Network discovery and mapping
Its been a few years since I looked at network discovery and mapping tools. Openview/et al did the job, but was always a pain to move all the boxes to the right spots on the resulting maps. Has network discovery and mapping improved for medium-scale wide area networks for ISPs (e.g. 1,000 networks, 100,000 network devices)? I've found lots of discovery tools, but intelligent mapping/layout still seems to be a problem. The usual requirements for SNMP smart discovery, interface/subnet mapping, device identification and connecting the right symbols with the right lines to all the other symbols.
Re: Network discovery and mapping
Sunday, June 22, 2003, 7:58:39 AM, Sean wrote: Its been a few years since I looked at network discovery and mapping tools. Openview/et al did the job, but was always a pain to move all the boxes to the right spots on the resulting maps. Has network discovery and mapping improved for medium-scale wide area networks for ISPs (e.g. 1,000 networks, 100,000 network devices)? I've found lots of discovery tools, but intelligent mapping/layout still seems to be a problem. The usual requirements for SNMP smart discovery, interface/subnet mapping, device identification and connecting the right symbols with the right lines to all the other symbols. www.solarwinds.net They have excellent collection of tools which is probably what you;re looking for. Windows only afaik -- Best regards, Subhi S Hashwa mailto:[EMAIL PROTECTED] Operations Manager Electronic Corner Limited
Re: Network discovery and mapping
On Sun, 22 Jun 2003, Sean Donelan wrote: Has network discovery and mapping improved for medium-scale wide area networks for ISPs (e.g. 1,000 networks, 100,000 network devices)? I've found lots of discovery tools, but intelligent mapping/layout still seems to be a problem. The usual requirements for SNMP smart discovery, interface/subnet mapping, device identification and connecting the right symbols with the right lines to all the other symbols. That's quite a medium-scale. Is there a single entity in the world that controls 1,000 networks and 100,000 network devices? Andy --- Andy Dills Xecunet, Inc. www.xecu.net 301-682-9972 ---
Re: Network discovery and mapping
On Sun, 22 Jun 2003, Andy Dills wrote: That's quite a medium-scale. Is there a single entity in the world that controls 1,000 networks and 100,000 network devices? WorldCom^Hn -- Jon Lewis [EMAIL PROTECTED]| I route System Administrator| therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_
Re: Network discovery and mapping
On Sun, Jun 22, 2003 at 01:16:38PM -0400, Andy Dills wrote: On Sun, 22 Jun 2003, Sean Donelan wrote: Has network discovery and mapping improved for medium-scale wide area networks for ISPs (e.g. 1,000 networks, 100,000 network devices)? I've found lots of discovery tools, but intelligent mapping/layout still seems to be a problem. The usual requirements for SNMP smart discovery, interface/subnet mapping, device identification and connecting the right symbols with the right lines to all the other symbols. That's quite a medium-scale. Is there a single entity in the world that controls 1,000 networks and 100,000 network devices? I am a network engineer for a cable ISP. We have over 50,000 cable modems and around 65,000 customer devices. We only have 200-250 networks, but well over 100,000 ip devices. -- Jonathan Crockett Network Engineer Midcontinent Communications
Re: Network discovery and mapping
On Sun, 22 Jun 2003, Sean Donelan wrote: Its been a few years since I looked at network discovery and mapping tools. Openview/et al did the job, but was always a pain to move all the boxes to the right spots on the resulting maps. Has network discovery and mapping improved for medium-scale wide area networks for ISPs (e.g. 1,000 networks, 100,000 network devices)? I've found lots of discovery tools, but intelligent mapping/layout still seems to be a problem. The usual requirements for SNMP smart discovery, interface/subnet mapping, device identification and connecting the right symbols with the right lines to all the other symbols. Cheops and Cheops-ng might be useful to you. http://www.marko.net/cheops/ http://cheops-ng.sourceforge.net/ Justin
RE: The Cidr Report
Not sure how relevent this may be but:
Interland has recently been in a major network
move
They boight out Communitech and are in the
process of moving datacenters to the Interland
centers..
This could explain it
But they should be doing a better job of it though...
Jim
-Original Message-
From: Hank Nussbacher [mailto:[EMAIL PROTECTED]
Sent: Saturday, June 21, 2003 3:41 PM
To: Haesu; [EMAIL PROTECTED]
Subject: Re: The Cidr Report
At 01:00 PM 21-06-03 -0400, Haesu wrote:
What is up with ASN11305 generating humongous loads of unaggregated /24's?
Sent them an email 11 days ago, no reply yet:
Date: Tue, 10 Jun 2003 10:56:46 +0200
To: [EMAIL PROTECTED], [EMAIL PROTECTED]
From: Hank Nussbacher [EMAIL PROTECTED]
Subject: AS11305 - routing table bloat
Cc: Terry Baranski [EMAIL PROTECTED], [EMAIL PROTECTED]
AS11305 has been lately seen to be sending out too many prefixes not based
on CIDR boundries, thereby increasing the global router table size:
ASnumNetsNow NetsAggr NetGain % Gain Description
AS11305 646 136 51078.9% INTERLAND-NET1 Interland
Incorporated
See http://www.mcvax.org/~jhma/routing/ and http://bgp.potaroo.net/cidr/
and http://bgp.potaroo.net/cgi-bin/as-report?as=as11305view=4637
for further details.
Regards,
Hank
-Hank
-hc
Aggregation Summary
The algorithm used in this report proposes aggregation only
when there is a precise match using the AS path, so as
to preserve traffic transit policies. Aggregation is also
proposed across non-advertised address space ('holes').
--- 20Jun03 ---
ASnumNetsNow NetsAggr NetGain % Gain Description
Table 122681877223495928.5% All ASes
AS7132 923 229 69475.2% SBIS-AS SBC Internet Services
- Southwest
AS11305 647 137 51078.8% INTERLAND-NET1 Interland
Incorporated
AS701 1514 1070 44429.3% ALTERNET-AS UUNET
Technologies, Inc.
AS7843 614 175 43971.5% ADELPHIA-AS Adelphia Corp.
AS4323 600 177 42370.5% TW-COMM Time Warner
Communications, Inc.
AS7018 1337 927 41030.7% ATT-INTERNET4 ATT WorldNet
Services
AS3908 889 521 36841.4% SUPERNETASBLK SuperNet, Inc.
AS1221 1062 756 30628.8% ASN-TELSTRA Telstra Pty Ltd
AS6197 518 225 29356.6% BATI-ATL BellSouth Network
Solutions, Inc
AS4355 397 111 28672.0% ERMS-EARTHLNK EARTHLINK, INC
AS6198 475 189 28660.2% BATI-MIA BellSouth Network
Solutions, Inc
AS1239 959 677 28229.4% SPRINTLINK Sprint
AS6347 367 92 27574.9% DIAMOND SAVVIS Communications
Corporation
AS27364 319 87 23272.7% ACS-INTERNET Armstrong Cable
Services
AS17676 250 24 22690.4% GIGAINFRA XTAGE CORPORATION
AS22773 2208 21296.4% CCINET-2 Cox Communications
Inc. Atlanta
AS209498 305 19338.8% ASN-QWEST Qwest
AS705508 331 17734.8% ALTERNET-AS UUNET
Technologies, Inc.
AS2386 406 235 17142.1% INS-AS ATT Data
Communications Services
AS2048 258 87 17166.3% LANET-1 State of Louisiana
AS17557 341 173 16849.3% PKTELECOM-AS-AP Pakistan
Telecom
AS6327 190 24 16687.4% SHAWFIBER Shaw Fiberlink
Limited
AS13601 205 46 15977.6% ASN-INNERHOST Innerhost, Inc.
AS690450 293 15734.9% MERIT-AS-27 Merit Network
Inc.
AS20115 463 311 15232.8% CHARTER-NET-HKY-NC Charter
Communications
AS3602 226 79 14765.0% SPRINT-CA-AS Sprint Canada
Inc.
AS2686 258 112 14656.6% AS2686 ATT Global Network
Services - EMEA
AS6140 297 155 14247.8% IMPSAT-USA ImpSat
AS7303 238 98 14058.8% AR-TAST-LACNIC Telecom
Re: Network discovery and mapping
On Sun, 22 Jun 2003 [EMAIL PROTECTED] wrote: On Sun, 22 Jun 2003, Andy Dills wrote: That's quite a medium-scale. Is there a single entity in the world that controls 1,000 networks and 100,000 network devices? WorldCom^Hn Well, sure, MCI is a single company that owns that many networks and possibly network devices, but are you saying there's a single group of people within MCI who are tasked with mapping, down to the host level, across their entire network? It would seem reasonable to me that there is some hierarchy involved...European division deals with its network, etc. On Sun, 22 Jun 2003, Jonathan Crockett wrote: I am a network engineer for a cable ISP. We have over 50,000 cable modems and around 65,000 customer devices. We only have 200-250 networks, but well over 100,000 ip devices. What would be the value of mapping all those CPEs? Not that there isn't one, I just don't know what it is. Andy --- Andy Dills Xecunet, Inc. www.xecu.net 301-682-9972 ---
Re: BTinternet problems?
Mike wrote: We're receiving multiple complaints about problems reaching anything @bt. Is anyone else experiencing this? GrrrThree days later, BT is now telling their customers that somehow, this is our fault. I find it rather odd that everyone in the world can reach us, *except* BT customers, yet it is our fault. You appear to be excessively deaggregating your space. Perhaps they are doing the responsible thing by filtering it? As announced, 13345 has 92 prefixes originated. After aggregation, there are 44 prefixes. There appear to be a couple of holes which you aren't announcing, which would further reduce this to 38 (after aggregation.) The space from the Colorado Internet Cooperative Association seems to be somewhat haphazard leading to the bulk of the remainder, and they aren't originating 207.174.0.0/16 at all, which may lead to issues. In this day and age of many people not doing the right thing, it is likely that one only see's the effects when looking at someone actually doing the right thing. I'ld suggest cleaning up your announcements and seeing if the problem persists. If nothing else, you become a part of the solution.
Re: BTinternet problems?
[EMAIL PROTECTED] wrote: You appear to be excessively deaggregating your space. Perhaps they are doing the responsible thing by filtering it? I had a /20 from which BT was unreachable, and a /24 working just fine, so this seems doubtful, unless they are doing it to be spiteful and punitive alone. I say had because we have since withdrawn almost all announcements that AS7018 was learning from AS19694 and temporarily fixed the problem. It still exists, and it is clear now that the problem lies somewhere between AS7018 and AS2856. AS7018 says that my traceroutes are confusing to them, so I have pretty much given up hope of ever seeing a resolution to this. As announced, 13345 has 92 prefixes originated. After aggregation, there are 44 prefixes. Policy for most of those were set prior to my involvement with BGP here. I have managed to prevent the (otherwise forgone) deaggregation of 204.188.96.0/20 and 199.45.236.0/22 since my involvement at this level. I am also working to prepare an application for PI space from ARIN and the end result will be to reduce our total # of prefixes. Due to topology, aggregation of the remainder at this time would require several hundred clients to renumber immediately. Yes, I know this sucks. Yes, I am working towards a solution. No, it isn't going to happen tomorrow. Yes, I feel the pain every time I look at what we advertise. Mike
Re: Network discovery and mapping
On Sun, 22 Jun 2003, Andy Dills wrote: That's quite a medium-scale. Is there a single entity in the world that controls 1,000 networks and 100,000 network devices? Its a bit like the fish that got away. People have varying ideas about how big is big. Its smaller than the Internet, but larger than a mompop network. Most americans consider themselves middle-class, no matter what their net worth is. As far as a single entity, obviously all large organizations have learned how to delegate responsibility. The US Military has about 3 million network devices connected to 3,000 networks. But no single person really controls all 3 million network devices. Its the organizational gaps between entities I'm interested in mapping. I want to discover and map the connections indviduals may know about, but no one realized how all the pieces were connected. So far the recommendations have included Cheops NetViz OpenView Intermapper
Re: Network discovery and mapping
On Sun, Jun 22, 2003 at 09:24:58PM -0400, Sean Donelan wrote: gaps between entities I'm interested in mapping. I want to discover and map the connections indviduals may know about, but no one realized how all the pieces were connected. So far the recommendations have included [...] I'm not one to push commercial products, but I don't know of a freely available tool that does the equivalent of what Lumeta http://www.lumeta.com does. This being the solution based on the original work of Cheswick and Burch. This may be just the kind of thing if you need to discover unexpected or even unknown paths. John
RE: IRR/RADB and BGP
On Fri, 20 Jun 2003, Andy Dills wrote: I dunno, there are plenty of smaller ASes who have yet to be forced to register their routes. We haven't yet been forced, but I finally got motivated to submit them to altdb last night. Altdb definitely rocks. Back when I got PI space in 1998, there were definitely some backbones ignoring routes not found in the IRR. I wonder if they gave up, or people just don't notice them anymore. -- Jon Lewis [EMAIL PROTECTED]| I route System Administrator| therefore you are Atlantic Net| _ http://www.lewis.org/~jlewis/pgp for PGP public key_
