 2001 9:40 PM >> Subject: Cogent Communications Info >> >> >> > >> > I have been reading about Cogent Communications recently and >> was wondering >> how >> > they can possibly offer 100mbps for $3000 ($1000 if you are an end-user, >> not a >> > service provider). It just seems too good to be true and we >> know how that >> > goes... >> > >> > Does anyone on NANOG have experience with them? >> > >> > Thoughts? >> > >> > For those who don't know: Cogent offers 100 mbps at $3000 to service >> provider >> > or 1000gbps for $20,000. http://www.cogentco.com has some info but not >> much. >> > >> > thanks, >> > -davidu >> > >> > Marshall Eubanks tme(a)21rst-century.com

1 0