- NANOG - lists.nanog.org

Re: Traffic engineering tools
by Sean Donelan 22 Oct '99

22 Oct '99

The votes are in, and the commercial winner for IP traffic engineering, planning, modeling, etc package is WAN DL <http://www.wandl.com/> The most frequent comment was "its expensive." Although WAN DL was the (only) commercial package mentioned, several people reported they are working with some homegrown tools, or SWAG. However no one was real happy with either the available commercial or homegrown tools they were using. Several vendor-types sent messages they are working on new tools for release "any day now." I was surprised I didn't hear about any projects from the academic world. The number of CS graduates continues to decline, have we dropped below critical mass?

1 0

DNS for 16.172.in-addr.arpa
by grisha＠verio.net 21 Oct '99

21 Oct '99

We were having a small problem here and I traced it down to a reverse DNS lookup on a 172.16 address hanging. I noticed that a.root-servers.net reports that BLACKHOLE.ISI.EDU is authoritative for 16.172.in-addr.arpa. Is this the way it should be, or is the yet another network solutions mistake? Who should really handle DNS for rfc1918 addresses? Grisha

3 3

RE: Traffic engineering tools
by Bora Akyol 21 Oct '99

21 Oct '99

I believe that WANDL has a tool that can be used for static TE purposes, but you are on your own for traffic models. Bora Akyol Pluris, http://www.pluris.com -----Original Message----- From: Sean Donelan [mailto:sean@donelan.com] Sent: Wednesday, October 20, 1999 11:21 PM To: nanog(a)merit.edu Subject: Traffic engineering tools At NANOG the lack of traffic engineering tools came up. Has anyone heard of any packages coming to market. Or looked if tools from other industries could be used as a starting point. Anything from the electric power transmission or road traffic world we could use? Or do IP packets have such different properties (e.g. re-transmission, independent next-hop behavior, etc), its not a good idea to even try.

1 0

Targeted LDP and Hierarchical LSPs in MPLS networks
by Bora Akyol 21 Oct '99

21 Oct '99

I apologize if this is off-topic. But I thought some of you may be interested in this question that I posed to the MPLS WG in IETF. Reply privately, I will post a summary to the list if I get any interesting replies: ---INCLUDED MESSAGE------- Hi all, I would like to get a poll on how targeted LDP sessions (and hierarchical tunnels) and MPLS data path tie together. Consider the following scenario: * denotes an MPLS tunnel between R_B and R_E R_A --> R_B *************>R_E-->R_F \ / \ / R_C --> R_D Here R_B and R_E have a targeted LDP session and an MPLS LSP between them. There is a label request traveling from R_A for an FEC_1. At R_A, MPLS does a routing table lookup and finds that next hop for this route is R_B, message is then passed on to R_B. Now at R_B, how does R_B know to associate the FEC_1 with R_E through the directed LDP session vs associating with R_C?? One way of doing this is to introduce a virtual interface between R_B and R_E with the lowest metric possible, so the routing lookup result is the virtual interface. The other way of doing this is to exchange all prefixes that the tunnel between is authorized to carry so that the LDP instance at R_B knows that FEC_1 belongs to this set. **What do you think is the best solution and are there any other solutions out there? **Why doesn't the LDP spec address this issue even though it is part of the framework document. ** Is targeted LDP important to ISPs, when you can do this much more cleanly with ER-LSPs in both CR-LDP and RSVP? Bora Akyol, Pluris http://www.pluris.com

1 0

Rochelle Park flood report filed with FCC
by Sean Donelan 21 Oct '99

21 Oct '99

I had hoped to put off doing this for a while, but since the Associated Press has already picked it up, here goes nothing. Bell Atlantic filed their FCC outage report last week. The Bergen County Record newspaper obtained a copy and printed this story. http://www.bergen.com/biz/bell20199910202.htm I speak only for myself.

1 0

NSI again removes services
by Ehud Gavron 21 Oct '99

21 Oct '99

The ever-changing NICNAMES/whois service once again does not show _ANY_ dates -- creation, update, or database update. Of course the obnoxious and non-legaly-binding statement has moved, now to the top of the response. I sure wish NSI got rid of their lawyers and hired some techies. I sure wish the alternate registries could display this kind of date data. Ehud

12 15

Interesting - No reaction to a smurf
by Richard Irving 21 Oct '99

21 Oct '99

We currently have an ongoing attack, and cannot get a directly adjacent network to respond. (Participate in a traceback) The list at puck.nether.net has a few errors.... Jared: The following phone number is disconnected: Concentric Network Corp. internex.net, concentric.net, concentric.com 2828 +1 408 327 2470 noc(a)internex.net The phone number appears to be no longer operational. The email gets an auto-responder. We have not received a human response to an "urgent" in over 24 hours. :( We have an ongoing attack that has been running for over 48 hours total so far....... Would someone with Concentrics (IP Clueful) Security please contact me! (And while you are at it, please update puck's NOC contact list) You appear to have a script kiddie pointing smurf at a downstream. Thanks! Richard Irving One Call Communications, Inc.

3 3

Your morning FUD
by Austin Schutz 21 Oct '99

21 Oct '99

http://news.excite.com/news/zd/991020/12/report-warns-of SAN FRANCISCO -- Internet performance is too low to support serious transactional business by ASPs, and trends indicate that improvements may be inadequate over the next several years, a recent study has found. Northeast Consulting Resources Inc., of Boston, reports that the growing complexity of Web pages and increased network delays will reverse improvements in Internet performance built over the last four years. The findings are bad news for application service providers, which promise to deliver enterprise resource planning and customer relationship management applications. "When the ASP applications come, the time of reckoning will appear," Peter Sevcik, an analyst for NCRI, said earlier this month at the 1999 Global Internet Performance Conference here. Router runaround NCRI says technology improvements have cut the average download time of basic business Web pages from 12 seconds to 6 seconds since 1995, a significant drop despite a 120 percent increase in page size. However, those improvements may not be sustainable over the next three years because overall delays within the Internet have become significantly higher, due primarily to the increasing number of routers. More routers are added as sites deploy additional hardware to provide more scalability. The increasing number of router hops imposed upon data could deteriorate availability to about 9 seconds, the study found. NCRI based its study on data provided by Keynote Systems Inc., a San Mateo, Calif., Web performance research company that publishes the Keynote Business 40 Internet Performance Index, an industry benchmark based on the performance and reliability of 40 leading business Web sites. Keynote announced here at the conference the Keynote Consumer 40 Internet Performance Index, which will be based on the performance of 40 consumer Web sites accessed through a 56K-bps modem. Indexes also will be available for digital subscriber line and cable modems, and the whole package will be sold as the Keynote Consumer Perspective service, which will be available later this quarter, company officials said. Pricing was not yet available.

1 0

Traffic engineering tools
by Sean Donelan 21 Oct '99

21 Oct '99

At NANOG the lack of traffic engineering tools came up. Has anyone heard of any packages coming to market. Or looked if tools from other industries could be used as a starting point. Anything from the electric power transmission or road traffic world we could use? Or do IP packets have such different properties (e.g. re-transmission, independent next-hop behavior, etc), its not a good idea to even try.

1 0

TACACS or Radius daemon on Linux
by mnolan＠southshore.com 20 Oct '99

20 Oct '99

Hi, We are an ISP running several Cisco 2500s, 5200s, and 5300s as access servers. Currently we are entering each user into each box. We are looking to set up a Red Hat Linux machine as either a TACACS or Radius server to centrally validate all our users. Does anyone have any experience running a TACACS or Radius daemon on Linux? Where is a good place to find a Linux TACACS or Radius daemon? I heard that although TACACS is a Cisco product, Radius has more accounting and statistics capability, runs well on Linux, and will validate for Ciscos. Any help would appreciated. Thanks, Mike

7 6